liz/rawTherapee
liz/rawTherapee
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ImProcFunctions cleanup and transition to OpenMP -- phase 1

Browse Source
This commit is contained in:
gabor 2010-05-21 16:14:18 +02:00
parent c427279ce3
commit cd91e7890e
19 changed files with 1944 additions and 2856 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
CMakeLists.txt
View File

@ -116,6 +116,11 @@ if (WITH_RAWZOR)
endif (WIN32) endif (WIN32)
endif (WITH_RAWZOR) endif (WITH_RAWZOR)
find_package(OpenMP)
if (OPENMP_FOUND)
set(CMAKE_CXX_FLAGS ${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS})
endif (OPENMP_FOUND)
add_subdirectory (rtexif) add_subdirectory (rtexif)
add_subdirectory (rtengine) add_subdirectory (rtengine)
add_subdirectory (rtgui) add_subdirectory (rtgui)

5
rtengine/CMakeLists.txt
View File

@ -6,11 +6,12 @@ link_directories (${CMAKE_CURRENT_SOURCE_DIR}/../rtexif ${EXTRA_LIBDIR} ${GTHREA
${GOBJECT_LIBRARY_DIRS} ${GLIB2_LIBRARY_DIRS} ${GLIBMM_LIBRARY_DIRS} ${GOBJECT_LIBRARY_DIRS} ${GLIB2_LIBRARY_DIRS} ${GLIBMM_LIBRARY_DIRS}
${IPTCDATA_LIBRARY_DIRS} ${LCMS_LIBRARY_DIRS}) ${IPTCDATA_LIBRARY_DIRS} ${LCMS_LIBRARY_DIRS})
set (RTENGINESOURCEFILES colortemp.cc curves.cc dcraw.cc gauss.cc iccstore.cc set (RTENGINESOURCEFILES colortemp.cc curves.cc dcraw.cc iccstore.cc
image8.cc image16.cc imagedata.cc imageio.cc improcfun.cc init.cc dcrop.cc image8.cc image16.cc imagedata.cc imageio.cc improcfun.cc init.cc dcrop.cc
loadinitial.cc procparams.cc rawimagesource.cc shmap.cc simpleprocess.cc refreshmap.cc loadinitial.cc procparams.cc rawimagesource.cc shmap.cc simpleprocess.cc refreshmap.cc
stdimagesource.cc myfile.cc iccjpeg.c hlmultipliers.cc improccoordinator.cc stdimagesource.cc myfile.cc iccjpeg.c hlmultipliers.cc improccoordinator.cc
processingjob.cc rtthumbnail.cc utils.cc bilateral2.cc) processingjob.cc rtthumbnail.cc utils.cc labimage.cc
iplab2rgb.cc ipsharpen.cc iptransform.cc ipresize.cc)
add_library (rtengine ${RTENGINESOURCEFILES}) add_library (rtengine ${RTENGINESOURCEFILES})
#It may be nice to store library version too #It may be nice to store library version too

458
rtengine/bilateral2.h
View File

@ -26,6 +26,9 @@
#include <mytime.h> #include <mytime.h>
#include <gauss.h> #include <gauss.h>
#include <glibmm.h> #include <glibmm.h>
#include <omp.h>
// This seems ugly, but way faster than any other solutions I tried
#define ELEM(a,b) (src[i - a][j - b] * ec[src[i - a][j - b]-src[i][j]+0x10000]) #define ELEM(a,b) (src[i - a][j - b] * ec[src[i - a][j - b]-src[i][j]+0x10000])
#define SULY(a,b) (ec[src[i - a][j - b]-src[i][j]+0x10000]) #define SULY(a,b) (ec[src[i - a][j - b]-src[i][j]+0x10000])
@ -34,22 +37,22 @@
int* ec = new int [0x20000]; \ int* ec = new int [0x20000]; \
for (int i=0; i<0x20000; i++) \ for (int i=0; i<0x20000; i++) \
ec[i] = (int)(exp(-(double)(i-0x10000)*(double)(i-0x10000) / (2.0*sens*sens))*scale); \ ec[i] = (int)(exp(-(double)(i-0x10000)*(double)(i-0x10000) / (2.0*sens*sens))*scale); \
int start = row_from; \ int rstart = b; \
if (start<(b)) start = (b); \ int rend = H-b; \
int end = row_to; \ int cstart = b; \
if (end>H-(b)) end = H-(b); \ int cend = W-b;
for (int i=start; i<end; i++) { \
for (int j=(b); j<W-(b); j++) {
#define BL_END(b) buffer[i][j] = v; }} delete [] ec; \ #define BL_END(b) buffer[i][j] = v; }} delete [] ec; \
for (int i=row_from; i<row_to; i++) \ for (int i=0; i<H; i++) \
for (int j=0; j<W; j++) \ for (int j=0; j<W; j++) \
if (i<start || j<(b) || i>=end || j>=W-(b)) \ if (i<rstart || j<cstart || i>=rend || j>=cend) \
dst[i][j] = src[i][j]; \ dst[i][j] = src[i][j]; \
else \ else \
dst[i][j] = buffer[i][j]; dst[i][j] = buffer[i][j];
#define BL_OPER3(a11,a12,a21,a22) A v = a11*ELEM(-1,-1) + a12*ELEM(-1,0) + a11*ELEM(-1,1) + \ #define BL_OPER3(a11,a12,a21,a22) for (int i=rstart; i<rend; i++) { \
for (int j=cstart; j<cend; j++) { \
A v = a11*ELEM(-1,-1) + a12*ELEM(-1,0) + a11*ELEM(-1,1) + \
a21*ELEM(0,-1) + a22*ELEM(0,0) + a21*ELEM(0,1) + \ a21*ELEM(0,-1) + a22*ELEM(0,0) + a21*ELEM(0,1) + \
a11*ELEM(1,-1) + a12*ELEM(1,0) + a11*ELEM(1,1); \ a11*ELEM(1,-1) + a12*ELEM(1,0) + a11*ELEM(1,1); \
v /= a11*SULY(-1,-1) + a12*SULY(-1,0) + a11*SULY(-1,1) + \ v /= a11*SULY(-1,-1) + a12*SULY(-1,0) + a11*SULY(-1,1) + \
@ -57,7 +60,9 @@
a11*SULY(1,-1) + a12*SULY(1,0) + a11*SULY(1,1); a11*SULY(1,-1) + a12*SULY(1,0) + a11*SULY(1,1);
#define BL_OPER5(a11,a12,a13,a21,a22,a23,a31,a32,a33) A v = a11*ELEM(-2,-2) + a12*ELEM(-2,-1) + a13*ELEM(-2,0) + a12*ELEM(-2,1) + a11*ELEM(-2,2) + \ #define BL_OPER5(a11,a12,a13,a21,a22,a23,a31,a32,a33) for (int i=rstart; i<rend; i++) { \
for (int j=cstart; j<cend; j++) { \
A v = a11*ELEM(-2,-2) + a12*ELEM(-2,-1) + a13*ELEM(-2,0) + a12*ELEM(-2,1) + a11*ELEM(-2,2) + \
a21*ELEM(-1,-2) + a22*ELEM(-1,-1) + a23*ELEM(-1,0) + a22*ELEM(-1,1) + a21*ELEM(-1,2) + \ a21*ELEM(-1,-2) + a22*ELEM(-1,-1) + a23*ELEM(-1,0) + a22*ELEM(-1,1) + a21*ELEM(-1,2) + \
a31*ELEM(0,-2) + a32*ELEM(0,-1) + a33*ELEM(0,0) + a32*ELEM(0,1) + a31*ELEM(0,2) + \ a31*ELEM(0,-2) + a32*ELEM(0,-1) + a33*ELEM(0,0) + a32*ELEM(0,1) + a31*ELEM(0,2) + \
a21*ELEM(1,-2) + a22*ELEM(1,-1) + a23*ELEM(1,0) + a22*ELEM(1,1) + a21*ELEM(1,2) + \ a21*ELEM(1,-2) + a22*ELEM(1,-1) + a23*ELEM(1,0) + a22*ELEM(1,1) + a21*ELEM(1,2) + \
@ -69,7 +74,9 @@
a11*SULY(2,-2) + a12*SULY(2,-1) + a13*SULY(2,0) + a12*SULY(2,1) + a11*SULY(2,2); a11*SULY(2,-2) + a12*SULY(2,-1) + a13*SULY(2,0) + a12*SULY(2,1) + a11*SULY(2,2);
#define BL_OPER7(a11,a12,a13,a14,a21,a22,a23,a24,a31,a32,a33,a34,a41,a42,a43,a44) \ #define BL_OPER7(a11,a12,a13,a14,a21,a22,a23,a24,a31,a32,a33,a34,a41,a42,a43,a44) \
A v = a11*ELEM(-3,-3) + a12*ELEM(-3,-2) + a13*ELEM(-3,-1) + a14*ELEM(-3,0) + a13*ELEM(-3,1) + a12*ELEM(-3,2) + a11*ELEM(-3,3) + \ for (int i=rstart; i<rend; i++) { \
for (int j=cstart; j<cend; j++) { \
A v = a11*ELEM(-3,-3) + a12*ELEM(-3,-2) + a13*ELEM(-3,-1) + a14*ELEM(-3,0) + a13*ELEM(-3,1) + a12*ELEM(-3,2) + a11*ELEM(-3,3) + \
a21*ELEM(-2,-3) + a22*ELEM(-2,-2) + a23*ELEM(-2,-1) + a24*ELEM(-2,0) + a23*ELEM(-2,1) + a22*ELEM(-2,2) + a21*ELEM(-2,3) + \ a21*ELEM(-2,-3) + a22*ELEM(-2,-2) + a23*ELEM(-2,-1) + a24*ELEM(-2,0) + a23*ELEM(-2,1) + a22*ELEM(-2,2) + a21*ELEM(-2,3) + \
a31*ELEM(-1,-3) + a32*ELEM(-1,-2) + a33*ELEM(-1,-1) + a34*ELEM(-1,0) + a33*ELEM(-1,1) + a32*ELEM(-1,2) + a31*ELEM(-1,3) + \ a31*ELEM(-1,-3) + a32*ELEM(-1,-2) + a33*ELEM(-1,-1) + a34*ELEM(-1,0) + a33*ELEM(-1,1) + a32*ELEM(-1,2) + a31*ELEM(-1,3) + \
a41*ELEM(0,-3) + a42*ELEM(0,-2) + a43*ELEM(0,-1) + a44*ELEM(0,0) + a43*ELEM(0,1) + a42*ELEM(0,2) + a41*ELEM(0,3) + \ a41*ELEM(0,-3) + a42*ELEM(0,-2) + a43*ELEM(0,-1) + a44*ELEM(0,0) + a43*ELEM(0,1) + a42*ELEM(0,2) + a41*ELEM(0,3) + \
@ -86,6 +93,8 @@
#define BL_OPER9(a11,a12,a13,a14,a15,a21,a22,a23,a24,a25,a31,a32,a33,a34,a35,a41,a42,a43,a44,a45,a51,a52,a53,a54,a55) \ #define BL_OPER9(a11,a12,a13,a14,a15,a21,a22,a23,a24,a25,a31,a32,a33,a34,a35,a41,a42,a43,a44,a45,a51,a52,a53,a54,a55) \
for (int i=rstart; i<rend; i++) { \
for (int j=cstart; j<cend; j++) { \
A v = a11*ELEM(-4,-4) + a12*ELEM(-4,-3) + a13*ELEM(-4,-2) + a14*ELEM(-4,-1) + a15*ELEM(-4,0) + a14*ELEM(-4,1) + a13*ELEM(-4,2) + a12*ELEM(-4,3) + a11*ELEM(-4,4) + \ A v = a11*ELEM(-4,-4) + a12*ELEM(-4,-3) + a13*ELEM(-4,-2) + a14*ELEM(-4,-1) + a15*ELEM(-4,0) + a14*ELEM(-4,1) + a13*ELEM(-4,2) + a12*ELEM(-4,3) + a11*ELEM(-4,4) + \
a21*ELEM(-3,-4) + a22*ELEM(-3,-3) + a23*ELEM(-3,-2) + a24*ELEM(-3,-1) + a25*ELEM(-3,0) + a24*ELEM(-3,1) + a23*ELEM(-3,2) + a22*ELEM(-3,3) + a21*ELEM(-3,4) + \ a21*ELEM(-3,-4) + a22*ELEM(-3,-3) + a23*ELEM(-3,-2) + a24*ELEM(-3,-1) + a25*ELEM(-3,0) + a24*ELEM(-3,1) + a23*ELEM(-3,2) + a22*ELEM(-3,3) + a21*ELEM(-3,4) + \
a31*ELEM(-2,-4) + a32*ELEM(-2,-3) + a33*ELEM(-2,-2) + a34*ELEM(-2,-1) + a35*ELEM(-2,0) + a34*ELEM(-2,1) + a33*ELEM(-2,2) + a32*ELEM(-2,3) + a31*ELEM(-2,4) + \ a31*ELEM(-2,-4) + a32*ELEM(-2,-3) + a33*ELEM(-2,-2) + a34*ELEM(-2,-1) + a35*ELEM(-2,0) + a34*ELEM(-2,1) + a33*ELEM(-2,2) + a32*ELEM(-2,3) + a31*ELEM(-2,4) + \
@ -106,6 +115,8 @@
a11*SULY(4,-4) + a12*SULY(4,-3) + a13*SULY(4,-2) + a14*SULY(4,-1) + a15*SULY(4,0) + a14*SULY(4,1) + a13*SULY(4,2) + a12*SULY(4,3) + a11*SULY(4,4); a11*SULY(4,-4) + a12*SULY(4,-3) + a13*SULY(4,-2) + a14*SULY(4,-1) + a15*SULY(4,0) + a14*SULY(4,1) + a13*SULY(4,2) + a12*SULY(4,3) + a11*SULY(4,4);
#define BL_OPER11(a11,a12,a13,a14,a15,a16,a21,a22,a23,a24,a25,a26,a31,a32,a33,a34,a35,a36,a41,a42,a43,a44,a45,a46,a51,a52,a53,a54,a55,a56,a61,a62,a63,a64,a65,a66) \ #define BL_OPER11(a11,a12,a13,a14,a15,a16,a21,a22,a23,a24,a25,a26,a31,a32,a33,a34,a35,a36,a41,a42,a43,a44,a45,a46,a51,a52,a53,a54,a55,a56,a61,a62,a63,a64,a65,a66) \
for (int i=rstart; i<rend; i++) { \
for (int j=cstart; j<cend; j++) { \
A v = a11*ELEM(-5,-5) + a12*ELEM(-5,-4) + a13*ELEM(-5,-3) + a14*ELEM(-5,-2) + a15*ELEM(-5,-1) + a16*ELEM(-5,0) + a15*ELEM(-5,1) + a14*ELEM(-5,2) + a13*ELEM(-5,3) + a12*ELEM(-5,4) + a11*ELEM(-5,5) + \ A v = a11*ELEM(-5,-5) + a12*ELEM(-5,-4) + a13*ELEM(-5,-3) + a14*ELEM(-5,-2) + a15*ELEM(-5,-1) + a16*ELEM(-5,0) + a15*ELEM(-5,1) + a14*ELEM(-5,2) + a13*ELEM(-5,3) + a12*ELEM(-5,4) + a11*ELEM(-5,5) + \
a21*ELEM(-4,-5) + a22*ELEM(-4,-4) + a23*ELEM(-4,-3) + a24*ELEM(-4,-2) + a25*ELEM(-4,-1) + a26*ELEM(-4,0) + a25*ELEM(-4,1) + a24*ELEM(-4,2) + a23*ELEM(-4,3) + a22*ELEM(-4,4) + a21*ELEM(-4,5) + \ a21*ELEM(-4,-5) + a22*ELEM(-4,-4) + a23*ELEM(-4,-3) + a24*ELEM(-4,-2) + a25*ELEM(-4,-1) + a26*ELEM(-4,0) + a25*ELEM(-4,1) + a24*ELEM(-4,2) + a23*ELEM(-4,3) + a22*ELEM(-4,4) + a21*ELEM(-4,5) + \
a31*ELEM(-3,-5) + a32*ELEM(-3,-4) + a33*ELEM(-3,-3) + a34*ELEM(-3,-2) + a35*ELEM(-3,-1) + a36*ELEM(-3,0) + a35*ELEM(-3,1) + a34*ELEM(-3,2) + a33*ELEM(-3,3) + a32*ELEM(-3,4) + a31*ELEM(-3,5) + \ a31*ELEM(-3,-5) + a32*ELEM(-3,-4) + a33*ELEM(-3,-3) + a34*ELEM(-3,-2) + a35*ELEM(-3,-1) + a36*ELEM(-3,0) + a35*ELEM(-3,1) + a34*ELEM(-3,2) + a33*ELEM(-3,3) + a32*ELEM(-3,4) + a31*ELEM(-3,5) + \
@ -131,452 +142,256 @@
// sigma = 0.5 // sigma = 0.5
template<class T, class A> void bilateral05 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral05 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(318,1) BL_BEGIN(318,1)
#pragma omp parallel for if (multiThread)
BL_OPER3(1,7,7,55) BL_OPER3(1,7,7,55)
BL_END(1) BL_END(1)
} }
// sigma = 0.6 // sigma = 0.6
template<class T, class A> void bilateral06 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral06 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(768,1) BL_BEGIN(768,1)
#pragma omp parallel for if (multiThread)
BL_OPER3(1,4,4,16) BL_OPER3(1,4,4,16)
BL_END(1) BL_END(1)
} }
// sigma = 0.7 // sigma = 0.7
template<class T, class A> void bilateral07 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral07 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(366,2) BL_BEGIN(366,2)
#pragma omp parallel for if (multiThread)
BL_OPER5(0,0,1,0,8,21,1,21,59) BL_OPER5(0,0,1,0,8,21,1,21,59)
BL_END(2) BL_END(2)
} }
// sigma = 0.8 // sigma = 0.8
template<class T, class A> void bilateral08 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral08 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(753,2) BL_BEGIN(753,2)
#pragma omp parallel for if (multiThread)
BL_OPER5(0,0,1,0,5,10,1,10,23) BL_OPER5(0,0,1,0,5,10,1,10,23)
BL_END(2) BL_END(2)
} }
// sigma = 0.9 // sigma = 0.9
template<class T, class A> void bilateral09 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral09 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(595,2) BL_BEGIN(595,2)
#pragma omp parallel for if (multiThread)
BL_OPER5(0,1,2,1,6,12,2,12,22) BL_OPER5(0,1,2,1,6,12,2,12,22)
BL_END(2) BL_END(2)
} }
// sigma = 1.0 // sigma = 1.0
template<class T, class A> void bilateral10 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral10 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(910,2) BL_BEGIN(910,2)
#pragma omp parallel for if (multiThread)
BL_OPER5(0,1,2,1,4,7,2,7,12) BL_OPER5(0,1,2,1,4,7,2,7,12)
BL_END(2) BL_END(2)
} }
// sigma = 1.1 // sigma = 1.1
template<class T, class A> void bilateral11 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral11 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(209,3) BL_BEGIN(209,3)
#pragma omp parallel for if (multiThread)
BL_OPER7(0,0,1,1,0,2,5,8,1,5,18,27,1,8,27,41) BL_OPER7(0,0,1,1,0,2,5,8,1,5,18,27,1,8,27,41)
BL_END(3) BL_END(3)
} }
// sigma = 1.2 // sigma = 1.2
template<class T, class A> void bilateral12 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral12 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(322,3) BL_BEGIN(322,3)
#pragma omp parallel for if (multiThread)
BL_OPER7(0,0,1,1,0,1,4,6,1,4,11,16,1,6,16,23) BL_OPER7(0,0,1,1,0,1,4,6,1,4,11,16,1,6,16,23)
BL_END(3) BL_END(3)
} }
// sigma = 1.3 // sigma = 1.3
template<class T, class A> void bilateral13 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral13 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(336,3) BL_BEGIN(336,3)
#pragma omp parallel for if (multiThread)
BL_OPER7(0,0,1,1,0,2,4,6,1,4,11,14,1,6,14,19) BL_OPER7(0,0,1,1,0,2,4,6,1,4,11,14,1,6,14,19)
BL_END(3) BL_END(3)
} }
// sigma = 1.4 // sigma = 1.4
template<class T, class A> void bilateral14 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral14 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(195,3) BL_BEGIN(195,3)
#pragma omp parallel for if (multiThread)
BL_OPER7(0,1,2,3,1,4,8,10,2,8,17,21,3,10,21,28) BL_OPER7(0,1,2,3,1,4,8,10,2,8,17,21,3,10,21,28)
BL_END(3) BL_END(3)
} }
// sigma = 1.5 // sigma = 1.5
template<class T, class A> void bilateral15 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral15 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(132,4) BL_BEGIN(132,4)
#pragma omp parallel for if (multiThread)
BL_OPER9(0,0,0,1,1,0,1,2,4,5,0,2,6,12,14,1,4,12,22,28,1,5,14,28,35) BL_OPER9(0,0,0,1,1,0,1,2,4,5,0,2,6,12,14,1,4,12,22,28,1,5,14,28,35)
BL_END(4) BL_END(4)
} }
// sigma = 1.6 // sigma = 1.6
template<class T, class A> void bilateral16 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral16 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(180,4) BL_BEGIN(180,4)
#pragma omp parallel for if (multiThread)
BL_OPER9(0,0,0,1,1,0,1,2,3,4,0,2,5,9,10,1,3,9,15,19,1,4,10,19,23) BL_OPER9(0,0,0,1,1,0,1,2,3,4,0,2,5,9,10,1,3,9,15,19,1,4,10,19,23)
BL_END(4) BL_END(4)
} }
// sigma = 1.7 // sigma = 1.7
template<class T, class A> void bilateral17 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral17 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(195,4) BL_BEGIN(195,4)
#pragma omp parallel for if (multiThread)
BL_OPER9(0,0,1,1,1,0,1,2,3,4,1,2,5,8,9,1,3,8,13,16,1,4,9,16,19) BL_OPER9(0,0,1,1,1,0,1,2,3,4,1,2,5,8,9,1,3,8,13,16,1,4,9,16,19)
BL_END(4) BL_END(4)
} }
// sigma = 1.8 // sigma = 1.8
template<class T, class A> void bilateral18 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral18 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(151,4) BL_BEGIN(151,4)
#pragma omp parallel for if (multiThread)
BL_OPER9(0,0,1,2,2,0,1,3,5,5,1,3,6,10,12,2,5,10,16,19,2,5,12,19,22) BL_OPER9(0,0,1,2,2,0,1,3,5,5,1,3,6,10,12,2,5,10,16,19,2,5,12,19,22)
BL_END(4) BL_END(4)
} }
// sigma = 1.9 // sigma = 1.9
template<class T, class A> void bilateral19 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral19 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(151,4) BL_BEGIN(151,4)
#pragma omp parallel for if (multiThread)
BL_OPER9(0,0,1,2,2,0,1,3,4,5,1,3,5,8,9,2,4,8,12,14,2,5,9,14,16) BL_OPER9(0,0,1,2,2,0,1,3,4,5,1,3,5,8,9,2,4,8,12,14,2,5,9,14,16)
BL_END(4) BL_END(4)
} }
// sigma = 2 // sigma = 2
template<class T, class A> void bilateral20 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral20 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(116,5) BL_BEGIN(116,5)
#pragma omp parallel for if (multiThread)
BL_OPER11(0,0,0,1,1,1,0,0,1,2,3,3,0,1,2,4,7,7,1,2,4,8,12,14,1,3,7,12,18,20,1,3,7,14,20,23) BL_OPER11(0,0,0,1,1,1,0,0,1,2,3,3,0,1,2,4,7,7,1,2,4,8,12,14,1,3,7,12,18,20,1,3,7,14,20,23)
BL_END(5) BL_END(5)
} }
// sigma = 2.1 // sigma = 2.1
template<class T, class A> void bilateral21 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral21 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(127,5) BL_BEGIN(127,5)
#pragma omp parallel for if (multiThread)
BL_OPER11(0,0,0,1,1,1,0,0,1,2,3,3,0,1,2,4,6,7,1,2,4,8,11,12,1,3,6,11,15,17,1,3,7,12,17,19) BL_OPER11(0,0,0,1,1,1,0,0,1,2,3,3,0,1,2,4,6,7,1,2,4,8,11,12,1,3,6,11,15,17,1,3,7,12,17,19)
BL_END(5) BL_END(5)
} }
// sigma = 2.2 // sigma = 2.2
template<class T, class A> void bilateral22 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral22 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(109,5) BL_BEGIN(109,5)
#pragma omp parallel for if (multiThread)
BL_OPER11(0,0,0,1,1,2,0,1,2,3,3,4,1,2,3,5,7,8,1,3,5,9,12,13,1,3,7,12,16,18,2,4,8,13,18,20) BL_OPER11(0,0,0,1,1,2,0,1,2,3,3,4,1,2,3,5,7,8,1,3,5,9,12,13,1,3,7,12,16,18,2,4,8,13,18,20)
BL_END(5) BL_END(5)
} }
// sigma = 2.3 // sigma = 2.3
template<class T, class A> void bilateral23 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral23 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(132,5) BL_BEGIN(132,5)
#pragma omp parallel for if (multiThread)
BL_OPER11(0,0,1,1,1,1,0,1,1,2,3,3,1,1,3,5,6,7,1,2,5,7,10,11,1,3,6,10,13,14,1,3,7,11,14,16) BL_OPER11(0,0,1,1,1,1,0,1,1,2,3,3,1,1,3,5,6,7,1,2,5,7,10,11,1,3,6,10,13,14,1,3,7,11,14,16)
BL_END(5) BL_END(5)
} }
// sigma = 2.4 // sigma = 2.4
template<class T, class A> void bilateral24 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral24 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(156,5) BL_BEGIN(156,5)
#pragma omp parallel for if (multiThread)
BL_OPER11(0,0,1,1,1,1,0,1,1,2,3,3,1,1,3,4,5,6,1,2,4,6,8,9,1,3,5,8,10,11,1,3,6,9,11,12) BL_OPER11(0,0,1,1,1,1,0,1,1,2,3,3,1,1,3,4,5,6,1,2,4,6,8,9,1,3,5,8,10,11,1,3,6,9,11,12)
BL_END(5) BL_END(5)
} }
// sigma = 2.5 // sigma = 2.5
template<class T, class A> void bilateral25 (T** src, T** dst, T** buffer, int W, int H, int row_from, int row_to, double sens) { template<class T, class A> void bilateral25 (T** src, T** dst, T** buffer, int W, int H, double sens, bool multiThread) {
BL_BEGIN(173,5) BL_BEGIN(173,5)
#pragma omp parallel for if (multiThread)
BL_OPER11(0,0,1,1,1,1,0,1,1,2,3,3,1,1,2,4,5,5,1,2,4,5,7,7,1,3,5,7,9,9,1,3,5,7,9,10) BL_OPER11(0,0,1,1,1,1,0,1,1,2,3,3,1,1,2,4,5,5,1,2,4,5,7,7,1,3,5,7,9,9,1,3,5,7,9,10)
BL_END(5) BL_END(5)
} }
class Dim {
public:
int W, H, row_from, row_to;
Dim (int w, int h, int rf, int rt) : W(w), H(h), row_from(rf), row_to(rt) {}
};
// main bilateral filter // main bilateral filter
template<class T, class A> void bilateral (T** src, T** dst, T** buffer, Dim dim, double sigma, double sens) { template<class T, class A> void bilateral (T** src, T** dst, T** buffer, int W, int H, double sigma, double sens, bool multiThread) {
int W = dim.W;
int H = dim.H;
int row_from = dim.row_from;
int row_to = dim.row_to;
if (sigma<0.45) if (sigma<0.45)
for (int i=row_from; i<row_to; i++) { #pragma omp parallel for if (multiThread)
for (int i=0; i<H; i++) {
memcpy (buffer[i], src[i], W*sizeof(T)); memcpy (buffer[i], src[i], W*sizeof(T));
memcpy (dst[i], buffer[i], W*sizeof(T)); memcpy (dst[i], buffer[i], W*sizeof(T));
} }
else if (sigma<0.55) else if (sigma<0.55)
bilateral05<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral05<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<0.65) else if (sigma<0.65)
bilateral06<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral06<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<0.75) else if (sigma<0.75)
bilateral07<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral07<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<0.85) else if (sigma<0.85)
bilateral08<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral08<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<0.95) else if (sigma<0.95)
bilateral09<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral09<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.05) else if (sigma<1.05)
bilateral10<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral10<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.15) else if (sigma<1.15)
bilateral11<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral11<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.25) else if (sigma<1.25)
bilateral12<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral12<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.35) else if (sigma<1.35)
bilateral13<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral13<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.45) else if (sigma<1.45)
bilateral14<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral14<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.55) else if (sigma<1.55)
bilateral15<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral15<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.65) else if (sigma<1.65)
bilateral16<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral16<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.75) else if (sigma<1.75)
bilateral17<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral17<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.85) else if (sigma<1.85)
bilateral18<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral18<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<1.95) else if (sigma<1.95)
bilateral19<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral19<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<2.05) else if (sigma<2.05)
bilateral20<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral20<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<2.15) else if (sigma<2.15)
bilateral21<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral21<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<2.25) else if (sigma<2.25)
bilateral22<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral22<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<2.35) else if (sigma<2.35)
bilateral23<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral23<T,A> (src, dst, buffer, W, H, sens, multiThread);
else if (sigma<2.45) else if (sigma<2.45)
bilateral24<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral24<T,A> (src, dst, buffer, W, H, sens, multiThread);
else else
bilateral25<T,A> (src, dst, buffer, W, H, row_from, row_to, sens); bilateral25<T,A> (src, dst, buffer, W, H, sens, multiThread);
} }
void bilateral_unsigned (unsigned short** src, unsigned short** dst, unsigned short** buffer, Dim dim, double sigma, double sens); // START OF EXPERIMENTAL CODE: O(1) bilateral box filter
void bilateral_signed (short** src, short** dst, short** buffer, Dim dim, double sigma, double sens); // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/*
template<class T> void bilateral (T** src, int** dst, int W, int H, int sigmar, double sigmas) {
time_t t1 = clock ();
int r = 2.0*sigmas + 0.5;
double scaleg = 1024.0;
double MAXINT = 65536.0*65536.0;
double sgmax = 275000/(MAXINT/2.0/sigmar/sigmar+(r+1)*(r+1)/sigmas/sigmas);
printf ("sgmax = %lf\n", sgmax);
if (scaleg>sgmax)
scaleg = sgmax;
// kernel vector for the spatial gaussian filter * 1024
int* kernel = new int[1+2*r];
for (int i=0; i<2*r+1; i++) {
kernel[i] = scaleg / (2.0*sigmas*sigmas) * (i-r)*(i-r) + 0.25;
printf ("kerneli = %d\n", kernel[i]);
}
// exponential lookup table
int scale = (2.0*sigmar*sigmar) / scaleg;
int scalem = 65535/((1+2*r)*(1+2*r));
int ec[256000];
for (int i=0; i<256000; i++)
ec[i] = exp (-i/scaleg) * scalem;
for (int i=r; i<H-r; i++) {
for (int j=r; j<W-r; j++) {
int sum = 0.0;
int val = 0.0;
int c = src[i][j];
for (int x=-r; x<=r; x++)
for (int y=-r; y<=r; y++) {
unsigned int d = src[i-x][j-y];
unsigned int mul = ec[(c-d)*(c-d)/scale + kernel[x+r] + kernel[y+r]];
// if (mul<275000) {
val += d*mul;
sum += mul;
// }
// else {
// printf ("out!!!\n");
// }
}
dst[i][j] = val / sum;
}
}
delete [] kernel;
time_t t2 = clock ();
printf ("bilateral: %d\n", t2-t1);
}
*/
#define MAXVAL 0xffff #define MAXVAL 0xffff
#define CLIP(a) ((a)>0?((a)<MAXVAL?(a):MAXVAL):0) #define CLIP(a) ((a)>0?((a)<MAXVAL?(a):MAXVAL):0)
/*
template<class T> void bilateral (T** src, T** dst, AlignedBuffer<float>* buffer, int W, int H, double sigmar, double sigmas) {
time_t t1 = clock ();
float alpha = 0.5 / sigmar / sigmar;
// buffer for the final image
float** buff_final = new float*[H];
for (int i=0; i<H; i++) {
buff_final[i] = new float[W];
memset (buff_final[i], 0, W*sizeof(float));
}
// buffer for the normalization
float** buff_norm = new float*[H];
for (int i=0; i<H; i++) {
buff_norm[i] = new float[W];
for (int j=0; j<W; j++)
buff_norm[i][j] = 1.0;
}
// temporary buffer
float** buff_temp = new float*[H];
for (int i=0; i<H; i++)
buff_temp[i] = new float[W];
// second order gaussian filter approximation
// first filtered image: temp <-- y_1
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
buff_temp[i][j] = src[i][j];
printf ("1: %g\n", buff_temp[100][100]);
gaussHorizontal_float (buff_temp, buff_temp, buffer, W, 0, H, sigmas);
gaussVertical_float (buff_temp, buff_temp, buffer, H, 0, W, sigmas);
printf ("2: %g\n", buff_temp[100][100]);
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
buff_norm[i][j] += 2.0*alpha*src[i][j]*buff_temp[i][j];
buff_final[i][j] += buff_temp[i][j];
}
printf ("3: %g\n", buff_norm[100][100]);
printf ("4: %g\n", buff_final[100][100]);
// second filtered image: temp <-- y_2
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
buff_temp[i][j] = (float)src[i][j]*(float)src[i][j];
gaussHorizontal_float (buff_temp, buff_temp, buffer, W, 0, H, sigmas);
gaussVertical_float (buff_temp, buff_temp, buffer, H, 0, W, sigmas);
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
buff_norm[i][j] += alpha*(2.0*alpha*src[i][j]*src[i][j]-1.0)*buff_temp[i][j];
buff_final[i][j] += 2.0*alpha*src[i][j]*buff_temp[i][j];
}
printf ("5: %g\n", buff_norm[100][100]);
printf ("6: %g\n", buff_final[100][100]);
// third filtered image: temp <-- y_3
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
buff_temp[i][j] = (float)src[i][j]*(float)src[i][j]*(float)src[i][j];
gaussHorizontal_float (buff_temp, buff_temp, buffer, W, 0, H, sigmas);
gaussVertical_float (buff_temp, buff_temp, buffer, H, 0, W, sigmas);
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
buff_norm[i][j] -= 2.0*alpha*alpha*src[i][j]*(1.0-2.0/3.0*alpha*src[i][j]*src[i][j])*buff_temp[i][j];
buff_final[i][j] += alpha*(2.0*alpha*src[i][j]*src[i][j]-1.0)*buff_temp[i][j];
}
// fourth filtered image: temp <-- y_4
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
buff_temp[i][j] = (float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j];
gaussHorizontal_float (buff_temp, buff_temp, buffer, W, 0, H, sigmas);
gaussVertical_float (buff_temp, buff_temp, buffer, H, 0, W, sigmas);
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
buff_norm[i][j] += alpha*alpha*(0.5-2.0*alpha*src[i][j]*src[i][j])*buff_temp[i][j];
buff_final[i][j] -= 2.0*alpha*alpha*src[i][j]*(1.0-2.0/3.0*alpha*src[i][j]*src[i][j])*buff_temp[i][j];
}
// fifth filtered image: temp <-- y_5
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
buff_temp[i][j] = (float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j];
gaussHorizontal_float (buff_temp, buff_temp, buffer, W, 0, H, sigmas);
gaussVertical_float (buff_temp, buff_temp, buffer, H, 0, W, sigmas);
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
buff_norm[i][j] += alpha*alpha*alpha*src[i][j]*buff_temp[i][j];
buff_final[i][j] += alpha*alpha*(0.5-2.0*alpha*src[i][j]*src[i][j])*buff_temp[i][j];
}
// sixth filtered image: temp <-- y_6
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
buff_temp[i][j] = (float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j];
gaussHorizontal_float (buff_temp, buff_temp, buffer, W, 0, H, sigmas);
gaussVertical_float (buff_temp, buff_temp, buffer, H, 0, W, sigmas);
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
buff_norm[i][j] -= alpha*alpha*alpha/6.0*buff_temp[i][j];
buff_final[i][j] += alpha*alpha*alpha*src[i][j]*buff_temp[i][j];
}
// seventh filtered image: temp <-- y_7, and finalizing
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
buff_temp[i][j] = (float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j]*(float)src[i][j];
gaussHorizontal_float (buff_temp, buff_temp, buffer, W, 0, H, sigmas);
gaussVertical_float (buff_temp, buff_temp, buffer, H, 0, W, sigmas);
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
buff_final[i][j] = (buff_final[i][j] - alpha*alpha*alpha/6.0*buff_temp[i][j]) / buff_norm[i][j];
dst[i][j] = (T)CLIP(buff_final[i][j]);
}
// cleanup
for (int i=0; i<H; i++) {
delete [] buff_temp[i];
delete [] buff_norm[i];
delete [] buff_final[i];
}
delete [] buff_temp;
delete [] buff_norm;
delete [] buff_final;
time_t t2 = clock ();
printf ("bilateral: %d\n", t2-t1);
}
*/
#define BINBIT 12 #define BINBIT 12
#define TRANSBIT 4 #define TRANSBIT 4
template<class T> void bilateral (T** src, T** dst, int W, int H, int sigmar, double sigmas, int row_from, int row_to) { template<class T> void bilateral (T** src, T** dst, int W, int H, int sigmar, double sigmas, int row_from, int row_to) {
// range weights // range weights
@ -654,95 +469,4 @@ template<class T> void bilateral (T** src, T** dst, int W, int H, int sigmar, do
delete [] buff_final[i]; delete [] buff_final[i];
} }
struct bilateralparams {
int row_from, row_to;
};
void bilateral_box_unsigned (unsigned short** src, unsigned short** dst, int W, int H, int sigmar, double sigmas, bilateralparams row);
/*
#define BINBIT 12
#define TRANSBIT 4
#define ADDHIST(a,b) { for (int k=0; k<(1<<BINBIT); k++) (a)[k] = (a)[k]+(b)[k]; }
#define SUBHIST(a,b) { for (int k=0; k<(1<<BINBIT); k++) (a)[k] = (a)[k]-(b)[k]; }
template<class T> void bilateral (T** src, T** dst, AlignedBuffer<float>* buffer, int W, int H, int sigmar, double sigmas) {
time_t t1 = clock ();
unsigned short** rowhist = new unsigned short* [H];
for (int i=0; i<H; i++) {
rowhist[i] = new unsigned short [1<<BINBIT];
memset (rowhist[i], 0, (1<<BINBIT)*sizeof(unsigned short));
}
unsigned short* hist = new unsigned short[1<<BINBIT];
int r = sigmas;
// buffer for the final image
float** buff_final = new float*[H];
for (int i=0; i<H; i++) {
buff_final[i] = new float[W];
memset (buff_final[i], 0, W*sizeof(float));
}
// fill the row histograms initially
for (int i=0; i<H; i++)
for (int j=0; j<2*r+1; j++)
rowhist[i][src[i][j]>>TRANSBIT]++;
// let the game begin...
for (int j=r+1; j<W-r-1; j++) {
// update row histograms
for (int i=0; i<H; i++) {
rowhist[i][src[i][j-r-1]>>TRANSBIT]--;
rowhist[i][src[i][j+r]>>TRANSBIT]++;
}
// sum up upper histograms
memset (hist, 0, (1<<BINBIT)*sizeof(unsigned short));
for (int i=0; i<2*r+1; i++)
ADDHIST(hist, rowhist[i]);
for (int i=r+1; i<H-r-1; i++) {
SUBHIST (hist, rowhist[i-r-1]);
ADDHIST (hist, rowhist[i+r]);
// calculate pixel value
float weight = 0.0;
for (int k=0; k<=(sigmar>>TRANSBIT); k++) {
float w = 1.0 - (double)k/(sigmar>>TRANSBIT);
int v = src[i][j]>>TRANSBIT;
if (v-k >= 0) {
weight += hist [v-k] * w;
buff_final[i][j] += hist [v-k] * w * (src[i][j]-(k<<TRANSBIT));
}
if (v+k < (1<<BINBIT)) {
weight += hist [v+k] * w;
buff_final[i][j] += hist [v+k] * w * (src[i][j]+(k<<TRANSBIT));
}
}
buff_final[i][j] /= weight;
}
}
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
if (i<r+1 || i>H-r-1 || j<r+1 || j>W-r-1)
dst[i][j] = src[i][j];
else
dst[i][j] = (T)CLIP(buff_final[i][j]);
delete [] hist;
for (int i=0; i<H; i++) {
delete [] buff_final[i];
delete [] rowhist[i];
}
delete [] buff_final;
delete [] rowhist;
time_t t2 = clock ();
printf ("bilateral: %d\n", t2-t1);
}
*/
#endif #endif

18
rtengine/dcrop.cc
View File

@ -74,6 +74,8 @@ void Crop::update (int todo, bool internal) {
MyTime t1,t2,t3,t4,t5,t6,t7,t8,t9; MyTime t1,t2,t3,t4,t5,t6,t7,t8,t9;
parent->ipf.setScale (skip);
t1.set (); t1.set ();
// give possibility to the listener to modify crop window (as the full image dimensions are already known at this point) // give possibility to the listener to modify crop window (as the full image dimensions are already known at this point)
int wx, wy, ww, wh, ws; int wx, wy, ww, wh, ws;
@ -130,7 +132,7 @@ void Crop::update (int todo, bool internal) {
} }
if (!resizeCrop) if (!resizeCrop)
resizeCrop = new Image16 (rcw, rch); resizeCrop = new Image16 (rcw, rch);
parent->ipf.resize (origCrop, resizeCrop, params.resize); parent->ipf.resize (origCrop, resizeCrop);
baseCrop = resizeCrop; baseCrop = resizeCrop;
} }
parent->minit.unlock (); parent->minit.unlock ();
@ -173,7 +175,7 @@ void Crop::update (int todo, bool internal) {
if (settings->verbose) printf ("C-BLURMAP: %d\n", t4.etime(t3)); if (settings->verbose) printf ("C-BLURMAP: %d\n", t4.etime(t3));
if (todo & M_RGBCURVE) { if (todo & M_RGBCURVE) {
parent->ipf.rgbProc (baseCrop, laboCrop, &params, parent->tonecurve, cshmap); parent->ipf.rgbProc (baseCrop, laboCrop, parent->tonecurve, cshmap);
} }
t5.set (); t5.set ();
@ -182,8 +184,8 @@ void Crop::update (int todo, bool internal) {
if (todo & M_LUMINANCE) { if (todo & M_LUMINANCE) {
parent->ipf.luminanceCurve (laboCrop, labnCrop, parent->lumacurve, 0, croph); parent->ipf.luminanceCurve (laboCrop, labnCrop, parent->lumacurve, 0, croph);
if (skip==1) { if (skip==1) {
parent->ipf.lumadenoise (labnCrop, &params, 1, cbuffer); parent->ipf.lumadenoise (labnCrop, cbuffer);
parent->ipf.sharpening (labnCrop, &params, 1, (unsigned short**)cbuffer); parent->ipf.sharpening (labnCrop, (unsigned short**)cbuffer);
} }
} }
@ -191,9 +193,9 @@ void Crop::update (int todo, bool internal) {
if (settings->verbose) printf ("C-LUMINANCE: %d\n", t6.etime(t5)); if (settings->verbose) printf ("C-LUMINANCE: %d\n", t6.etime(t5));
if (todo & M_COLOR) { if (todo & M_COLOR) {
parent->ipf.colorCurve (laboCrop, labnCrop, &params); parent->ipf.colorCurve (laboCrop, labnCrop);
if (skip==1) if (skip==1)
parent->ipf.colordenoise (labnCrop, &params, 1, cbuffer); parent->ipf.colordenoise (labnCrop, cbuffer);
} }
t7.set (); t7.set ();
@ -290,7 +292,7 @@ if (settings->verbose) printf ("setcropsizes before lock\n");
// determine which part of the source image is required to compute the crop rectangle // determine which part of the source image is required to compute the crop rectangle
int orx, ory, orw, orh; int orx, ory, orw, orh;
ProcParams& params = parent->params; ProcParams& params = parent->params;
parent->ipf.transCoord (&params, parent->fw, parent->fh, bx1, by1, bw, bh, orx, ory, orw, orh); ImProcFunctions::transCoord (&params, parent->fw, parent->fh, bx1, by1, bw, bh, orx, ory, orw, orh);
int tr = TR_NONE; int tr = TR_NONE;
if (params.coarse.rotate==90) tr |= TR_R90; if (params.coarse.rotate==90) tr |= TR_R90;
@ -324,7 +326,7 @@ if (settings->verbose) printf ("setcropsizes before lock\n");
laboCrop = new LabImage (cropw, croph); laboCrop = new LabImage (cropw, croph);
labnCrop = new LabImage (cropw, croph); labnCrop = new LabImage (cropw, croph);
cropImg = new Image8 (cropw, croph); cropImg = new Image8 (cropw, croph);
cshmap = new SHMap (cropw, croph); cshmap = new SHMap (cropw, croph, true);
cbuffer = new int*[croph]; cbuffer = new int*[croph];
for (int i=0; i<croph; i++) for (int i=0; i<croph; i++)

49
rtengine/gauss.cc
View File

@ -1,49 +0,0 @@
/*
* This file is part of RawTherapee.
*
* Copyright (c) 2004-2010 Gabor Horvath <hgabor@rawtherapee.com>
*
* RawTherapee is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* RawTherapee is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with RawTherapee. If not, see <http://www.gnu.org/licenses/>.
*/
#include <gauss.h>
void gaussHorizontal_unsigned (unsigned short** src, unsigned short** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma) {
gaussHorizontal<unsigned short> (src, dst, buffer, W, row_from, row_to, sigma);
}
void gaussVertical_unsigned (unsigned short** src, unsigned short** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma) {
gaussVertical<unsigned short> (src, dst, buffer, H, col_from, col_to, sigma);
}
void gaussHorizontal_signed (short** src, short** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma) {
gaussHorizontal<short> (src, dst, buffer, W, row_from, row_to, sigma);
}
void gaussVertical_signed (short** src, short** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma) {
gaussVertical<short> (src, dst, buffer, H, col_from, col_to, sigma);
}
void gaussHorizontal_float (float** src, float** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma) {
gaussHorizontal<float> (src, dst, buffer, W, row_from, row_to, sigma);
}
void gaussVertical_float (float** src, float** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma) {
gaussVertical<float> (src, dst, buffer, H, col_from, col_to, sigma);
}

480
rtengine/gauss.h
View File

@ -23,453 +23,45 @@
#include <string.h> #include <string.h>
#include <math.h> #include <math.h>
#include <alignedbuffer.h> #include <alignedbuffer.h>
#include <omp.h>
#define NOSSE 1 // classical filtering if the support window is small:
#ifndef NOSSE template<class T> void gaussHorizontal3 (T** src, T** dst, T* buffer, int W, int H, const float c0, const float c1, bool multiThread) {
#include <xmmintrin.h>
template<class T> void gaussHorizontal3 (T** src, T** dst, T* buffer, int W, int row_from, int row_to, const float c0, const float c1) {
typedef float pfloat[4];
pfloat* temp = (pfloat*)buffer + 1;
pfloat* tmp = (pfloat*)buffer;
__m128 xmm1; xmm1 = _mm_load1_ps (&c0); #pragma omp parallel for if (multiThread)
__m128 xmm2; xmm2 = _mm_load1_ps (&c1); for (int i=0; i<H; i++) {
__m128 xmm3; __m128 xmm4; __m128 xmm5; __m128 xmm6; T* temp = buffer + omp_get_thread_num() * W;
__m128 xmm0;
int i;
for (i = row_from; i<row_to-3; i+=4) {
T* row1 = src[i];
T* row2 = src[i+1];
T* row3 = src[i+2];
T* row4 = src[i+3];
(*tmp)[0] = (float)row1[0];
(*tmp)[1] = (float)row2[0];
(*tmp)[2] = (float)row3[0];
(*tmp)[3] = (float)row4[0];
xmm3 = _mm_load_ps ((float*)tmp); // previous element
(*tmp)[0] = (float)row1[1];
(*tmp)[1] = (float)row2[1];
(*tmp)[2] = (float)row3[1];
(*tmp)[3] = (float)row4[1];
xmm4 = _mm_load_ps ((float*)tmp); // current element
for (int j=1; j<W-1; j++) {
(*tmp)[0] = (float)row1[j+1];
(*tmp)[1] = (float)row2[j+1];
(*tmp)[2] = (float)row3[j+1];
(*tmp)[3] = (float)row4[j+1];
xmm5 = _mm_load_ps ((float*)tmp); // next element
// compute blured pixel
xmm0 = _mm_mul_ps (xmm3, xmm2);
xmm6 = _mm_mul_ps (xmm5, xmm2);
xmm0 = _mm_add_ps (xmm6, xmm0);
xmm6 = _mm_mul_ps (xmm4, xmm1);
xmm0 = _mm_add_ps (xmm6, xmm0);
// store blured pixel
_mm_store_ps ((float*)&temp[j], xmm0);
// update prev and current
xmm3 = xmm4;
xmm4 = xmm5;
}
for (int j=1; j<W-1; j++) {
dst[i][j] = (T)temp[j][0];
dst[i+1][j] = (T)temp[j][1];
dst[i+2][j] = (T)temp[j][2];
dst[i+3][j] = (T)temp[j][3];
}
dst[i][0] = src[i][0];
dst[i+1][0] = src[i+1][0];
dst[i+2][0] = src[i+2][0];
dst[i+3][0] = src[i+3][0];
dst[i][W-1] = src[i][W-1];
dst[i+1][W-1] = src[i+1][W-1];
dst[i+2][W-1] = src[i+2][W-1];
dst[i+3][W-1] = src[i+3][W-1];
}
for (; i<row_to; i++) {
for (int j=1; j<W-1; j++) for (int j=1; j<W-1; j++)
buffer[j] = (T)(c1 * (src[i][j-1] + src[i][j+1]) + c0 * src[i][j]); temp[j] = (T)(c1 * (src[i][j-1] + src[i][j+1]) + c0 * src[i][j]);
dst[i][0] = src[i][0]; dst[i][0] = src[i][0];
memcpy (dst[i]+1, buffer+1, (W-2)*sizeof(T)); memcpy (dst[i]+1, temp+1, (W-2)*sizeof(T));
dst[i][W-1] = src[i][W-1]; dst[i][W-1] = src[i][W-1];
} }
} }
/*template<class T> void gaussHorizontal3 (T** src, T** dst, T* buffer, int W, int row_from, int row_to, const int c0, const int c1) { template<class T> void gaussVertical3 (T** src, T** dst, T* buffer, int W, int H, const float c0, const float c1, bool multiThread) {
time_t t1 = clock (); #pragma omp parallel for if (multiThread)
for (int i=0; i<W; i++) {
const int csum = c0 + 2 * c1; T* temp = buffer + omp_get_thread_num() * H;
for (int i = row_from; i<row_to; i++) {
for (int j=1; j<W-1; j++)
buffer[j] = (c1 * (src[i][j-1] + src[i][j+1]) + c0 * src[i][j]) / csum;
dst[i][0] = src[i][0];
memcpy (dst[i]+1, buffer+1, (W-2)*sizeof(T));
dst[i][W-1] = src[i][W-1];
}
printf ("horizontal time = %d\n", clock()-t1);
}
*/
template<class T> void gaussVertical3 (T** src, T** dst, T* buffer, int H, int col_from, int col_to, const float c0, const float c1) {
typedef float pfloat[4];
pfloat* temp = (pfloat*)buffer + 1;
pfloat* tmp = (pfloat*)buffer;
__m128 xmm1; xmm1 = _mm_load1_ps (&c0);
__m128 xmm2; xmm2 = _mm_load1_ps (&c1);
__m128 xmm3; __m128 xmm4; __m128 xmm5; __m128 xmm6;
__m128 xmm0;
int i;
for (i = col_from; i<col_to-3; i+=4) {
(*tmp)[0] = (float)src[0][i];
(*tmp)[1] = (float)src[0][i+1];
(*tmp)[2] = (float)src[0][i+2];
(*tmp)[3] = (float)src[0][i+3];
xmm3 = _mm_load_ps ((float*)tmp); // previous element
(*tmp)[0] = (float)src[1][i];
(*tmp)[1] = (float)src[1][i+1];
(*tmp)[2] = (float)src[1][i+2];
(*tmp)[3] = (float)src[1][i+3];
xmm4 = _mm_load_ps ((float*)tmp); // current element
for (int j=1; j<H-1; j++) {
(*tmp)[0] = (float)src[j+1][i];
(*tmp)[1] = (float)src[j+1][i+1];
(*tmp)[2] = (float)src[j+1][i+2];
(*tmp)[3] = (float)src[j+1][i+3];
xmm5 = _mm_load_ps ((float*)tmp); // next element
// compute blured pixel
xmm0 = _mm_mul_ps (xmm3, xmm2);
xmm6 = _mm_mul_ps (xmm5, xmm2);
xmm0 = _mm_add_ps (xmm6, xmm0);
xmm6 = _mm_mul_ps (xmm4, xmm1);
xmm0 = _mm_add_ps (xmm6, xmm0);
// store blured pixel
_mm_store_ps ((float*)&temp[j], xmm0);
// update prev and current
xmm3 = xmm4;
xmm4 = xmm5;
}
for (int j=1; j<H-1; j++) {
dst[j][i] = (T)temp[j][0];
dst[j][i+1] = (T)temp[j][1];
dst[j][i+2] = (T)temp[j][2];
dst[j][i+3] = (T)temp[j][3];
}
dst[0][i] = src[0][i];
dst[0][i+1] = src[0][i+1];
dst[0][i+2] = src[0][i+2];
dst[0][i+3] = src[0][i+3];
dst[H-1][i] = src[H-1][i];
dst[H-1][i+1] = src[H-1][i+1];
dst[H-1][i+2] = src[H-1][i+2];
dst[H-1][i+3] = src[H-1][i+3];
}
// int stride = dst[1] - dst[0];
for (; i<col_to; i++) {
for (int j = 1; j<H-1; j++) for (int j = 1; j<H-1; j++)
buffer[j] = (T)(c1 * (src[j-1][i] + src[j+1][i]) + c0 * src[j][i]); temp[j] = (T)(c1 * (src[j-1][i] + src[j+1][i]) + c0 * src[j][i]);
dst[0][i] = src[0][i];
for (int j=1; j<H-1; j++)
dst[j][i] = buffer[j];
// T* crow = &dst[1][i];
// T* cbuff = &buffer[1];
// for (int j = 1; j<H-1; j++, crow += stride, cbuff++)
// (*crow) = (*cbuff);
dst[H-1][i] = src[H-1][i];
}
}
/*
template<class T> void gaussVertical3 (T** src, T** dst, T* buffer, int H, int col_from, int col_to, const float c0, const float c1) {
time_t t1 = clock ();
int stride = dst[1] - dst[0];
for (int j=col_from; j<col_to; j++) {
for (int i = 1; i<H-1; i++)
buffer[i] = (c1 * (src[i-1][j] + src[i+1][j]) + c0 * src[i][j]);
dst[0][j] = src[0][j];
for (int i=1; i<H-1; i++)
dst[i][j] = buffer[i];
// T* crow = &dst[1][j];
// T* cbuff = &buffer[1];
// for (int i = 1; i<H-1; i++, crow += stride, cbuff++)
// (*crow) = (*cbuff);
dst[H-1][j] = src[H-1][j];
}
printf ("vertical time = %d\n", clock()-t1);
}
*/
template<class T> void gaussHorizontal (T** src, T** dst, AlignedBuffer<float>* buffer, int W, int row_from, int row_to, double sigma) {
if (sigma<0.25) {
// dont perform filtering
if (src!=dst)
for (int i = row_from; i<row_to; i++)
memcpy (dst[i], src[i], W*sizeof(T));
return;
}
if (sigma<0.6) {
// compute 3x3 kernel
// double c0 = 2.0 / exp (-1.0 / (2.0 * sigma * sigma));
// printf ("c0=%g\n", c0);
// gaussHorizontal3<T> (src, dst, (T*)(buffer->data), W, row_from, row_to, (int) round(c0), 2);
double c1 = exp (-1.0 / (2.0 * sigma * sigma));
double csum = 2.0 * c1 + 1.0;
c1 /= csum;
double c0 = 1.0 / csum;
gaussHorizontal3<T> (src, dst, (T*)(buffer->data), W, row_from, row_to, c0, c1);
return;
}
// horizontal
float q = 0.98711 * sigma - 0.96330;
if (sigma<2.5)
q = 3.97156 - 4.14554 * sqrt (1.0 - 0.26891 * sigma);
float b0 = 1.57825 + 2.44413*q + 1.4281*q*q + 0.422205*q*q*q;
float b1 = 2.44413*q + 2.85619*q*q + 1.26661*q*q*q;
float b2 = -1.4281*q*q - 1.26661*q*q*q;
float b3 = 0.422205*q*q*q;
float B = 1.0 - (b1+b2+b3) / b0;
b1 /= b0;
b2 /= b0;
b3 /= b0;
// SSE optimized version:
__m128 xmm1; xmm1 = _mm_load1_ps (&B);
__m128 xmm2; xmm2 = _mm_load1_ps (&b1);
__m128 xmm3; xmm3 = _mm_load1_ps (&b2);
__m128 xmm4; xmm4 = _mm_load1_ps (&b3);
__m128 xmm5;
__m128 xmm6;
typedef float pfloat[4];
pfloat* temp = (pfloat*)buffer->data + 1;
pfloat* tmp = (pfloat*)buffer->data;
memset (temp, 0, W*sizeof(pfloat));
int i;
for (i=row_from; i<row_to-3; i+=4) {
T* row1 = src[i];
T* row2 = src[i+1];
T* row3 = src[i+2];
T* row4 = src[i+3];
for (int j=0; j<3; j++) {
(*tmp)[0] = (float)row1[3];
(*tmp)[1] = (float)row2[3];
(*tmp)[2] = (float)row3[3];
(*tmp)[3] = (float)row4[3];
xmm5 = _mm_load_ps ((float*)tmp);
_mm_store_ps ((float*)&temp[j], xmm5);
}
for (int j=3; j<W; j++) {
(*tmp)[0] = (float)row1[j];
(*tmp)[1] = (float)row2[j];
(*tmp)[2] = (float)row3[j];
(*tmp)[3] = (float)row4[j];
xmm5 = _mm_load_ps ((float*)tmp);
xmm5 =_mm_mul_ps (xmm1, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j-1]);
xmm6 = _mm_mul_ps (xmm2, xmm6);
xmm5 = _mm_add_ps (xmm6, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j-2]);
xmm6 = _mm_mul_ps (xmm6, xmm3);
xmm5 = _mm_add_ps (xmm5, xmm6);
xmm6 = _mm_load_ps ((float*)&temp[j-3]);
xmm6 = _mm_mul_ps (xmm6, xmm4);
xmm5 = _mm_add_ps (xmm5, xmm6);
_mm_store_ps ((float*)&temp[j], xmm5);
}
for (int j=W-4; j>=0; j--) {
xmm5 = _mm_load_ps ((float*)&temp[j]);
xmm5 =_mm_mul_ps (xmm1, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j+1]);
xmm6 = _mm_mul_ps (xmm2, xmm6);
xmm5 = _mm_add_ps (xmm6, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j+2]);
xmm6 = _mm_mul_ps (xmm6, xmm3);
xmm5 = _mm_add_ps (xmm5, xmm6);
xmm6 = _mm_load_ps ((float*)&temp[j+3]);
xmm6 = _mm_mul_ps (xmm6, xmm4);
xmm5 = _mm_add_ps (xmm5, xmm6);
_mm_store_ps ((float*)&temp[j], xmm5);
}
for (int j=0; j<W; j++) {
dst[i][j] = (T)temp[j][0];
dst[i+1][j] = (T)temp[j][1];
dst[i+2][j] = (T)temp[j][2];
dst[i+3][j] = (T)temp[j][3];
}
}
// blur remaining rows
float* temp2 = (float*)buffer->data;
for (; i<row_to; i++) {
for (int j=0; j<3; j++)
temp2[j] = src[i][3];
for (int j=3; j<W; j++)
temp2[j] = B * src[i][j] + b1*temp2[j-1] + b2*temp2[j-2] + b3*temp2[j-3];
for (int j=W-4; j>=0; j--)
temp2[j] = B * temp2[j] + b1*temp2[j+1] + b2*temp2[j+2] + b3*temp2[j+3];
for (int j=0; j<W; j++)
dst[i][j] = (T)temp2[j];
}
}
template<class T> void gaussVertical (T** src, T** dst, AlignedBuffer<float>* buffer, int H, int col_from, int col_to, double sigma) {
if (sigma<0.25) {
// dont perform filtering
if (src!=dst)
for (int i = 0; i<H; i++)
for (int j=col_from; j<col_to; j++)
dst[i][j] = src[i][j];
return;
}
if (sigma<0.6) {
// compute 3x3 kernel
double c1 = exp (-1.0 / (2.0 * sigma * sigma));
double csum = 2.0 * c1 + 1.0;
c1 /= csum;
double c0 = 1.0 / csum;
gaussVertical3<T> (src, dst, (T*)(buffer->data), H, col_from, col_to, c0, c1);
// double c0 = 2.0 / exp (-1.0 / (2.0 * sigma * sigma));
// gaussVertical3<T> (src, dst, (T*)(buffer->data), H, col_from, col_to, (int) round(c0), 2);
return;
}
// vertical
float q = 0.98711 * sigma - 0.96330;
if (sigma<2.5)
q = 3.97156 - 4.14554 * sqrt (1.0 - 0.26891 * sigma);
float b0 = 1.57825 + 2.44413*q + 1.4281*q*q + 0.422205*q*q*q;
float b1 = 2.44413*q + 2.85619*q*q + 1.26661*q*q*q;
float b2 = -1.4281*q*q - 1.26661*q*q*q;
float b3 = 0.422205*q*q*q;
float B = 1.0 - (b1+b2+b3) / b0;
b1 /= b0;
b2 /= b0;
b3 /= b0;
// SSE optimized version:
__m128 xmm1; xmm1 = _mm_load1_ps (&B);
__m128 xmm2; xmm2 = _mm_load1_ps (&b1);
__m128 xmm3; xmm3 = _mm_load1_ps (&b2);
__m128 xmm4; xmm4 = _mm_load1_ps (&b3);
__m128 xmm5;
__m128 xmm6;
typedef float pfloat[4];
pfloat* temp = (pfloat*)buffer->data + 1;
pfloat* tmp = (pfloat*)buffer->data;
memset (temp, 0, H*sizeof(pfloat));
int i;
for (i=col_from; i<col_to-3; i+=4) {
for (int j=0; j<3; j++) {
(*tmp)[0] = (float)src[3][i];
(*tmp)[1] = (float)src[3][i+1];
(*tmp)[2] = (float)src[3][i+2];
(*tmp)[3] = (float)src[3][i+3];
xmm5 = _mm_load_ps ((float*)tmp);
_mm_store_ps ((float*)&temp[j], xmm5);
}
for (int j=3; j<H; j++) {
(*tmp)[0] = (float)src[j][i];
(*tmp)[1] = (float)src[j][i+1];
(*tmp)[2] = (float)src[j][i+2];
(*tmp)[3] = (float)src[j][i+3];
xmm5 = _mm_load_ps ((float*)tmp);
xmm5 =_mm_mul_ps (xmm1, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j-1]);
xmm6 = _mm_mul_ps (xmm2, xmm6);
xmm5 = _mm_add_ps (xmm6, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j-2]);
xmm6 = _mm_mul_ps (xmm6, xmm3);
xmm5 = _mm_add_ps (xmm5, xmm6);
xmm6 = _mm_load_ps ((float*)&temp[j-3]);
xmm6 = _mm_mul_ps (xmm6, xmm4);
xmm5 = _mm_add_ps (xmm5, xmm6);
_mm_store_ps ((float*)&temp[j], xmm5);
}
for (int j=H-4; j>=0; j--) {
xmm5 = _mm_load_ps ((float*)&temp[j]);
xmm5 =_mm_mul_ps (xmm1, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j+1]);
xmm6 = _mm_mul_ps (xmm2, xmm6);
xmm5 = _mm_add_ps (xmm6, xmm5);
xmm6 = _mm_load_ps ((float*)&temp[j+2]);
xmm6 = _mm_mul_ps (xmm6, xmm3);
xmm5 = _mm_add_ps (xmm5, xmm6);
xmm6 = _mm_load_ps ((float*)&temp[j+3]);
xmm6 = _mm_mul_ps (xmm6, xmm4);
xmm5 = _mm_add_ps (xmm5, xmm6);
_mm_store_ps ((float*)&temp[j], xmm5);
}
for (int j=0; j<H; j++) {
dst[j][i] = (T)temp[j][0];
dst[j][i+1] = (T)temp[j][1];
dst[j][i+2] = (T)temp[j][2];
dst[j][i+3] = (T)temp[j][3];
}
}
// blur remaining columns
float* temp2 = (float*)buffer->data;
for (; i<col_to; i++) {
for (int j=0; j<3; j++)
temp2[j] = src[3][i];
for (int j=3; j<H; j++)
temp2[j] = B * src[j][i] + b1*temp2[j-1] + b2*temp2[j-2] + b3*temp2[j-3];
for (int j=H-4; j>=0; j--)
temp2[j] = B * temp2[j] + b1*temp2[j+1] + b2*temp2[j+2] + b3*temp2[j+3];
for (int j=0; j<H; j++)
dst[j][i] = (T)temp2[j];
}
}
#else
template<class T> void gaussHorizontal3 (T** src, T** dst, T* buffer, int W, int row_from, int row_to, const float c0, const float c1) {
for (int i=row_from; i<row_to; i++) {
for (int j=1; j<W-1; j++)
buffer[j] = (T)(c1 * (src[i][j-1] + src[i][j+1]) + c0 * src[i][j]);
dst[i][0] = src[i][0];
memcpy (dst[i]+1, buffer+1, (W-2)*sizeof(T));
dst[i][W-1] = src[i][W-1];
}
}
template<class T> void gaussVertical3 (T** src, T** dst, T* buffer, int H, int col_from, int col_to, const float c0, const float c1) {
for (int i=col_from; i<col_to; i++) {
for (int j = 1; j<H-1; j++)
buffer[j] = (T)(c1 * (src[j-1][i] + src[j+1][i]) + c0 * src[j][i]);
dst[0][i] = src[0][i]; dst[0][i] = src[0][i];
for (int j=1; j<H-1; j++) for (int j=1; j<H-1; j++)
dst[j][i] = buffer[j]; dst[j][i] = temp[j];
dst[H-1][i] = src[H-1][i]; dst[H-1][i] = src[H-1][i];
} }
} }
template<class T> void gaussHorizontal (T** src, T** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma) { // fast gaussian approximation if the support window is large
template<class T> void gaussHorizontal (T** src, T** dst, AlignedBuffer<double>* buffer, int W, int H, double sigma, bool multiThread) {
if (sigma<0.25) { if (sigma<0.25) {
// dont perform filtering // dont perform filtering
if (src!=dst) if (src!=dst)
for (int i = row_from; i<row_to; i++) for (int i = 0; i<H; i++)
memcpy (dst[i], src[i], W*sizeof(T)); memcpy (dst[i], src[i], W*sizeof(T));
return; return;
} }
@ -480,11 +72,11 @@ template<class T> void gaussHorizontal (T** src, T** dst, AlignedBuffer<double>*
double csum = 2.0 * c1 + 1.0; double csum = 2.0 * c1 + 1.0;
c1 /= csum; c1 /= csum;
double c0 = 1.0 / csum; double c0 = 1.0 / csum;
gaussHorizontal3<T> (src, dst, (T*)(buffer->data), W, row_from, row_to, c0, c1); gaussHorizontal3<T> (src, dst, (T*)(buffer->data), W, H, c0, c1, multiThread);
return; return;
} }
// horizontal // coefficient calculation
double q = 0.98711 * sigma - 0.96330; double q = 0.98711 * sigma - 0.96330;
if (sigma<2.5) if (sigma<2.5)
q = 3.97156 - 4.14554 * sqrt (1.0 - 0.26891 * sigma); q = 3.97156 - 4.14554 * sqrt (1.0 - 0.26891 * sigma);
@ -513,8 +105,11 @@ template<class T> void gaussHorizontal (T** src, T** dst, AlignedBuffer<double>*
for (int j=0; j<3; j++) for (int j=0; j<3; j++)
M[i][j] /= (1.0+b1-b2+b3)*(1.0+b2+(b1-b3)*b3); M[i][j] /= (1.0+b1-b2+b3)*(1.0+b2+(b1-b3)*b3);
double* temp2 = (double*)buffer->data; #pragma omp parallel for if (multiThread)
for (int i=row_from; i<row_to; i++) { for (int i=0; i<H; i++) {
double* temp2 = buffer->data + omp_get_thread_num() * W;
temp2[0] = B * src[i][0] + b1*src[i][0] + b2*src[i][0] + b3*src[i][0]; temp2[0] = B * src[i][0] + b1*src[i][0] + b2*src[i][0] + b3*src[i][0];
temp2[1] = B * src[i][1] + b1*temp2[0] + b2*src[i][0] + b3*src[i][0]; temp2[1] = B * src[i][1] + b1*temp2[0] + b2*src[i][0] + b3*src[i][0];
temp2[2] = B * src[i][2] + b1*temp2[1] + b2*temp2[0] + b3*src[i][0]; temp2[2] = B * src[i][2] + b1*temp2[1] + b2*temp2[0] + b3*src[i][0];
@ -537,14 +132,13 @@ template<class T> void gaussHorizontal (T** src, T** dst, AlignedBuffer<double>*
} }
} }
template<class T> void gaussVertical (T** src, T** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma) { template<class T> void gaussVertical (T** src, T** dst, AlignedBuffer<double>* buffer, int W, int H, double sigma, bool multiThread) {
if (sigma<0.25) { if (sigma<0.25) {
// dont perform filtering // dont perform filtering
if (src!=dst) if (src!=dst)
for (int i = 0; i<H; i++) for (int i = 0; i<H; i++)
for (int j=col_from; j<col_to; j++) memcpy (dst[i], src[i], W*sizeof(T));
dst[i][j] = src[i][j];
return; return;
} }
@ -554,11 +148,11 @@ template<class T> void gaussVertical (T** src, T** dst, AlignedBuffer<double>* b
double csum = 2.0 * c1 + 1.0; double csum = 2.0 * c1 + 1.0;
c1 /= csum; c1 /= csum;
double c0 = 1.0 / csum; double c0 = 1.0 / csum;
gaussVertical3<T> (src, dst, (T*)(buffer->data), H, col_from, col_to, c0, c1); gaussVertical3<T> (src, dst, (T*)(buffer->data), W, H, c0, c1, multiThread);
return; return;
} }
// vertical // coefficient calculation
double q = 0.98711 * sigma - 0.96330; double q = 0.98711 * sigma - 0.96330;
if (sigma<2.5) if (sigma<2.5)
q = 3.97156 - 4.14554 * sqrt (1.0 - 0.26891 * sigma); q = 3.97156 - 4.14554 * sqrt (1.0 - 0.26891 * sigma);
@ -587,9 +181,12 @@ template<class T> void gaussVertical (T** src, T** dst, AlignedBuffer<double>* b
for (int j=0; j<3; j++) for (int j=0; j<3; j++)
M[i][j] /= (1.0+b1-b2+b3)*(1.0+b2+(b1-b3)*b3); M[i][j] /= (1.0+b1-b2+b3)*(1.0+b2+(b1-b3)*b3);
double* temp2 = (double*)buffer->data; #pragma omp parallel for if (multiThread)
for (int i=col_from; i<col_to; i++) { for (int i=0; i<W; i++) {
temp2[0] = B * src[0][i] + b1*src[0][i] + b2*src[0][i] + b3*src[0][i];
double* temp2 = buffer->data + omp_get_thread_num() * H;
temp2[0] = B * src[0][i] + b1*src[0][i] + b2*src[0][i] + b3*src[0][i];
temp2[1] = B * src[1][i] + b1*temp2[0] + b2*src[0][i] + b3*src[0][i]; temp2[1] = B * src[1][i] + b1*temp2[0] + b2*src[0][i] + b3*src[0][i];
temp2[2] = B * src[2][i] + b1*temp2[1] + b2*temp2[0] + b3*src[0][i]; temp2[2] = B * src[2][i] + b1*temp2[1] + b2*temp2[0] + b3*src[0][i];
@ -611,14 +208,13 @@ template<class T> void gaussVertical (T** src, T** dst, AlignedBuffer<double>* b
dst[j][i] = (T)temp2[j]; dst[j][i] = (T)temp2[j];
} }
} }
#endif
/*
void gaussHorizontal_unsigned (unsigned short** src, unsigned short** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma); void gaussHorizontal_unsigned (unsigned short** src, unsigned short** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma);
void gaussVertical_unsigned (unsigned short** src, unsigned short** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma); void gaussVertical_unsigned (unsigned short** src, unsigned short** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma);
void gaussHorizontal_signed (short** src, short** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma); void gaussHorizontal_signed (short** src, short** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma);
void gaussVertical_signed (short** src, short** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma); void gaussVertical_signed (short** src, short** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma);
void gaussHorizontal_float (float** src, float** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma); void gaussHorizontal_float (float** src, float** dst, AlignedBuffer<double>* buffer, int W, int row_from, int row_to, double sigma);
void gaussVertical_float (float** src, float** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma); void gaussVertical_float (float** src, float** dst, AlignedBuffer<double>* buffer, int H, int col_from, int col_to, double sigma);
*/
#endif #endif

21
rtengine/improccoordinator.cc
View File

@ -28,7 +28,7 @@ namespace rtengine {
extern Settings* settings; extern Settings* settings;
ImProcCoordinator::ImProcCoordinator () ImProcCoordinator::ImProcCoordinator ()
: allocated(false), scale(-1), pW(-1), pH(-1), : ipf(&params, true), allocated(false), scale(-1), pW(-1), pH(-1),
plistener(NULL), imageListener(NULL), aeListener(NULL), hListener(NULL), plistener(NULL), imageListener(NULL), aeListener(NULL), hListener(NULL),
resultValid(false), awbComputed(false), resultValid(false), awbComputed(false),
changeSinceLast(0), updaterRunning(false), changeSinceLast(0), updaterRunning(false),
@ -54,7 +54,6 @@ ImProcCoordinator::~ImProcCoordinator () {
delete toDel[i]; delete toDel[i];
imgsrc->decreaseRef (); imgsrc->decreaseRef ();
ipf.release ();
updaterThreadStart.unlock (); updaterThreadStart.unlock ();
} }
@ -79,6 +78,8 @@ void ImProcCoordinator::updatePreviewImage (int todo) {
else if (params.resize.dataspec==2) else if (params.resize.dataspec==2)
params.resize.scale = (double)params.resize.height / (params.coarse.rotate==90 || params.coarse.rotate==270 ? fw : fh); params.resize.scale = (double)params.resize.height / (params.coarse.rotate==90 || params.coarse.rotate==270 ? fw : fh);
ipf.setScale (scale);
progress ("Applying white balance, color correction & sRBG conversion...",100*readyphase/numofphases); progress ("Applying white balance, color correction & sRBG conversion...",100*readyphase/numofphases);
if (todo & M_INIT) { if (todo & M_INIT) {
minit.lock (); minit.lock ();
@ -164,7 +165,7 @@ void ImProcCoordinator::updatePreviewImage (int todo) {
progress ("Exposure curve & CIELAB conversion...",100*readyphase/numofphases); progress ("Exposure curve & CIELAB conversion...",100*readyphase/numofphases);
if (todo & M_RGBCURVE) { if (todo & M_RGBCURVE) {
CurveFactory::complexCurve (params.toneCurve.expcomp, params.toneCurve.black/65535.0, params.toneCurve.hlcompr, params.toneCurve.shcompr, params.toneCurve.brightness, params.toneCurve.contrast, imgsrc->getDefGain(), imgsrc->getGamma(), true, params.toneCurve.curve, vhist16, tonecurve, bcrgbhist, scale==1 ? 1 : 1); CurveFactory::complexCurve (params.toneCurve.expcomp, params.toneCurve.black/65535.0, params.toneCurve.hlcompr, params.toneCurve.shcompr, params.toneCurve.brightness, params.toneCurve.contrast, imgsrc->getDefGain(), imgsrc->getGamma(), true, params.toneCurve.curve, vhist16, tonecurve, bcrgbhist, scale==1 ? 1 : 1);
ipf.rgbProc (oprevi, oprevl, &params, tonecurve, shmap); ipf.rgbProc (oprevi, oprevl, tonecurve, shmap);
// recompute luminance histogram // recompute luminance histogram
memset (lhist16, 0, 65536*sizeof(int)); memset (lhist16, 0, 65536*sizeof(int));
@ -186,12 +187,12 @@ void ImProcCoordinator::updatePreviewImage (int todo) {
readyphase++; readyphase++;
if (scale==1) { if (scale==1) {
progress ("Denoising luminance...",100*readyphase/numofphases); progress ("Denoising luminance...",100*readyphase/numofphases);
ipf.lumadenoise (nprevl, &params, scale*params.resize.scale, buffer); ipf.lumadenoise (nprevl, buffer);
} }
readyphase++; readyphase++;
if (scale==1) { if (scale==1) {
progress ("Sharpening...",100*readyphase/numofphases); progress ("Sharpening...",100*readyphase/numofphases);
ipf.sharpening (nprevl, &params, scale*params.resize.scale, (unsigned short**)buffer); ipf.sharpening (nprevl, (unsigned short**)buffer);
} }
readyphase++; readyphase++;
} }
@ -201,11 +202,11 @@ void ImProcCoordinator::updatePreviewImage (int todo) {
if (todo & M_COLOR) { if (todo & M_COLOR) {
progress ("Applying Color Boost...",100*readyphase/numofphases); progress ("Applying Color Boost...",100*readyphase/numofphases);
ipf.colorCurve (oprevl, nprevl, &params); ipf.colorCurve (oprevl, nprevl);
readyphase++; readyphase++;
if (scale==1) { if (scale==1) {
progress ("Denoising color...",100*readyphase/numofphases); progress ("Denoising color...",100*readyphase/numofphases);
ipf.colordenoise (nprevl, &params, scale*params.resize.scale, buffer); ipf.colordenoise (nprevl, buffer);
} }
readyphase++; readyphase++;
} }
@ -314,7 +315,7 @@ if (settings->verbose) printf ("setscale before lock\n");
oprevl = new LabImage (pW, pH); oprevl = new LabImage (pW, pH);
nprevl = new LabImage (pW, pH); nprevl = new LabImage (pW, pH);
previmg = new Image8 (pW, pH); previmg = new Image8 (pW, pH);
shmap = new SHMap (pW, pH); shmap = new SHMap (pW, pH, true);
buffer = new int*[pH]; buffer = new int*[pH];
for (int i=0; i<pH; i++) for (int i=0; i<pH; i++)
@ -410,7 +411,7 @@ void ImProcCoordinator::getSpotWB (int x, int y, int rect, double& temp, double&
for (int j=x-rect; j<=x+rect; j++) for (int j=x-rect; j<=x+rect; j++)
points.push_back (Coord2D (j, i)); points.push_back (Coord2D (j, i));
ipf.transCoord (&params, fw, fh, points, red, green, blue); ImProcFunctions::transCoord (&params, fw, fh, points, red, green, blue);
int tr = TR_NONE; int tr = TR_NONE;
if (params.coarse.rotate==90) tr |= TR_R90; if (params.coarse.rotate==90) tr |= TR_R90;
if (params.coarse.rotate==180) tr |= TR_R180; if (params.coarse.rotate==180) tr |= TR_R180;
@ -437,7 +438,7 @@ void ImProcCoordinator::getAutoCrop (double ratio, int &x, int &y, int &w, int &
x = (fullw - w) / 2; x = (fullw - w) / 2;
y = (fullh - h) / 2; y = (fullh - h) / 2;
int orx, ory, orw, orh; int orx, ory, orw, orh;
clipped = ipf.transCoord (&params, fw, fh, x, y, w, h, orx, ory, orw, orh); clipped = ImProcFunctions::transCoord (&params, fw, fh, x, y, w, h, orx, ory, orw, orh);
w -= 4; w -= 4;
} }
if (ratio>0) if (ratio>0)

1929
rtengine/improcfun.cc
View File

File diff suppressed because it is too large Load Diff

86
rtengine/improcfun.h
View File

@ -24,89 +24,75 @@
#include <procparams.h> #include <procparams.h>
#include <shmap.h> #include <shmap.h>
#include <coord2d.h> #include <coord2d.h>
#include <labimage.h>
namespace rtengine { namespace rtengine {
using namespace procparams; using namespace procparams;
class LabImage {
private:
bool fromImage;
public:
int W, H;
unsigned short** L;
short** a;
short** b;
LabImage (int w, int h);
LabImage (Image16* im);
~LabImage ();
};
class ImProcFunctions { class ImProcFunctions {
protected:
struct STemp {
int cx, cy, sx, sy, oW, oH;
};
cmsHTRANSFORM monitorTransform;
void transform_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void simpltransform_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void vignetting_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void transform_sep_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void rgbProc_ (Image16* working, LabImage* lab, const ProcParams* params, int* tonecurve, SHMap* shmap, int row_from, int row_to);
void lab2rgb_ (LabImage* lab, Image8* image, int row_from, int row_to);
void colorCurve_ (LabImage* lold, LabImage* lnew, const ProcParams* params, int row_from, int row_to, double* cmultiplier);
void sharpenHaloCtrl (LabImage* lab, const ProcParams* params, unsigned short** blurmap, unsigned short** base, int W, int row_from, int row_to);
void firstAnalysis_ (Image16* original, Glib::ustring wprofile, unsigned int* histogram, int* chroma_radius, int row_from, int row_to);
void resize_ (Image16* src, Image16* dst, ResizeParams params, int row_from, int row_to);
void damping_ (float** aI, unsigned short** aO, float damping, int W, int rowfrom, int rowto);
public:
static int* cacheL; static int* cacheL;
static int* cachea; static int* cachea;
static int* cacheb; static int* cacheb;
static int* xcache; static int* xcache;
static int* ycache; static int* ycache;
static int* zcache; static int* zcache;
static unsigned short gamma2curve[65536];
struct STemp {
int cx, cy, sx, sy, oW, oH;
};
cmsHTRANSFORM monitorTransform;
int chroma_scale; int chroma_scale;
int chroma_radius; int chroma_radius;
const ProcParams* params;
double scale;
bool multiThread;
void transform_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void simpltransform_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void vignetting_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void transform_sep_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to);
void sharpenHaloCtrl (LabImage* lab, unsigned short** blurmap, unsigned short** base, int W, int H);
void firstAnalysis_ (Image16* original, Glib::ustring wprofile, unsigned int* histogram, int* chroma_radius, int row_from, int row_to);
void dcdamping (float** aI, unsigned short** aO, float damping, int W, int H);
public:
double lumimul[3]; double lumimul[3];
static unsigned short gamma2curve[65536];
static void initCache (); static void initCache ();
ImProcFunctions () : monitorTransform(NULL) {} ImProcFunctions (const ProcParams* iparams, bool imultiThread=true)
void release (); : monitorTransform(NULL), params(iparams), scale(1), multiThread(imultiThread) {}
~ImProcFunctions ();
void setScale (double iscale);
void firstAnalysis (Image16* working, const ProcParams* params, unsigned int* vhist16, double gamma); void firstAnalysis (Image16* working, const ProcParams* params, unsigned int* vhist16, double gamma);
void rgbProc (Image16* working, LabImage* lab, const ProcParams* params, int* tonecurve, SHMap* shmap); void rgbProc (Image16* working, LabImage* lab, int* tonecurve, SHMap* shmap);
void luminanceCurve (LabImage* lold, LabImage* lnew, int* curve, int row_from, int row_to); void luminanceCurve (LabImage* lold, LabImage* lnew, int* curve, int row_from, int row_to);
void colorCurve (LabImage* lold, LabImage* lnew, const ProcParams* params); void colorCurve (LabImage* lold, LabImage* lnew);
void sharpening (LabImage* lab, const ProcParams* params, double scale, unsigned short** buffer); void sharpening (LabImage* lab, unsigned short** buffer);
void lumadenoise (LabImage* lab, const ProcParams* params, double scale, int** buffer); void lumadenoise (LabImage* lab, int** buffer);
void colordenoise (LabImage* lab, const ProcParams* params, double scale, int** buffer); void colordenoise (LabImage* lab, int** buffer);
void transform (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int sx, int sy, int oW, int oH); void transform (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int sx, int sy, int oW, int oH);
void simpltransform (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int sx, int sy, int oW, int oH); void simpltransform (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int sx, int sy, int oW, int oH);
void vignetting (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int oW, int oH); void vignetting (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int oW, int oH);
void lab2rgb (LabImage* lab, Image8* image); void lab2rgb (LabImage* lab, Image8* image);
void resize (Image16* src, Image16* dst, ResizeParams params); void resize (Image16* src, Image16* dst);
bool transCoord (const ProcParams* params, int W, int H, int x, int y, int w, int h, int& xv, int& yv, int& wv, int& hv); void deconvsharpening(LabImage* lab, unsigned short** buffer);
bool transCoord (const ProcParams* params, int W, int H, std::vector<Coord2D> &src, std::vector<Coord2D> &red, std::vector<Coord2D> &green, std::vector<Coord2D> &blue);
void deconvsharpening(LabImage* lab, const ProcParams* params, double scale, unsigned short** buffer);
Image8* lab2rgb (LabImage* lab, int cx, int cy, int cw, int ch, Glib::ustring profile); Image8* lab2rgb (LabImage* lab, int cx, int cy, int cw, int ch, Glib::ustring profile);
Image16* lab2rgb16 (LabImage* lab, int cx, int cy, int cw, int ch, Glib::ustring profile); Image16* lab2rgb16 (LabImage* lab, int cx, int cy, int cw, int ch, Glib::ustring profile);
static bool transCoord (const ProcParams* params, int W, int H, int x, int y, int w, int h, int& xv, int& yv, int& wv, int& hv);
static bool transCoord (const ProcParams* params, int W, int H, std::vector<Coord2D> &src, std::vector<Coord2D> &red, std::vector<Coord2D> &green, std::vector<Coord2D> &blue);
static void getAutoExp (unsigned int* histogram, int histcompr, double expcomp, double clip, double& br, int& bl); static void getAutoExp (unsigned int* histogram, int histcompr, double expcomp, double clip, double& br, int& bl);
}; };
}; };

259
rtengine/iplab2rgb.cc Normal file
View File

@ -0,0 +1,259 @@
/*
* This file is part of RawTherapee.
*
* Copyright (c) 2004-2010 Gabor Horvath <hgabor@rawtherapee.com>
*
* RawTherapee is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* RawTherapee is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with RawTherapee. If not, see <http://www.gnu.org/licenses/>.
*/
#include <rtengine.h>
#include <improcfun.h>
#include <glibmm.h>
#include <iccstore.h>
#include <omp.h>
namespace rtengine {
#undef CLIP
#undef CLIPTO
#undef CMAXVAL
#define CMAXVAL 0xffff
#define CLIP(a) ((a)>0?((a)<CMAXVAL?(a):CMAXVAL):0)
#define CLIPTO(a,b,c) ((a)>(b)?((a)<(c)?(a):(c)):(b))
extern const Settings* settings;
void ImProcFunctions::lab2rgb (LabImage* lab, Image8* image) {
if (monitorTransform) {
int ix = 0;
short* buffer = new short [3*lab->W];
for (int i=0; i<lab->H; i++) {
unsigned short* rL = lab->L[i];
short* ra = lab->a[i];
short* rb = lab->b[i];
int iy = 0;
for (int j=0; j<lab->W; j++) {
int y_ = rL[j];
int x_ = rL[j]+10486+ra[j]*152/chroma_scale+141556;
int z_ = rL[j]+10486-rb[j]*380/chroma_scale+369619;
x_ = CLIPTO(x_,0,369820);
y_ = CLIPTO(y_,0,825745);
y_ = ycache[y_];
x_ = xcache[x_];
z_ = zcache[z_];
buffer[iy++] = CLIP(x_);
buffer[iy++] = CLIP(y_);
buffer[iy++] = CLIP(z_);
}
cmsDoTransform (monitorTransform, buffer, image->data + ix, lab->W);
ix += 3*lab->W;
}
delete [] buffer;
}
else {
#pragma omp parallel for if (multiThread)
for (int i=0; i<lab->H; i++) {
unsigned short* rL = lab->L[i];
short* ra = lab->a[i];
short* rb = lab->b[i];
int ix = 3*i*lab->W;
for (int j=0; j<lab->W; j++) {
int y_ = rL[j];
int x_ = rL[j]+10486+ra[j]*152/chroma_scale+141556;
int z_ = rL[j]+10486-rb[j]*380/chroma_scale+369619;
x_ = CLIPTO(x_,0,369820);
y_ = CLIPTO(y_,0,825745);
y_ = ycache[y_];
x_ = xcache[x_];
z_ = zcache[z_];
/* XYZ-D50 to RGB */
int R = (25689*x_-13261*y_-4022*z_) >> 13;
int G = (-8017*x_+15697*y_+274*z_) >> 13;
int B = (590*x_-1877*y_+11517*z_) >> 13;
/* copy RGB */
image->data[ix++] = gamma2curve[CLIP(R)] >> 8;
image->data[ix++] = gamma2curve[CLIP(G)] >> 8;
image->data[ix++] = gamma2curve[CLIP(B)] >> 8;
}
}
}
}
Image8* ImProcFunctions::lab2rgb (LabImage* lab, int cx, int cy, int cw, int ch, Glib::ustring profile) {
if (cx<0) cx = 0;
if (cy<0) cy = 0;
if (cx+cw>lab->W) cw = lab->W-cx;
if (cy+ch>lab->H) ch = lab->H-cy;
Image8* image = new Image8 (cw, ch);
cmsHPROFILE oprof = iccStore.getProfile (profile);
if (oprof) {
cmsHPROFILE iprof = iccStore.getXYZProfile ();
lcmsMutex->lock ();
cmsHTRANSFORM hTransform = cmsCreateTransform (iprof, TYPE_RGB_16, oprof, TYPE_RGB_8, settings->colorimetricIntent, 0);
lcmsMutex->unlock ();
int ix = 0;
short* buffer = new short [3*cw];
for (int i=cy; i<cy+ch; i++) {
unsigned short* rL = lab->L[i];
short* ra = lab->a[i];
short* rb = lab->b[i];
int iy = 0;
for (int j=cx; j<cx+cw; j++) {
int y_ = rL[j];
int x_ = rL[j]+10486+ra[j]*152/chroma_scale+141556;
int z_ = rL[j]+10486-rb[j]*380/chroma_scale+369619;
x_ = CLIPTO(x_,0,369820);
y_ = CLIPTO(y_,0,825745);
y_ = ycache[y_];
x_ = xcache[x_];
z_ = zcache[z_];
buffer[iy++] = CLIP(x_);
buffer[iy++] = CLIP(y_);
buffer[iy++] = CLIP(z_);
}
cmsDoTransform (hTransform, buffer, image->data + ix, cw);
ix += 3*cw;
}
delete [] buffer;
cmsDeleteTransform(hTransform);
}
else {
#pragma omp parallel for if (multiThread)
for (int i=cy; i<cy+ch; i++) {
unsigned short* rL = lab->L[i];
short* ra = lab->a[i];
short* rb = lab->b[i];
int ix = 3*i*cw;
for (int j=cx; j<cx+cw; j++) {
int y_ = rL[j];
int x_ = rL[j]+10486+ra[j]*152/chroma_scale+141556;
int z_ = rL[j]+10486-rb[j]*380/chroma_scale+369619;
x_ = CLIPTO(x_,0,369820);
y_ = CLIPTO(y_,0,825745);
y_ = ycache[y_];
x_ = xcache[x_];
z_ = zcache[z_];
int R = (25689*x_-13261*y_-4022*z_) >> 13;
int G = (-8017*x_+15697*y_+274*z_) >> 13;
int B = (590*x_-1877*y_+11517*z_) >> 13;
image->data[ix++] = gamma2curve[CLIP(R)] >> 8;
image->data[ix++] = gamma2curve[CLIP(G)] >> 8;
image->data[ix++] = gamma2curve[CLIP(B)] >> 8;
}
}
}
return image;
}
Image16* ImProcFunctions::lab2rgb16 (LabImage* lab, int cx, int cy, int cw, int ch, Glib::ustring profile) {
if (cx<0) cx = 0;
if (cy<0) cy = 0;
if (cx+cw>lab->W) cw = lab->W-cx;
if (cy+ch>lab->H) ch = lab->H-cy;
Image16* image = new Image16 (cw, ch);
cmsHPROFILE oprof = iccStore.getProfile (profile);
if (oprof) {
#pragma omp parallel for if (multiThread)
for (int i=cy; i<cy+ch; i++) {
unsigned short* rL = lab->L[i];
short* ra = lab->a[i];
short* rb = lab->b[i];
short* xa = (short*)image->r[i-cy];
short* ya = (short*)image->g[i-cy];
short* za = (short*)image->b[i-cy];
for (int j=cx; j<cx+cw; j++) {
int y_ = rL[j];
int x_ = rL[j]+10486+ra[j]*152/chroma_scale+141556;
int z_ = rL[j]+10486-rb[j]*380/chroma_scale+369619;
x_ = CLIPTO(x_,0,369820);
y_ = CLIPTO(y_,0,825745);
y_ = ycache[y_];
x_ = xcache[x_];
z_ = zcache[z_];
xa[j-cx] = CLIP(x_);
ya[j-cx] = CLIP(y_);
za[j-cx] = CLIP(z_);
}
}
cmsHPROFILE iprof = iccStore.getXYZProfile ();
lcmsMutex->lock ();
cmsHTRANSFORM hTransform = cmsCreateTransform (iprof, TYPE_RGB_16_PLANAR, oprof, TYPE_RGB_16_PLANAR, settings->colorimetricIntent, 0);
lcmsMutex->unlock ();
cmsDoTransform (hTransform, image->data, image->data, image->planestride/2);
cmsDeleteTransform(hTransform);
}
else {
#pragma omp parallel for if (multiThread)
for (int i=cy; i<cy+ch; i++) {
unsigned short* rL = lab->L[i];
short* ra = lab->a[i];
short* rb = lab->b[i];
for (int j=cx; j<cx+cw; j++) {
int y_ = rL[j];
int x_ = rL[j]+10486+ra[j]*152/chroma_scale+141556;
int z_ = rL[j]+10486-rb[j]*380/chroma_scale+369619;
x_ = CLIPTO(x_,0,369820);
y_ = CLIPTO(y_,0,825745);
y_ = ycache[y_];
x_ = xcache[x_];
z_ = zcache[z_];
int R = (25689*x_-13261*y_-4022*z_) >> 13;
int G = (-8017*x_+15697*y_+274*z_) >> 13;
int B = (590*x_-1877*y_+11517*z_) >> 13;
image->r[i-cy][j-cx] = gamma2curve[CLIP(R)];
image->g[i-cy][j-cx] = gamma2curve[CLIP(G)];
image->b[i-cy][j-cx] = gamma2curve[CLIP(B)];
}
}
}
return image;
}
}

293
rtengine/ipresize.cc Normal file
View File

@ -0,0 +1,293 @@
/*
* This file is part of RawTherapee.
*
* Copyright (c) 2004-2010 Gabor Horvath <hgabor@rawtherapee.com>
*
* RawTherapee is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* RawTherapee is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with RawTherapee. If not, see <http://www.gnu.org/licenses/>.
*/
#include <rtengine.h>
#include <improcfun.h>
#include <glibmm.h>
#include <omp.h>
namespace rtengine {
#undef CLIP
#undef CLIPTO
#undef CMAXVAL
#define CMAXVAL 0xffff
#define CLIP(a) ((a)>0?((a)<CMAXVAL?(a):CMAXVAL):0)
#define CLIPTO(a,b,c) ((a)>(b)?((a)<(c)?(a):(c)):(b))
void ImProcFunctions::resize (Image16* src, Image16* dst) {
if(params->resize.method == "Downscale (Better)") {
// small-scale algorithm by Ilia
// provides much better quality on small scales
// calculates mean value over source pixels which current destination pixel covers
// works only for scales < 1
// for scales ~1 it is analogous to bilinear
// possibly, for even less scale factors (< 0.2 possibly) boundary pixels are not needed, omitting them can give a speedup
// this algorithm is much slower on small factors than others, because it uses all pixels of the SOURCE image
// Ilia Popov ilia_popov@rambler.ru 2010
double delta = 1.0 / params->resize.scale;
double k = params->resize.scale * params->resize.scale;
#pragma omp parallel for if (multiThread)
for(int i = 0; i < dst->height; i++) {
// top and bottom boundary coordinates
double y0 = i * delta;
double y1 = (i + 1) * delta;
int m0 = y0;
m0 = CLIPTO(m0, 0, src->height-1);
int m1 = y1;
m1 = CLIPTO(m1, 0, src->height-1);
// weights of boundary pixels
double wy0 = 1.0 - (y0 - m0);
double wy1 = y1 - m1;
for(int j = 0; j < dst->width; j++) {
// left and right boundary coordinates
double x0 = j * delta;
double x1 = (j + 1) * delta;
int n0 = x0;
n0 = CLIPTO(n0, 0, src->width-1);
int n1 = x1;
n1 = CLIPTO(n1, 0, src->width-1);
double wx0 = 1.0 - (x0 - n0);
double wx1 = x1 - n1;
double r = 0;
double g = 0;
double b = 0;
// integration
// corners
r += wy0 * wx0 * src->r[m0][n0] + wy0 * wx1 * src->r[m0][n1] + wy1 * wx0 * src->r[m1][n0] + wy1 * wx1 * src->r[m1][n1];
g += wy0 * wx0 * src->g[m0][n0] + wy0 * wx1 * src->g[m0][n1] + wy1 * wx0 * src->g[m1][n0] + wy1 * wx1 * src->g[m1][n1];
b += wy0 * wx0 * src->b[m0][n0] + wy0 * wx1 * src->b[m0][n1] + wy1 * wx0 * src->b[m1][n0] + wy1 * wx1 * src->b[m1][n1];
// top and bottom boundaries
for(int n = n0 + 1; n < n1; n++) {
r += wy0 * src->r[m0][n] + wy1 * src->r[m1][n];
g += wy0 * src->g[m0][n] + wy1 * src->g[m1][n];
b += wy0 * src->b[m0][n] + wy1 * src->b[m1][n];
}
// inner rows
for(int m = m0 + 1; m < m1; m++) {
// left and right boundaries
r += wx0 * src->r[m][n0] + wx1 * src->r[m][n1];
g += wx0 * src->g[m][n0] + wx1 * src->g[m][n1];
b += wx0 * src->b[m][n0] + wx1 * src->b[m][n1];
// inner pixels
for(int n = n0 + 1; n < n1; n++) {
r += src->r[m][n];
g += src->g[m][n];
b += src->b[m][n];
}
}
// overall weight is equal to the DST pixel area in SRC coordinates
r *= k;
g *= k;
b *= k;
dst->r[i][j] = CLIP((int)r);
dst->g[i][j] = CLIP((int)g);
dst->b[i][j] = CLIP((int)b);
}
}
return;
}
if(params->resize.method == "Downscale (Faster)")
{
// faster version of algo above, does not take into account border pixels,
// which are summed with non-unity weights in slow algo. So, no need
// for weights at all
// Ilia Popov ilia_popov@rambler.ru 5.04.2010
double delta = 1.0 / params->resize.scale;
int p = (int) delta;
// if actually we are doing upscaling, behave like Nearest
if(p == 0)
p = 1;
int q = p/2;
// may cause problems on 32-bit systems on extremely small factors.
// In that case change 1024 to smth less
const int divider = 1024;
// scaling factor after summation
int k = divider / (p * p);
#pragma omp parallel for if (multiThread)
for(int i = 0; i < dst->height; i++) {
// y coordinate of center of destination pixel
double y = (i + 0.5) * delta;
int m0 = (int) (y) - q;
m0 = CLIPTO(m0, 0, src->height-1);
int m1 = m0 + p;
if(m1 > src->height) {
m1 = src->height;
m0 = m1 - p;
}
m1 = CLIPTO(m1, 0, src->height);
for(int j = 0; j < dst->width; j++) {
// x coordinate of center of destination pixel
double x = (j + 0.5) * delta;
int n0 = (int) (x) - q;
n0 = CLIPTO(n0, 0, src->width-1);
int n1 = n0 + p;
if(n1 > src->width) {
n1 = src->width;
n0 = n1 - p;
}
n1 = CLIPTO(n1, 0, src->width);
int r = 0;
int g = 0;
int b = 0;
// integration
for(int m = m0; m < m1; m++) {
for(int n = n0; n < n1; n++) {
r += src->r[m][n];
g += src->g[m][n];
b += src->b[m][n];
}
}
dst->r[i][j] = CLIP( r * k / divider);
dst->g[i][j] = CLIP( g * k / divider);
dst->b[i][j] = CLIP( b * k / divider);
}
}
return;
}
if (params->resize.method.substr(0,7)=="Bicubic") {
double Av = -0.5;
if (params->resize.method=="Bicubic (Sharper)")
Av = -0.75;
else if (params->resize.method=="Bicubic (Softer)")
Av = -0.25;
#pragma omp parallel for if (multiThread)
for (int i=0; i<dst->height; i++) {
double wx[4], wy[4];
double Dy = i / params->resize.scale;
int yc = (int) Dy; Dy -= (double)yc;
int ys = yc - 1; // smallest y-index used for interpolation
// compute vertical weights
double t1y = -Av*(Dy-1.0)*Dy;
double t2y = (3.0-2.0*Dy)*Dy*Dy;
wy[3] = t1y*Dy;
wy[2] = t1y*(Dy-1.0) + t2y;
wy[1] = -t1y*Dy + 1.0 - t2y;
wy[0] = -t1y*(Dy-1.0);
for (int j=0; j<dst->width; j++) {
double Dx = j / params->resize.scale;
int xc = (int) Dx; Dx -= (double)xc;
int xs = xc - 1; // smallest x-index used for interpolation
if (ys >= 0 && ys <src->height-3 && xs >= 0 && xs <= src->width-3) {
// compute horizontal weights
double t1 = -Av*(Dx-1.0)*Dx;
double t2 = (3.0-2.0*Dx)*Dx*Dx;
wx[3] = t1*Dx;
wx[2] = t1*(Dx-1.0) + t2;
wx[1] = -t1*Dx + 1.0 - t2;
wx[0] = -t1*(Dx-1.0);
// compute weighted sum
int r = 0;
int g = 0;
int b = 0;
for (int x=0; x<4; x++)
for (int y=0; y<4; y++) {
double w = wx[x]*wy[y];
r += w*src->r[ys+y][xs+x];
g += w*src->g[ys+y][xs+x];
b += w*src->b[ys+y][xs+x];
}
dst->r[i][j] = CLIP(r);
dst->g[i][j] = CLIP(g);
dst->b[i][j] = CLIP(b);
}
else {
xc = CLIPTO(xc, 0, src->width-1);
yc = CLIPTO(yc, 0, src->height-1);
int nx = xc + 1;
if (nx>=src->width)
nx = xc;
int ny = yc + 1;
if (ny>=src->height)
ny = yc;
dst->r[i][j] = (1-Dx)*(1-Dy)*src->r[yc][xc] + (1-Dx)*Dy*src->r[ny][xc] + Dx*(1-Dy)*src->r[yc][nx] + Dx*Dy*src->r[ny][nx];
dst->g[i][j] = (1-Dx)*(1-Dy)*src->g[yc][xc] + (1-Dx)*Dy*src->g[ny][xc] + Dx*(1-Dy)*src->g[yc][nx] + Dx*Dy*src->g[ny][nx];
dst->b[i][j] = (1-Dx)*(1-Dy)*src->b[yc][xc] + (1-Dx)*Dy*src->b[ny][xc] + Dx*(1-Dy)*src->b[yc][nx] + Dx*Dy*src->b[ny][nx];
}
}
}
}
else if (params->resize.method=="Bilinear") {
#pragma omp parallel for if (multiThread)
for (int i=0; i<dst->height; i++) {
int sy = i/params->resize.scale;
sy = CLIPTO(sy, 0, src->height-1);
double dy = i/params->resize.scale - sy;
int ny = sy+1;
if (ny>=src->height)
ny = sy;
for (int j=0; j<dst->width; j++) {
int sx = j/params->resize.scale;
sx = CLIPTO(sx, 0, src->width-1);
double dx = j/params->resize.scale - sx;
int nx = sx+1;
if (nx>=src->width)
nx = sx;
dst->r[i][j] = (1-dx)*(1-dy)*src->r[sy][sx] + (1-dx)*dy*src->r[ny][sx] + dx*(1-dy)*src->r[sy][nx] + dx*dy*src->r[ny][nx];
dst->g[i][j] = (1-dx)*(1-dy)*src->g[sy][sx] + (1-dx)*dy*src->g[ny][sx] + dx*(1-dy)*src->g[sy][nx] + dx*dy*src->g[ny][nx];
dst->b[i][j] = (1-dx)*(1-dy)*src->b[sy][sx] + (1-dx)*dy*src->b[ny][sx] + dx*(1-dy)*src->b[sy][nx] + dx*dy*src->b[ny][nx];
}
}
}
else {
#pragma omp parallel for if (multiThread)
for (int i=0; i<dst->height; i++) {
int sy = i/params->resize.scale;
sy = CLIPTO(sy, 0, src->height-1);
for (int j=0; j<dst->width; j++) {
int sx = j/params->resize.scale;
sx = CLIPTO(sx, 0, src->width-1);
dst->r[i][j] = src->r[sy][sx];
dst->g[i][j] = src->g[sy][sx];
dst->b[i][j] = src->b[sy][sx];
}
}
}
}
}

201
rtengine/ipsharpen.cc Normal file
View File

@ -0,0 +1,201 @@
/*
* This file is part of RawTherapee.
*
* Copyright (c) 2004-2010 Gabor Horvath <hgabor@rawtherapee.com>
*
* RawTherapee is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* RawTherapee is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with RawTherapee. If not, see <http://www.gnu.org/licenses/>.
*/
#include <rtengine.h>
#include <improcfun.h>
#include <omp.h>
#include <minmax.h>
#include <gauss.h>
#include <bilateral2.h>
namespace rtengine {
#undef CLIP
#undef CMAXVAL
#undef ABS
#define CMAXVAL 0xffff
#define CLIP(a) ((a)>0?((a)<CMAXVAL?(a):CMAXVAL):0)
#define ABS(a) ((a)<0?-(a):(a))
void ImProcFunctions::dcdamping (float** aI, unsigned short** aO, float damping, int W, int H) {
#pragma omp parallel for if (multiThread)
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
float I = aI[i][j];
float O = (float)aO[i][j];
if (O==0.0 || I==0.0) {
aI[i][j] = 0.0;
continue;
}
float U = -(O * log(I/O) - I + O) * 2.0 / (damping*damping);
U = MIN(U,1.0);
U = U*U*U*U*(5.0-U*4.0);
aI[i][j] = (O - I) / I * U + 1.0;
}
}
void ImProcFunctions::deconvsharpening (LabImage* lab, unsigned short** b2) {
if (params->sharpening.enabled==false || params->sharpening.deconvamount<1)
return;
int W = lab->W, H = lab->H;
float** tmpI = new float*[H];
for (int i=0; i<H; i++) {
tmpI[i] = new float[W];
for (int j=0; j<W; j++)
tmpI[i][j] = (float)lab->L[i][j];
}
float** tmp = (float**)b2;
AlignedBuffer<double>* buffer = new AlignedBuffer<double> (MAX(W,H)*omp_get_max_threads());
float damping = params->sharpening.deconvdamping / 5.0;
bool needdamp = params->sharpening.deconvdamping > 0;
for (int k=0; k<params->sharpening.deconviter; k++) {
// apply blur function (gaussian blur)
gaussHorizontal<float> (tmpI, tmp, buffer, W, H, params->sharpening.deconvradius / scale, multiThread);
gaussVertical<float> (tmp, tmp, buffer, W, H, params->sharpening.deconvradius / scale, multiThread);
if (!needdamp) {
#pragma omp parallel for if (multiThread)
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
if (tmp[i][j]>0)
tmp[i][j] = (float)lab->L[i][j] / tmp[i][j];
}
else
dcdamping (tmp, lab->L, damping, W, H);
gaussHorizontal<float> (tmp, tmp, buffer, W, H, params->sharpening.deconvradius / scale, multiThread);
gaussVertical<float> (tmp, tmp, buffer, W, H, params->sharpening.deconvradius / scale, multiThread);
#pragma omp parallel for if (multiThread)
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
tmpI[i][j] = tmpI[i][j] * tmp[i][j];
}
#pragma omp parallel for if (multiThread)
for (int i=0; i<H; i++)
for (int j=0; j<W; j++)
lab->L[i][j] = lab->L[i][j]*(100-params->sharpening.deconvamount) / 100 + (int)CLIP(tmpI[i][j])*params->sharpening.deconvamount / 100;
for (int i=0; i<H; i++)
delete [] tmpI[i];
delete [] tmpI;
}
void ImProcFunctions::sharpening (LabImage* lab, unsigned short** b2) {
if (params->sharpening.method=="rld") {
deconvsharpening (lab, b2);
return;
}
if (params->sharpening.enabled==false || params->sharpening.amount<1 || lab->W<8 || lab->H<8)
return;
int W = lab->W, H = lab->H;
unsigned short** b3;
AlignedBuffer<double>* buffer = new AlignedBuffer<double> (MAX(W,H)*omp_get_max_threads());
if (params->sharpening.edgesonly==false) {
gaussHorizontal<unsigned short> (lab->L, b2, buffer, W, H, params->sharpening.radius / scale, multiThread);
gaussVertical<unsigned short> (b2, b2, buffer, W, H, params->sharpening.radius / scale, multiThread);
}
else {
b3 = new unsigned short*[H];
for (int i=0; i<H; i++)
b3[i] = new unsigned short[W];
bilateral<unsigned short, unsigned int> (lab->L, (unsigned short**)b3, b2, W, H, params->sharpening.edges_radius / scale, params->sharpening.edges_tolerance, multiThread);
gaussHorizontal<unsigned short> (b3, b2, buffer, W, H, params->sharpening.radius / scale, multiThread);
gaussVertical<unsigned short> (b2, b2, buffer, W, H, params->sharpening.radius / scale, multiThread);
}
delete buffer;
unsigned short** base = lab->L;
if (params->sharpening.edgesonly)
base = b3;
if (params->sharpening.halocontrol==false) {
#pragma omp parallel for if (multiThread)
for (int i=0; i<H; i++)
for (int j=0; j<W; j++) {
int diff = base[i][j] - b2[i][j];
if (ABS(diff)>params->sharpening.threshold) {
int val = lab->L[i][j] + params->sharpening.amount * diff / 100;
lab->L[i][j] = CLIP(val);
}
}
}
else
sharpenHaloCtrl (lab, b2, base, W, H);
if (params->sharpening.edgesonly) {
for (int i=0; i<H; i++)
delete [] b3[i];
delete [] b3;
}
}
void ImProcFunctions::sharpenHaloCtrl (LabImage* lab, unsigned short** blurmap, unsigned short** base, int W, int H) {
int scale = 100 * (100-params->sharpening.halocontrol_amount);
unsigned short** nL = base;
#pragma omp parallel for if (multiThread)
for (int i=2; i<H-2; i++) {
int max1 = 0, max2 = 0, min1 = 0, min2 = 0, maxn, minn, np1, np2, np3, min, max;
for (int j=2; j<W-2; j++) {
int diff = base[i][j] - blurmap[i][j];
if (ABS(diff) > params->sharpening.threshold) {
// compute maximum/minimum in a delta environment
np1 = 2*(nL[i-2][j] + nL[i-2][j+1] + nL[i-2][j+2] + nL[i-1][j] + nL[i-1][j+1] + nL[i-1][j+2] + nL[i][j] + nL[i][j+1] + nL[i][j+2]) / 27 + nL[i-1][j+1] / 3;
np2 = 2*(nL[i-1][j] + nL[i-1][j+1] + nL[i-1][j+2] + nL[i][j] + nL[i][j+1] + nL[i][j+2] + nL[i+1][j] + nL[i+1][j+1] + nL[i+1][j+2]) / 27 + nL[i][j+1] / 3;
np3 = 2*(nL[i][j] + nL[i][j+1] + nL[i][j+2] + nL[i+1][j] + nL[i+1][j+1] + nL[i+1][j+2] + nL[i+2][j] + nL[i+2][j+1] + nL[i+2][j+2]) / 27 + nL[i+1][j+1] / 3;
MINMAX3(np1,np2,np3,maxn,minn);
MAX3(max1,max2,maxn,max);
MIN3(min1,min2,minn,min);
max1 = max2; max2 = maxn;
min1 = min2; min2 = minn;
if (max < lab->L[i][j])
max = lab->L[i][j];
if (min > lab->L[i][j])
min = lab->L[i][j];
int val = lab->L[i][j] + params->sharpening.amount * diff / 100;
int newL = CLIP(val);
// applying halo control
if (newL > max)
newL = max + (newL-max) * scale / 10000;
else if (newL<min)
newL = min - (min-newL) * scale / 10000;
lab->L[i][j] = newL;
}
}
}
}
}

821
rtengine/iptransform.cc Normal file
View File

@ -0,0 +1,821 @@
/*
* This file is part of RawTherapee.
*
* Copyright (c) 2004-2010 Gabor Horvath <hgabor@rawtherapee.com>
*
* RawTherapee is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* RawTherapee is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with RawTherapee. If not, see <http://www.gnu.org/licenses/>.
*/
#include <rtengine.h>
#include <improcfun.h>
#include <omp.h>
namespace rtengine {
#undef CMAXVAL
#undef MAX
#undef MIN
#undef CLIP
#undef CLIPTOC
#define CMAXVAL 0xffff
#define MAX(a,b) ((a)<(b)?(b):(a))
#define MIN(a,b) ((a)>(b)?(b):(a))
#define CLIP(a) ((a)>0?((a)<CMAXVAL?(a):CMAXVAL):0)
#define CLIPTOC(a,b,c,d) ((a)>=(b)?((a)<=(c)?(a):((c),d=true)):((b),d=true))
extern const Settings* settings;
void ImProcFunctions::vignetting_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to) {
int oW = sizes.oW;
int oH = sizes.oH;
int cx = sizes.cx;
int cy = sizes.cy;
double w2 = (double) oW / 2.0 - 0.5;
double h2 = (double) oH / 2.0 - 0.5;
double maxRadius = sqrt( (double)( oW*oW + oH*oH ) ) / 2;
double v = 1.0 - params->vignetting.amount * 3.0 / 400.0;
double b = 1.0 + params->vignetting.radius * 7.0 / 100.0;
double mul = (1.0-v) / tanh(b);
int val;
for (int y=row_from; y<row_to; y++) {
double y_d = (double) (y + cy) - h2 ;
for (int x=0; x<transformed->width; x++) {
double x_d = (double) (x + cx) - w2 ;
double r = sqrt(x_d*x_d + y_d*y_d);
double vign = v + mul * tanh (b*(maxRadius-r) / maxRadius);
val = original->r[y][x] / vign;
transformed->r[y][x] = CLIP(val);
val = original->g[y][x] / vign;
transformed->g[y][x] = CLIP(val);
val = original->b[y][x] / vign;
transformed->b[y][x] = CLIP(val);
}
}
}
void ImProcFunctions::vignetting (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int oW, int oH) {
STemp sizes;
sizes.cx = cx;
sizes.cy = cy;
sizes.oW = oW;
sizes.oH = oH;
if (settings->dualThreadEnabled) {
Glib::Thread *thread1 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::vignetting_), original, transformed, params, sizes, 0, transformed->height/2), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
Glib::Thread *thread2 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::vignetting_), original, transformed, params, sizes, transformed->height/2, transformed->height), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
thread1->join ();
thread2->join ();
}
else
vignetting_ (original, transformed, params, sizes, 0, transformed->height);
}
#include "cubint.cc"
void ImProcFunctions::transform_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to) {
int oW = sizes.oW;
int oH = sizes.oH;
int cx = sizes.cx;
int cy = sizes.cy;
int sx = sizes.sx;
int sy = sizes.sy;
double w2 = (double) oW / 2.0 - 0.5;
double h2 = (double) oH / 2.0 - 0.5;
double cost = cos(params->rotate.degree * 3.14/180.0);
double sint = sin(params->rotate.degree * 3.14/180.0);
double max_x = (double) (sx + original->width - 1);
double max_y = (double) (sy + original->height - 1);
double min_x = (double) sx;
double min_y = (double) sy;
const int n2 = 2;
const int n = 4;
int mix = original->width - 1; // maximum x-index src
int miy = original->height - 1;// maximum y-index src
int mix2 = mix +1 - n;
int miy2 = miy +1 - n;
double scale = (oW>oH) ? (double)oW / 2.0 : (double)oH / 2.0 ;
double radius = sqrt( (double)( oW*oW + oH*oH ) );
radius /= (oW<oH) ? oW : oH;
double a = params->distortion.amount;
double d = 1.0 - a;
// magnify image to keep size
double rotmagn = 1.0;
if (params->rotate.fill) {
double beta = atan((double)MIN(oH,oW)/MAX(oW,oH));
rotmagn = sin(beta) / sin(fabs(params->rotate.degree) * 3.14/180.0 + beta);
}
// 1. check upper and lower border
double d1 = rotmagn - a*h2/scale;
double d2 = rotmagn - a*w2/scale;
double d3 = rotmagn - a*sqrt(h2*h2+w2*w2) / scale;
d = MIN(d,MIN(d1,MIN(d2,d3)));
// auxilary variables for vignetting
double maxRadius = sqrt( (double)( oW*oW + oH*oH ) ) / 2 / scale;
double v = 1.0 - params->vignetting.amount * 3.0 / 400.0;
double b = 1.0 + params->vignetting.radius * 7.0 / 100.0;
double mul = (1.0-v) / tanh(b);
// main cycle
double eps = 1e-10;
bool calc_r=( (fabs(a)>eps) || (fabs(1.0-v)>eps) );
bool do_vign = (fabs(1.0-v)>eps);
for (int y=row_from; y<row_to; y++) {
double y_d = (double) (y + cy) - h2 ;
for (int x=0; x<transformed->width; x++) {
double x_d = (double) (x + cx) - w2 ;
double r=0.0;
double s = d;//10000.0;
if (calc_r)
{
r=(sqrt(x_d*x_d + y_d*y_d)) / scale;
if (r<radius)
s += a * r ;
}
double Dx = s*(x_d * cost - y_d * sint) + w2;
double Dy = s*(x_d * sint + y_d * cost) + h2;
if (fabs(Dx)<eps) Dx = 0;
if (fabs(Dy)<eps) Dy = 0;
if (fabs(Dx-max_x)<eps) Dx = nextafter(max_x,0);
if (fabs(Dy-max_y)<eps) Dy = nextafter(max_y,0);
bool valid = !((Dx >= max_x) || (Dy >= max_y) || (Dx < min_x) || (Dy < min_y));
// Convert only valid pixels
if (valid) {
// Extract integer and fractions of source screen coordinates
int xc = (int) (Dx); Dx -= (double)xc;
int yc = (int) (Dy); Dy -= (double)yc;
int ys = yc +1 - n2 - sy; // smallest y-index used for interpolation
int xs = xc +1 - n2 - sx; // smallest x-index used for interpolation
double vignmul = 1.0;
if (do_vign) vignmul /= (v + mul * tanh (b*(maxRadius-s*r) / maxRadius));
if (ys >= 0 && ys <= miy2 && xs >= 0 && xs <= mix2) // all interpolation pixels inside image
cubint (original, xs, ys, Dx, Dy, &(transformed->r[y][x]), &(transformed->g[y][x]), &(transformed->b[y][x]), vignmul);
else { // edge pixels
int y1 = (yc>0) ? yc : 0;
if (y1>miy) y1 = miy;
int y2 = (yc<miy) ? yc+1 : miy;
if (y2<0) y2 = 0;
int x1 = (xc>0) ? xc : 0;
if (x1>mix) x1 = mix;
int x2 = (xc<mix) ? xc+1 : mix;
if (x2<0) x2 = 0;
int r = vignmul*(original->r[y1][x1]*(1.0-Dx)*(1.0-Dy) + original->r[y1][x2]*Dx*(1.0-Dy) + original->r[y2][x1]*(1.0-Dx)*Dy + original->r[y2][x2]*Dx*Dy);
int g = vignmul*(original->g[y1][x1]*(1.0-Dx)*(1.0-Dy) + original->g[y1][x2]*Dx*(1.0-Dy) + original->g[y2][x1]*(1.0-Dx)*Dy + original->g[y2][x2]*Dx*Dy);
int b = vignmul*(original->b[y1][x1]*(1.0-Dx)*(1.0-Dy) + original->b[y1][x2]*Dx*(1.0-Dy) + original->b[y2][x1]*(1.0-Dx)*Dy + original->b[y2][x2]*Dx*Dy);
transformed->r[y][x] = CLIP(r);
transformed->g[y][x] = CLIP(g);
transformed->b[y][x] = CLIP(b);
}
}
else {
// not valid (source pixel x,y not inside source image, etc.)
transformed->r[y][x] = 0;
transformed->g[y][x] = 0;
transformed->b[y][x] = 0;
}
}
}
}
void ImProcFunctions::simpltransform_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to) {
int oW = sizes.oW;
int oH = sizes.oH;
int cx = sizes.cx;
int cy = sizes.cy;
int sx = sizes.sx;
int sy = sizes.sy;
double w2 = (double) oW / 2.0 - 0.5;
double h2 = (double) oH / 2.0 - 0.5;
double cost = cos(params->rotate.degree * 3.14/180.0);
double sint = sin(params->rotate.degree * 3.14/180.0);
double max_x = (double) (sx + original->width - 1);
double max_y = (double) (sy + original->height - 1);
double min_x = (double) sx;
double min_y = (double) sy;
const int n2 = 2;
const int n = 2;
int mix = original->width - 1; // maximum x-index src
int miy = original->height - 1;// maximum y-index src
int mix2 = mix +1 - n;
int miy2 = miy +1 - n;
double scale = (oW>oH) ? (double)oW / 2.0 : (double)oH / 2.0 ;
double radius = sqrt( (double)( oW*oW + oH*oH ) );
radius /= (oW<oH) ? oW : oH;
double a = params->distortion.amount;
double d = 1.0 - a;
// magnify image to keep size
double rotmagn = 1.0;
if (params->rotate.fill) {
double beta = atan((double)MIN(oH,oW)/MAX(oW,oH));
rotmagn = sin(beta) / sin(fabs(params->rotate.degree) * 3.14/180.0 + beta);
}
// 1. check upper and lower border
double d1r = rotmagn - a*h2/scale - params->cacorrection.red;
double d2r = rotmagn - a*w2/scale - params->cacorrection.red;
double d3r = rotmagn - a*sqrt(h2*h2+w2*w2) / scale - params->cacorrection.red;
double dr = MIN(d,MIN(d1r,MIN(d2r,d3r)));
double d1b = rotmagn - a*h2/scale - params->cacorrection.blue;
double d2b = rotmagn - a*w2/scale - params->cacorrection.blue;
double d3b = rotmagn - a*sqrt(h2*h2+w2*w2) / scale - params->cacorrection.blue;
double db = MIN(d,MIN(d1b,MIN(d2b,d3b)));
double d1g = rotmagn - a*h2/scale;
double d2g = rotmagn - a*w2/scale;
double d3g = rotmagn - a*sqrt(h2*h2+w2*w2) / scale;
double dg = MIN(d,MIN(d1g,MIN(d2g,d3g)));
d = MIN(dg,MIN(dr,db));
// auxilary variables for vignetting
double maxRadius = sqrt( (double)( oW*oW + oH*oH ) ) / 2 / scale;
double v = 1.0 - params->vignetting.amount * 3.0 / 400.0;
double b = 1.0 + params->vignetting.radius * 7.0 / 100.0;
double mul = (1.0-v) / tanh(b);
// main cycle
double eps = 1e-10;
bool calc_r=( (fabs(a)>eps) || (fabs(1.0-v)>eps) );
bool do_vign = (fabs(1.0-v)>eps);
for (int y=row_from; y<row_to; y++) {
double y_d = (double) (y + cy) - h2 ;
for (int x=0; x<transformed->width; x++) {
double x_d = (double) (x + cx) - w2 ;
double r=0.0;
double s = d;//10000.0;
if (calc_r)
{
r=(sqrt(x_d*x_d + y_d*y_d)) / scale;
if (r<radius)
s += a * r ;
}
double Dx = s*(x_d * cost - y_d * sint) + w2;
double Dy = s*(x_d * sint + y_d * cost) + h2;
if (fabs(Dx)<eps) Dx = 0;
if (fabs(Dy)<eps) Dy = 0;
if (fabs(Dx-max_x)<eps) Dx = nextafter(max_x,0);
if (fabs(Dy-max_y)<eps) Dy = nextafter(max_y,0);
bool valid = !((Dx >= max_x) || (Dy >= max_y) || (Dx < min_x) || (Dy < min_y));
// Convert only valid pixels
if (valid) {
// Extract integer and fractions of source screen coordinates
int xc = (int) (Dx); Dx -= (double)xc;
int yc = (int) (Dy); Dy -= (double)yc;
int ys = yc +1 - n2 - sy; // smallest y-index used for interpolation
int xs = xc +1 - n2 - sx; // smallest x-index used for interpolation
double vignmul = 1.0;
if (do_vign) vignmul /= (v + mul * tanh (b*(maxRadius-s*r) / maxRadius));
if (ys >= 0 && ys <= miy2 && xs >= 0 && xs <= mix2 && yc < miy-1) { // all interpolation pixels inside image
int r = vignmul*(original->r[yc][xc]*(1.0-Dx)*(1.0-Dy) + original->r[yc][xc+1]*Dx*(1.0-Dy) + original->r[yc+1][xc]*(1.0-Dx)*Dy + original->r[yc+1][xc+1]*Dx*Dy);
int g = vignmul*(original->g[yc][xc]*(1.0-Dx)*(1.0-Dy) + original->g[yc][xc+1]*Dx*(1.0-Dy) + original->g[yc+1][xc]*(1.0-Dx)*Dy + original->g[yc+1][xc+1]*Dx*Dy);
int b = vignmul*(original->b[yc][xc]*(1.0-Dx)*(1.0-Dy) + original->b[yc][xc+1]*Dx*(1.0-Dy) + original->b[yc+1][xc]*(1.0-Dx)*Dy + original->b[yc+1][xc+1]*Dx*Dy);
transformed->r[y][x] = CLIP(r);
transformed->g[y][x] = CLIP(g);
transformed->b[y][x] = CLIP(b);
}
else { // edge pixels
int y1 = (yc>0) ? yc : 0;
if (y1>miy) y1 = miy;
int y2 = (yc<miy) ? yc+1 : miy;
if (y2<0) y2 = 0;
int x1 = (xc>0) ? xc : 0;
if (x1>mix) x1 = mix;
int x2 = (xc<mix) ? xc+1 : mix;
if (x2<0) x2 = 0;
int r = vignmul*(original->r[y1][x1]*(1.0-Dx)*(1.0-Dy) + original->r[y1][x2]*Dx*(1.0-Dy) + original->r[y2][x1]*(1.0-Dx)*Dy + original->r[y2][x2]*Dx*Dy);
int g = vignmul*(original->g[y1][x1]*(1.0-Dx)*(1.0-Dy) + original->g[y1][x2]*Dx*(1.0-Dy) + original->g[y2][x1]*(1.0-Dx)*Dy + original->g[y2][x2]*Dx*Dy);
int b = vignmul*(original->b[y1][x1]*(1.0-Dx)*(1.0-Dy) + original->b[y1][x2]*Dx*(1.0-Dy) + original->b[y2][x1]*(1.0-Dx)*Dy + original->b[y2][x2]*Dx*Dy);
transformed->r[y][x] = CLIP(r);
transformed->g[y][x] = CLIP(g);
transformed->b[y][x] = CLIP(b);
}
}
else {
// not valid (source pixel x,y not inside source image, etc.)
transformed->r[y][x] = 0;
transformed->g[y][x] = 0;
transformed->b[y][x] = 0;
}
}
}
}
#include "cubintch.cc"
void ImProcFunctions::transform_sep_ (Image16* original, Image16* transformed, const ProcParams* params, STemp sizes, int row_from, int row_to) {
int oW = sizes.oW;
int oH = sizes.oH;
int cx = sizes.cx;
int cy = sizes.cy;
int sx = sizes.sx;
int sy = sizes.sy;
double w2 = (double) oW / 2.0 - 0.5;
double h2 = (double) oH / 2.0 - 0.5;
double cost = cos(params->rotate.degree * 3.14/180.0);
double sint = sin(params->rotate.degree * 3.14/180.0);
double max_x = (double) (sx + original->width - 1);
double max_y = (double) (sy + original->height - 1);
double min_x = (double) sx;
double min_y = (double) sy;
const int n2 = 2;
const int n = 4;
int mix = original->width - 1; // maximum x-index src
int miy = original->height - 1;// maximum y-index src
int mix2 = mix +1 - n;
int miy2 = miy +1 - n;
double scale = (oW>oH) ? (double)oW / 2.0 : (double)oH / 2.0 ;
double radius = sqrt( (double)( oW*oW + oH*oH ) );
radius /= (oW<oH) ? oW : oH;
double a = params->distortion.amount;
double d = 1.0 - a;
double cdist[3];
cdist[0] = params->cacorrection.red;
cdist[1] = 0.0;
cdist[2] = params->cacorrection.blue;
// magnify image to keep size
double rotmagn = 1.0;
if (params->rotate.fill) {
double beta = atan((double)MIN(oH,oW)/MAX(oW,oH));
rotmagn = sin(beta) / sin(fabs(params->rotate.degree) * 3.14/180.0 + beta);
}
// 1. check upper and lower border
double d1r = rotmagn - a*h2/scale - params->cacorrection.red;
double d2r = rotmagn - a*w2/scale - params->cacorrection.red;
double d3r = rotmagn - a*sqrt(h2*h2+w2*w2) / scale - params->cacorrection.red;
double dr = MIN(d,MIN(d1r,MIN(d2r,d3r)));
double d1b = rotmagn - a*h2/scale - params->cacorrection.blue;
double d2b = rotmagn - a*w2/scale - params->cacorrection.blue;
double d3b = rotmagn - a*sqrt(h2*h2+w2*w2) / scale - params->cacorrection.blue;
double db = MIN(d,MIN(d1b,MIN(d2b,d3b)));
double d1g = rotmagn - a*h2/scale;
double d2g = rotmagn - a*w2/scale;
double d3g = rotmagn - a*sqrt(h2*h2+w2*w2) / scale;
double dg = MIN(d,MIN(d1g,MIN(d2g,d3g)));
d = MIN(dg,MIN(dr,db));
unsigned short** chorig[3];
chorig[0] = original->r;
chorig[1] = original->g;
chorig[2] = original->b;
unsigned short** chtrans[3];
chtrans[0] = transformed->r;
chtrans[1] = transformed->g;
chtrans[2] = transformed->b;
// auxilary variables for vignetting
double maxRadius = sqrt( (double)( oW*oW + oH*oH ) ) / 2 / scale;
double v = 1.0 - params->vignetting.amount * 3.0 / 400.0;
double b = 1.0 + params->vignetting.radius * 7.0 / 100.0;
double mul = (1.0-v) / tanh(b);
// main cycle
double eps = 1e-10;
for (int y=row_from; y<row_to; y++) {
double y_d = (double) (y + cy) - h2 ;
for (int x=0; x<transformed->width; x++) {
double x_d = (double) (x + cx) - w2 ;
double r = (sqrt(x_d*x_d + y_d*y_d)) / scale;
double s = 10000.0;
if (r<radius)
s = a * r + d;
double vignmul = 1.0 / (v + mul * tanh (b*(maxRadius-s*r) / maxRadius));
for (int c=0; c<3; c++) {
double Dx = (s + cdist[c]) * (x_d * cost - y_d * sint) + w2;
double Dy = (s + cdist[c]) * (x_d * sint + y_d * cost) + h2;
if (fabs(Dx)<eps) Dx = 0;
if (fabs(Dy)<eps) Dy = 0;
if (fabs(Dx-max_x)<eps) Dx = nextafter(max_x,0);
if (fabs(Dy-max_y)<eps) Dy = nextafter(max_y,0);
bool valid = !((Dx >= max_x) || (Dy >= max_y) || (Dx < min_x) || (Dy < min_y));
// Convert only valid pixels
if (valid) {
// Extract integer and fractions of source screen coordinates
int xc = (int) (Dx); Dx -= (double)xc;
int yc = (int) (Dy); Dy -= (double)yc;
int ys = yc +1 - n2 - sy; // smallest y-index used for interpolation
int xs = xc +1 - n2 - sx; // smallest x-index used for interpolation
if (ys >= 0 && ys <= miy2 && xs >= 0 && xs <= mix2) // all interpolation pixels inside image
cubintch (chorig[c], xs, ys, Dx, Dy, &(chtrans[c][y][x]), vignmul);
else {// edge pixels, linear interpolation
int y1 = (yc>0) ? yc : 0;
if (y1>miy) y1 = miy;
int y2 = (yc<miy) ? yc+1 : miy;
if (y2<0) y2 = 0;
int x1 = (xc>0) ? xc : 0;
if (x1>mix) x1 = mix;
int x2 = (xc<mix) ? xc+1 : mix;
if (x2<0) x2 = 0;
int val = vignmul*(chorig[c][y1][x1]*(1.0-Dx)*(1.0-Dy) + chorig[c][y1][x2]*Dx*(1.0-Dy) + chorig[c][y2][x1]*(1.0-Dx)*Dy + chorig[c][y2][x2]*Dx*Dy);
chtrans[c][y][x] = CLIP(val);
}
}
else // not valid (source pixel x,y not inside source image, etc.)
chtrans[c][y][x] = 0;
}
}
}
}
bool ImProcFunctions::transCoord (const ProcParams* params, int W, int H, std::vector<Coord2D> &src, std::vector<Coord2D> &red, std::vector<Coord2D> &green, std::vector<Coord2D> &blue) {
bool clipresize = true;
bool clipped = false;
red.clear ();
green.clear ();
blue.clear ();
bool needstransform = 0;// fabs(params->rotate.degree)>1e-15 || fabs(params->distortion.amount)>1e-15 || fabs(params->cacorrection.red)>1e-15 || fabs(params->cacorrection.blue)>1e-15;
if (!needstransform) {
if (clipresize) {
// Apply resizing
if (fabs(params->resize.scale-1.0)>=1e-7) {
for (int i=0; i<src.size(); i++) {
red.push_back (Coord2D (src[i].x / params->resize.scale, src[i].y / params->resize.scale));
green.push_back (Coord2D (src[i].x / params->resize.scale, src[i].y / params->resize.scale));
blue.push_back (Coord2D (src[i].x / params->resize.scale, src[i].y / params->resize.scale));
}
for (int i=0; i<src.size(); i++) {
red[i].x = CLIPTOC(red[i].x,0,W-1,clipped);
red[i].y = CLIPTOC(red[i].y,0,H-1,clipped);
green[i].x = CLIPTOC(green[i].x,0,W-1,clipped);
green[i].y = CLIPTOC(green[i].y,0,H-1,clipped);
blue[i].x = CLIPTOC(blue[i].x,0,W-1,clipped);
blue[i].y = CLIPTOC(blue[i].y,0,H-1,clipped);
}
}
else
for (int i=0; i<src.size(); i++) {
red.push_back (Coord2D (src[i].x, src[i].y));
green.push_back (Coord2D (src[i].x, src[i].y));
blue.push_back (Coord2D (src[i].x, src[i].y));
}
}
return clipped;
}
double rW = W*params->resize.scale;
double rH = H*params->resize.scale;
double w2 = (double) rW / 2.0 - 0.5;
double h2 = (double) rH / 2.0 - 0.5;
double cost = cos(params->rotate.degree * 3.14/180.0);
double sint = sin(params->rotate.degree * 3.14/180.0);
double scale = (rW>rH) ? rW / 2.0 : rH / 2.0 ;
double radius = sqrt ((double)(rW*rW + rH*rH ));
radius /= (rW<rH) ? rW : rH;
double a = params->distortion.amount;
double d = 1.0 - a;
// magnify image to keep size
double rotmagn = 1.0;
if (params->rotate.fill) {
double beta = atan(MIN(rH,rW)/MAX(rW,rH));
rotmagn = sin(beta) / sin(fabs(params->rotate.degree) * 3.14/180.0 + beta);
}
if (params->cacorrection.red==0 && params->cacorrection.blue==0) {
// 1. check upper and lower border
double d1 = rotmagn - a*h2/scale;
double d2 = rotmagn - a*w2/scale;
double d3 = rotmagn - a*sqrt(h2*h2+w2*w2) / scale;
d = MIN(d,MIN(d1,MIN(d2,d3)));
for (int i=0; i<src.size(); i++) {
double y_d = src[i].y - h2 ;
double x_d = src[i].x - w2 ;
double r = (sqrt(x_d*x_d + y_d*y_d)) / scale;
double s = 10000.0;
if (r<radius)
s = a * r + d;
red.push_back (Coord2D(s*(x_d * cost - y_d * sint) + w2, s*(x_d * sint + y_d * cost) + h2));
green.push_back (Coord2D(s*(x_d * cost - y_d * sint) + w2, s*(x_d * sint + y_d * cost) + h2));
blue.push_back (Coord2D(s*(x_d * cost - y_d * sint) + w2, s*(x_d * sint + y_d * cost) + h2));
}
}
else {
double cdist[3];
cdist[0] = params->cacorrection.red;
cdist[1] = 0.0;
cdist[2] = params->cacorrection.blue;
// 1. check upper and lower border
double d1r = rotmagn - a*h2/scale - params->cacorrection.red;
double d2r = rotmagn - a*w2/scale - params->cacorrection.red;
double d3r = rotmagn - a*sqrt(h2*h2+w2*w2) / scale - params->cacorrection.red;
double dr = MIN(d,MIN(d1r,MIN(d2r,d3r)));
double d1b = rotmagn - a*h2/scale - params->cacorrection.blue;
double d2b = rotmagn - a*w2/scale - params->cacorrection.blue;
double d3b = rotmagn - a*sqrt(h2*h2+w2*w2) / scale - params->cacorrection.blue;
double db = MIN(d,MIN(d1b,MIN(d2b,d3b)));
double d1g = rotmagn - a*h2/scale;
double d2g = rotmagn - a*w2/scale;
double d3g = rotmagn - a*sqrt(h2*h2+w2*w2) / scale;
double dg = MIN(d,MIN(d1g,MIN(d2g,d3g)));
d = MIN(dg,MIN(dr,db));
for (int i=0; i<src.size(); i++) {
double y_d = src[i].y - h2 ;
double x_d = src[i].x - w2 ;
double r = (sqrt(x_d*x_d + y_d*y_d)) / scale;
double s = 10000.0;
if (r<radius)
s = a * r + d;
src[i].x = s*(x_d * cost - y_d * sint) + w2;
src[i].y = s*(x_d * sint + y_d * cost) + h2;
red.push_back (Coord2D((s+cdist[0])*(x_d * cost - y_d * sint) + w2, (s+cdist[0])*(x_d * sint + y_d * cost) + h2));
green.push_back (Coord2D((s+cdist[1])*(x_d * cost - y_d * sint) + w2, (s+cdist[1])*(x_d * sint + y_d * cost) + h2));
blue.push_back (Coord2D((s+cdist[2])*(x_d * cost - y_d * sint) + w2, (s+cdist[2])*(x_d * sint + y_d * cost) + h2));
}
}
if (clipresize) {
if (fabs(params->resize.scale-1.0)>=1e-7) {
for (int i=0; i<src.size(); i++) {
red[i].x /= params->resize.scale;
red[i].y /= params->resize.scale;
green[i].x /= params->resize.scale;
green[i].y /= params->resize.scale;
blue[i].x /= params->resize.scale;
blue[i].y /= params->resize.scale;
}
}
for (int i=0; i<src.size(); i++) {
red[i].x = CLIPTOC(red[i].x,0,W-1,clipped);
red[i].y = CLIPTOC(red[i].y,0,H-1,clipped);
green[i].x = CLIPTOC(green[i].x,0,W-1,clipped);
green[i].y = CLIPTOC(green[i].y,0,H-1,clipped);
blue[i].x = CLIPTOC(blue[i].x,0,W-1,clipped);
blue[i].y = CLIPTOC(blue[i].y,0,H-1,clipped);
}
}
return clipped;
}
bool ImProcFunctions::transCoord (const ProcParams* params, int W, int H, int x, int y, int w, int h, int& xv, int& yv, int& wv, int& hv) {
int x1 = x, y1 = y;
int x2 = x1 + w - 1;
int y2 = y1 + h - 1;
std::vector<Coord2D> corners (8);
corners[0].set (x1, y1);
corners[1].set (x1, y2);
corners[2].set (x2, y2);
corners[3].set (x2, y1);
corners[4].set ((x1+x2)/2, y1);
corners[5].set ((x1+x2)/2, y2);
corners[6].set (x1, (y1+y2)/2);
corners[7].set (x2, (y1+y2)/2);
std::vector<Coord2D> r, g, b;
bool result = transCoord (params, W, H, corners, r, g, b);
std::vector<Coord2D> transCorners;
transCorners.insert (transCorners.end(), r.begin(), r.end());
transCorners.insert (transCorners.end(), g.begin(), g.end());
transCorners.insert (transCorners.end(), b.begin(), b.end());
double x1d = transCorners[0].x;
for (int i=1; i<transCorners.size(); i++)
if (transCorners[i].x<x1d)
x1d = transCorners[i].x;
int x1v = (int)(x1d);
double y1d = transCorners[0].y;
for (int i=1; i<transCorners.size(); i++)
if (transCorners[i].y<y1d)
y1d = transCorners[i].y;
int y1v = (int)(y1d);
double x2d = transCorners[0].x;
for (int i=1; i<transCorners.size(); i++)
if (transCorners[i].x>x2d)
x2d = transCorners[i].x;
int x2v = (int)ceil(x2d);
double y2d = transCorners[0].y;
for (int i=1; i<transCorners.size(); i++)
if (transCorners[i].y>y2d)
y2d = transCorners[i].y;
int y2v = (int)ceil(y2d);
xv = x1v;
yv = y1v;
wv = x2v - x1v + 1;
hv = y2v - y1v + 1;
return result;
}
void ImProcFunctions::transform (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int sx, int sy, int oW, int oH) {
STemp sizes;
sizes.cx = 0;//cx;
sizes.cy = 0;//cy;
sizes.oW = oW;
sizes.oH = oH;
sizes.sx = 0;//sx;
sizes.sy = 0;//sy;
if (params->cacorrection.red==0 && params->cacorrection.blue==0) {
if (settings->dualThreadEnabled) {
Glib::Thread *thread1 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::transform_), original, transformed, params, sizes, 0, transformed->height/2), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
Glib::Thread *thread2 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::transform_), original, transformed, params, sizes, transformed->height/2, transformed->height), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
thread1->join ();
thread2->join ();
}
else
transform_ (original, transformed, params, sizes, 0, transformed->height);
}
else {
if (settings->dualThreadEnabled) {
Glib::Thread *thread1 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::transform_sep_), original, transformed, params, sizes, 0, transformed->height/2), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
Glib::Thread *thread2 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::transform_sep_), original, transformed, params, sizes, transformed->height/2, transformed->height), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
thread1->join ();
thread2->join ();
}
else
transform_sep_ (original, transformed, params, sizes, 0, transformed->height);
}
}
void ImProcFunctions::simpltransform (Image16* original, Image16* transformed, const ProcParams* params, int cx, int cy, int sx, int sy, int oW, int oH) {
STemp sizes;
sizes.cx = 0;//cx;
sizes.cy = 0;//cy;
sizes.oW = oW;
sizes.oH = oH;
sizes.sx = 0;//sx;
sizes.sy = 0;//sy;
if (settings->dualThreadEnabled) {
Glib::Thread *thread1 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::simpltransform_), original, transformed, params, sizes, 0, transformed->height/2), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
Glib::Thread *thread2 = Glib::Thread::create(sigc::bind(sigc::mem_fun(*this, &ImProcFunctions::simpltransform_), original, transformed, params, sizes, transformed->height/2, transformed->height), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
thread1->join ();
thread2->join ();
}
else
simpltransform_ (original, transformed, params, sizes, 0, transformed->height);
}
/*void ImProcFunctions::transform (Image16* original, Image16* transformed, const ProcParams* params, int ox, int oy) {
if (!transformed)
return;
int oW = W, oH = H, tW = W, tH = H;
double w2 = (double) tW / 2.0 - 0.5;
double h2 = (double) tH / 2.0 - 0.5;
double sw2 = (double) oW / 2.0 - 0.5;
double sh2 = (double) oH / 2.0 - 0.5;
double cost = cos(params->rotate_fine * 3.14/180.0);
double sint = sin(params->rotate_fine * 3.14/180.0);
double max_x = (double) oW;
double max_y = (double) oH;
double min_x = 0.0;
double min_y = 0.0;
const int n2 = 2;
const int n = 4;
int mix = oW - 1; // maximum x-index src
int miy = oH - 1;// maximum y-index src
int mix2 = mix +1 - n;
int miy2 = miy +1 - n;
double scale = (tW>tH) ? (double)tW / 2.0 : (double)tH / 2.0 ;
double radius = sqrt( (double)( tW*tW + tH*tH ) );
radius /= (tW<tH) ? tW : tH;
double a = params->lens_distortion;
for (int y=0; y<transformed->height; y++) {
double y_d = (double) y + oy - h2 ;
for (int x=0; x<transformed->width; x++) {
double x_d = (double) x + ox - w2 ;
double r = (sqrt(x_d*x_d + y_d*y_d)) / scale;
double s = 10000.0;
if (r<radius)
s = a * r + 1.0 - a;
double Dx = s*(x_d * cost - y_d * sint) + sw2;
double Dy = s*(x_d * sint + y_d * cost) + sh2;
bool valid = !((Dx >= max_x) || (Dy >= max_y) || (Dx < min_x) || (Dy < min_y));
// Convert only valid pixels
if (valid) {
// Extract integer and fractions of source screen coordinates
int xc = (int) floor (Dx) ; Dx -= (double)xc;
int yc = (int) floor (Dy) ; Dy -= (double)yc;
int ys = yc +1 - n2 ; // smallest y-index used for interpolation
int xs = xc +1 - n2 ; // smallest x-index used for interpolation
unsigned short sr[2][2], sg[2][2], sb[2][2];
if (ys >= 0 && ys <= miy2 && xs >= 0 && xs <= mix2) // all interpolation pixels inside image
cubint (original, xs, ys, Dx, Dy, &(transformed->r[y][x]), &(transformed->g[y][x]), &(transformed->b[y][x]));
else { // edge pixels
transformed->r[y][x] = 0;
transformed->g[y][x] = 0;
transformed->b[y][x] = 0;
}
}
else {
// not valid (source pixel x,y not inside source image, etc.)
transformed->r[y][x] = 0;
transformed->g[y][x] = 0;
transformed->b[y][x] = 0;
}
}
}
}*/
}

42
rtengine/labimage.cc Normal file
View File

@ -0,0 +1,42 @@
#include <labimage.h>
namespace rtengine {
LabImage::LabImage (int w, int h) : W(w), H(h), fromImage(false) {
L = new unsigned short*[H];
for (int i=0; i<H; i++)
L[i] = new unsigned short[W];
a = new short*[H];
for (int i=0; i<H; i++)
a[i] = new short[W];
b = new short*[H];
for (int i=0; i<H; i++)
b[i] = new short[W];
}
LabImage::LabImage (Image16* im) {
W = im->width;
H = im->height;
L = im->r;
a = (short**) im->g;
b = (short**) im->b;
fromImage = true;
}
LabImage::~LabImage () {
if (!fromImage) {
for (int i=0; i<H; i++) {
delete [] L[i];
delete [] a[i];
delete [] b[i];
}
delete [] L;
delete [] a;
delete [] b;
}
}
}

36
rtengine/bilateral2.cc → rtengine/labimage.h
View File

@ -7,7 +7,7 @@
* it under the terms of the GNU General Public License as published by * it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or * the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version. * (at your option) any later version.
* *
* RawTherapee is distributed in the hope that it will be useful, * RawTherapee is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of * but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
@ -16,20 +16,26 @@
* You should have received a copy of the GNU General Public License * You should have received a copy of the GNU General Public License
* along with RawTherapee. If not, see <http://www.gnu.org/licenses/>. * along with RawTherapee. If not, see <http://www.gnu.org/licenses/>.
*/ */
#include <bilateral2.h> #ifndef _LABIMAGE_H_
#define _LABIMAGE_H_
void bilateral_unsigned (unsigned short** src, unsigned short** dst, unsigned short** buffer, Dim dim, double sigma, double sens) { #include <image16.h>
bilateral<unsigned short, unsigned int> (src, dst, buffer, dim, sigma, sens); namespace rtengine {
class LabImage {
private:
bool fromImage;
public:
int W, H;
unsigned short** L;
short** a;
short** b;
LabImage (int w, int h);
LabImage (Image16* im);
~LabImage ();
};
} }
#endif
void bilateral_signed (short** src, short** dst, short** buffer, Dim dim, double sigma, double sens) {
bilateral<short, int> (src, dst, buffer, dim, sigma, sens);
}
void bilateral_box_unsigned (unsigned short** src, unsigned short** dst, int W, int H, int sigmar, double sigmas, bilateralparams row) {
bilateral<unsigned short> (src, dst, W, H, sigmar, sigmas, row.row_from, row.row_to);
}

16
rtengine/rtthumbnail.cc
View File

@ -206,7 +206,7 @@ IImage8* Thumbnail::processImage (const procparams::ProcParams& params, int rhei
gmi = 1024.0 * gm / mul_lum; gmi = 1024.0 * gm / mul_lum;
bmi = 1024.0 * bm / mul_lum; bmi = 1024.0 * bm / mul_lum;
} }
// resize to requested with and perform coarse transformation // resize to requested width and perform coarse transformation
int rwidth; int rwidth;
if (params.coarse.rotate==90 || params.coarse.rotate==270) { if (params.coarse.rotate==90 || params.coarse.rotate==270) {
rwidth = rheight; rwidth = rheight;
@ -269,7 +269,9 @@ IImage8* Thumbnail::processImage (const procparams::ProcParams& params, int rhei
int fw = baseImg->width; int fw = baseImg->width;
int fh = baseImg->height; int fh = baseImg->height;
ImProcFunctions ipf; ImProcFunctions ipf (&params, false);
ipf.setScale (sqrt(double(fw*fw+fh*fh))/sqrt(double(thumbImg->width*thumbImg->width+thumbImg->height*thumbImg->height))*scale);
unsigned int* hist16 = new unsigned int [65536]; unsigned int* hist16 = new unsigned int [65536];
ipf.firstAnalysis (baseImg, &params, hist16, isRaw ? 2.2 : 0.0); ipf.firstAnalysis (baseImg, &params, hist16, isRaw ? 2.2 : 0.0);
@ -299,7 +301,7 @@ IImage8* Thumbnail::processImage (const procparams::ProcParams& params, int rhei
for (int i=0; i<fh; i++) for (int i=0; i<fh; i++)
buffer[i] = new unsigned short[fw]; buffer[i] = new unsigned short[fw];
} }
shmap = new SHMap (fw, fh); shmap = new SHMap (fw, fh, false);
double radius = sqrt (double(fw*fw+fh*fh)) / 2.0; double radius = sqrt (double(fw*fw+fh*fh)) / 2.0;
double shradius = radius / 1800.0 * params.sh.radius; double shradius = radius / 1800.0 * params.sh.radius;
shmap->update (baseImg, buffer, shradius, ipf.lumimul, params.sh.hq); shmap->update (baseImg, buffer, shradius, ipf.lumimul, params.sh.hq);
@ -321,7 +323,7 @@ IImage8* Thumbnail::processImage (const procparams::ProcParams& params, int rhei
CurveFactory::complexCurve (br, bl/65535.0, params.toneCurve.hlcompr, params.toneCurve.shcompr, params.toneCurve.brightness, params.toneCurve.contrast, logDefGain, isRaw ? 2.2 : 0, true, params.toneCurve.curve, hist16, curve, NULL, 16); CurveFactory::complexCurve (br, bl/65535.0, params.toneCurve.hlcompr, params.toneCurve.shcompr, params.toneCurve.brightness, params.toneCurve.contrast, logDefGain, isRaw ? 2.2 : 0, true, params.toneCurve.curve, hist16, curve, NULL, 16);
LabImage* labView = new LabImage (baseImg); LabImage* labView = new LabImage (baseImg);
ipf.rgbProc (baseImg, labView, &params, curve, shmap); ipf.rgbProc (baseImg, labView, curve, shmap);
if (shmap) if (shmap)
delete shmap; delete shmap;
@ -341,12 +343,11 @@ IImage8* Thumbnail::processImage (const procparams::ProcParams& params, int rhei
delete [] hist16; delete [] hist16;
// color processing // color processing
ipf.colorCurve (labView, labView, &params); ipf.colorCurve (labView, labView);
// obtain final image // obtain final image
Image8* readyImg = new Image8 (fw, fh); Image8* readyImg = new Image8 (fw, fh);
ipf.lab2rgb (labView, readyImg); ipf.lab2rgb (labView, readyImg);
ipf.release ();
delete baseImg; delete baseImg;
// calculate scale // calculate scale
@ -453,8 +454,7 @@ void Thumbnail::getSpotWB (const procparams::ProcParams& params, int xp, int yp,
fw = thumbImg->height; fw = thumbImg->height;
fh = thumbImg->width; fh = thumbImg->width;
} }
ImProcFunctions ipf; ImProcFunctions::transCoord (&params, fw, fh, points, red, green, blue);
ipf.transCoord (&params, fw, fh, points, red, green, blue);
int tr = TR_NONE; int tr = TR_NONE;
if (params.coarse.rotate==90) tr |= TR_R90; if (params.coarse.rotate==90) tr |= TR_R90;
if (params.coarse.rotate==180) tr |= TR_R180; if (params.coarse.rotate==180) tr |= TR_R180;

61
rtengine/shmap.cc
View File

@ -29,7 +29,7 @@ namespace rtengine {
extern const Settings* settings; extern const Settings* settings;
SHMap::SHMap (int w, int h) : W(w), H(h) { SHMap::SHMap (int w, int h, bool multiThread) : W(w), H(h), multiThread(multiThread) {
map = new unsigned short*[H]; map = new unsigned short*[H];
for (int i=0; i<H; i++) for (int i=0; i<H; i++)
@ -52,51 +52,27 @@ void SHMap::update (Image16* img, unsigned short** buffer, double radius, double
map[i][j] = CLIP(val); map[i][j] = CLIP(val);
} }
//MyTime t1,t2;
//t1.set ();
if (!hq) { if (!hq) {
AlignedBuffer<double>* buffer = new AlignedBuffer<double> (MAX(W,H)*omp_get_max_threads());
AlignedBuffer<double>* buffer1 = new AlignedBuffer<double> (MAX(W,H)*5); gaussHorizontal<unsigned short> (map, map, buffer, W, H, radius, multiThread);
AlignedBuffer<double>* buffer2 = new AlignedBuffer<double> (MAX(W,H)*5); gaussVertical<unsigned short> (map, map, buffer, W, H, radius, multiThread);
// blur delete buffer;
if (settings->dualThreadEnabled) {
Glib::Thread *thread1 = Glib::Thread::create(sigc::bind(sigc::ptr_fun(gaussHorizontal_unsigned), map, map, buffer1, W, 0, H/2, radius), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
Glib::Thread *thread2 = Glib::Thread::create(sigc::bind(sigc::ptr_fun(gaussHorizontal_unsigned), map, map, buffer2, W, H/2, H, radius), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
thread1->join ();
thread2->join ();
thread1 = Glib::Thread::create(sigc::bind(sigc::ptr_fun(gaussVertical_unsigned), map, map, buffer1, H, 0, W/2, radius), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
thread2 = Glib::Thread::create(sigc::bind(sigc::ptr_fun(gaussVertical_unsigned), map, map, buffer2, H, W/2, W, radius), 0, true, true, Glib::THREAD_PRIORITY_NORMAL);
thread1->join ();
thread2->join ();
}
else {
gaussHorizontal_unsigned (map, map, buffer1, W, 0, H, radius);
gaussVertical_unsigned (map, map, buffer1, H, 0, W, radius);
}
delete buffer1;
delete buffer2;
} }
else { else {
if (settings->dualThreadEnabled) { #pragma omp parallel if (multiThread)
bilateralparams r1, r2; {
r1.row_from = 0; int tid = omp_get_thread_num();
r1.row_to = H/2; int nthreads = omp_get_num_threads();
r2.row_from = H/2; int blk = H/nthreads;
r2.row_to = H;
Glib::Thread *thread1 = Glib::Thread::create(sigc::bind(sigc::ptr_fun(bilateral_box_unsigned), map, buffer, W, H, 8000, radius, r1), 0, true, true, Glib::THREAD_PRIORITY_NORMAL); if (tid<nthreads-1)
Glib::Thread *thread2 = Glib::Thread::create(sigc::bind(sigc::ptr_fun(bilateral_box_unsigned), map, buffer, W, H, 8000, radius, r2), 0, true, true, Glib::THREAD_PRIORITY_NORMAL); bilateral<unsigned short> (map, buffer, W, H, 8000, radius, tid*blk, (tid+1)*blk);
thread1->join (); else
thread2->join (); bilateral<unsigned short> (map, buffer, W, H, 8000, radius, tid*blk, H);
} }
else { // anti-alias filtering the result
bilateralparams r1;
r1.row_from = 0;
r1.row_to = H;
bilateral_box_unsigned (map, buffer, W, H, 8000, radius, r1);
}
for (int i=0; i<H; i++) for (int i=0; i<H; i++)
for (int j=0; j<W; j++) for (int j=0; j<W; j++)
if (i>0 && j>0 && i<H-1 && j<W-1) if (i>0 && j>0 && i<H-1 && j<W-1)
@ -105,9 +81,6 @@ void SHMap::update (Image16* img, unsigned short** buffer, double radius, double
map[i][j] = buffer[i][j]; map[i][j] = buffer[i][j];
} }
// t2.set ();
// printf ("shmap: %d\n", t2.etime (t1));
// update average, minimum, maximum // update average, minimum, maximum
double _avg = 0; double _avg = 0;
int n = 1; int n = 1;

3
rtengine/shmap.h
View File

@ -29,8 +29,9 @@ class SHMap {
int W, H; int W, H;
unsigned short** map; unsigned short** map;
unsigned short max, min, avg; unsigned short max, min, avg;
bool multiThread;
SHMap (int w, int h); SHMap (int w, int h, bool multiThread);
~SHMap (); ~SHMap ();
void update (Image16* img, unsigned short** buffer, double radius, double lumi[3], bool hq); void update (Image16* img, unsigned short** buffer, double radius, double lumi[3], bool hq);

17
rtengine/simpleprocess.cc
View File

@ -69,7 +69,7 @@ IImage16* processImage (ProcessingJob* pjob, int& errorCode, ProgressListener* p
int fw, fh; int fw, fh;
imgsrc->getFullSize (fw, fh, tr); imgsrc->getFullSize (fw, fh, tr);
ImProcFunctions ipf; ImProcFunctions ipf (&params, true);
Image16* baseImg; Image16* baseImg;
PreviewProps pp (0, 0, fw, fh, 1); PreviewProps pp (0, 0, fw, fh, 1);
@ -83,7 +83,7 @@ IImage16* processImage (ProcessingJob* pjob, int& errorCode, ProgressListener* p
fw *= params.resize.scale; fw *= params.resize.scale;
fh *= params.resize.scale; fh *= params.resize.scale;
baseImg = new Image16 (fw, fh); baseImg = new Image16 (fw, fh);
ipf.resize (oorig, baseImg, params.resize); ipf.resize (oorig, baseImg);
delete oorig; delete oorig;
} }
if (pl) if (pl)
@ -117,7 +117,7 @@ IImage16* processImage (ProcessingJob* pjob, int& errorCode, ProgressListener* p
SHMap* shmap = NULL; SHMap* shmap = NULL;
if (params.sh.enabled) { if (params.sh.enabled) {
shmap = new SHMap (fw, fh); shmap = new SHMap (fw, fh, true);
double radius = sqrt (double(fw*fw+fh*fh)) / 2.0; double radius = sqrt (double(fw*fw+fh*fh)) / 2.0;
double shradius = radius / 1800.0 * params.sh.radius; double shradius = radius / 1800.0 * params.sh.radius;
shmap->update (baseImg, (unsigned short**)buffer, shradius, ipf.lumimul, params.sh.hq); shmap->update (baseImg, (unsigned short**)buffer, shradius, ipf.lumimul, params.sh.hq);
@ -138,7 +138,7 @@ IImage16* processImage (ProcessingJob* pjob, int& errorCode, ProgressListener* p
CurveFactory::complexCurve (br, bl/65535.0, params.toneCurve.hlcompr, params.toneCurve.shcompr, params.toneCurve.brightness, params.toneCurve.contrast, imgsrc->getDefGain(), imgsrc->getGamma(), true, params.toneCurve.curve, hist16, curve, NULL); CurveFactory::complexCurve (br, bl/65535.0, params.toneCurve.hlcompr, params.toneCurve.shcompr, params.toneCurve.brightness, params.toneCurve.contrast, imgsrc->getDefGain(), imgsrc->getGamma(), true, params.toneCurve.curve, hist16, curve, NULL);
LabImage* labView = new LabImage (baseImg); LabImage* labView = new LabImage (baseImg);
ipf.rgbProc (baseImg, labView, &params, curve, shmap); ipf.rgbProc (baseImg, labView, curve, shmap);
if (shmap) if (shmap)
delete shmap; delete shmap;
@ -155,15 +155,15 @@ IImage16* processImage (ProcessingJob* pjob, int& errorCode, ProgressListener* p
// luminance processing // luminance processing
CurveFactory::complexCurve (0.0, 0.0, 0.0, 0.0, params.lumaCurve.brightness, params.lumaCurve.contrast, 0.0, 0.0, false, params.lumaCurve.curve, hist16, curve, NULL); CurveFactory::complexCurve (0.0, 0.0, 0.0, 0.0, params.lumaCurve.brightness, params.lumaCurve.contrast, 0.0, 0.0, false, params.lumaCurve.curve, hist16, curve, NULL);
ipf.luminanceCurve (labView, labView, curve, 0, fh); ipf.luminanceCurve (labView, labView, curve, 0, fh);
ipf.lumadenoise (labView, &params, 1, buffer); ipf.lumadenoise (labView, buffer);
ipf.sharpening (labView, &params, 1, (unsigned short**)buffer); ipf.sharpening (labView, (unsigned short**)buffer);
delete [] curve; delete [] curve;
delete [] hist16; delete [] hist16;
// color processing // color processing
ipf.colorCurve (labView, labView, &params); ipf.colorCurve (labView, labView);
ipf.colordenoise (labView, &params, 1, buffer); ipf.colordenoise (labView, buffer);
for (int i=0; i<fh; i++) for (int i=0; i<fh; i++)
delete [] buffer[i]; delete [] buffer[i];
@ -182,7 +182,6 @@ IImage16* processImage (ProcessingJob* pjob, int& errorCode, ProgressListener* p
ch = params.crop.h; ch = params.crop.h;
} }
readyImg = ipf.lab2rgb16 (labView, cx, cy, cw, ch, params.icm.output); readyImg = ipf.lab2rgb16 (labView, cx, cy, cw, ch, params.icm.output);
ipf.release ();
if (pl) if (pl)
pl->setProgress (1.0); pl->setProgress (1.0);