Changes to black compression and saturation controls. Black compression from 0-50 acts the same as 0-100 on the previous version, compressing dark tones without crushing blacks. 50-100 then starts crushing blacks until by 100 on the slider, all tones up to the set black point are sent to zero. In the new saturation control, negative values of the slider set a linear curve rather than an inverted S curve, and smoothly decrease saturation to zero across the board.

This commit is contained in:
Emil Martinec
2010-10-26 22:59:18 -05:00
commit 926056c2c2
620 changed files with 130476 additions and 0 deletions

312
rtengine/wavelet.h Normal file
View File

@@ -0,0 +1,312 @@
/*
* This file is part of RawTherapee.
*
* 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/>.
*
* 2010 Ilya Popov <ilia_popov@rambler.ru>
*/
#ifndef WAVELET_H_INCLUDED
#define WAVELET_H_INCLUDED
/////////////////////////////////////////////////////////////////////////////
// Haar wavelets //
/////////////////////////////////////////////////////////////////////////////
// Bad, strong block effect
template<typename T>
void dwt_haar(T * data, size_t pitch, T * buffer, size_t n)
{
size_t n2a = (n + 1) / 2;
size_t n2 = n/2;
for(size_t i = 0, j = 0; i < n2; i++, j += 2 * pitch)
{
T a = data[j];
T b = data[j + pitch];
buffer[i] = (a + b) / 2;
buffer[n2a + i] = (a - b);
}
if(n2 < n2a)
{
buffer[n2] = data[pitch * (n-1)];
}
for(size_t k = 0, q = 0; k < n; k++, q += pitch)
{
data[q] = buffer[k];
}
}
template<typename T>
void idwt_haar(T * data, size_t pitch, T * buffer, size_t n, int alpha)
{
size_t n2a = (n + 1) / 2;
size_t n2 = n/2;
for(size_t i = 0, j = 0; i < n2; i++, j += 2)
{
T p = data[i * pitch];
T q = (alpha * data[(n2a + i)*pitch]) / 1024;
buffer[j] = p + q / 2;
buffer[j + 1] = p - q / 2;
}
if(n2 < n2a)
{
buffer[n-1] = data[pitch * n2];
}
for(size_t k = 0, q = 0; k < n; k++, q += pitch)
{
data[q] = buffer[k];
}
}
/////////////////////////////////////////////////////////////////////////////
// CDF 5/3 wavelets //
/////////////////////////////////////////////////////////////////////////////
// buffer must be of length (n + 4)
template<typename T>
void dwt_53(T * data, size_t pitch, T * buffer, size_t n)
{
size_t n2 = n/2;
size_t n2a = (n + 1) / 2;
T * tmp = buffer + 2;
// copy data
for(size_t i = 0, j = 0; i < n; i++, j += pitch)
{
tmp[i] = data[j];
}
// extend mirror-like
tmp[-1] = tmp[1];
tmp[-2] = tmp[2];
tmp[n] = tmp[n-2];
tmp[n+1] = tmp[n-3];
// calculate coefficients
for(ptrdiff_t i = -1; i < (ptrdiff_t)n + 1; i += 2)
{
tmp[i] = tmp[i] - (tmp[i-1] + tmp[i+1]) / 2;
}
for(ptrdiff_t i = 0; i < (ptrdiff_t)n; i += 2)
{
tmp[i] = tmp[i] + (tmp[i-1] + tmp[i+1] + 2) / 4;
}
// copy with reordering
for(size_t i = 0, j = 0; i < n; i+=2, j += pitch)
{
data[j] = tmp[i];
}
for(size_t i = 1, j = n2a*pitch; i < n; i+=2, j += pitch)
{
data[j] = tmp[i];
}
}
template<typename T>
void idwt_53(T * data, size_t pitch, T * buffer, size_t n, int alpha)
{
size_t n2 = n/2;
size_t n2a = (n + 1) / 2;
T * tmp = buffer + 2;
// copy with reordering
for(size_t i = 0, j = 0; i < n; i+=2, j += pitch)
{
tmp[i] = data[j];
}
for(size_t i = 1, j = n2a*pitch; i < n; i+=2, j += pitch)
{
tmp[i] = (alpha * data[j]) / 1024;
}
// extend mirror-like
tmp[-1] = tmp[1];
tmp[-2] = tmp[2];
tmp[n] = tmp[n-2];
tmp[n+1] = tmp[n-3];
// calculate coefficients
for(ptrdiff_t i = 0; i < (ptrdiff_t)n + 1; i += 2)
{
tmp[i] = tmp[i] - (tmp[i-1] + tmp[i+1] + 2) / 4;
}
for(ptrdiff_t i = 1; i < (ptrdiff_t)n; i += 2)
{
tmp[i] = tmp[i] + (tmp[i-1] + tmp[i+1]) / 2;
}
// copy data
for(size_t i = 0, j = 0; i < n; i++, j += pitch)
{
data[j] = tmp[i];
}
}
/////////////////////////////////////////////////////////////////////////////
// Edge-avoiding wavelets //
/////////////////////////////////////////////////////////////////////////////
// based on
// Edge-Avoiding Wavelets and their Applications
// Raanan Fattal <raananf@cs.huji.ac.il>,
// Hebrew University of Jerusalem, Israel
//
// WCDF variant from this paper is used here
// T must be one of floating-point types
/////////////////////////////////////////////////////////////////////////////
template<typename T>
inline T wcdf_weight(T a, T b)
{
static const T eps = 1;
static const T one = 1.0;
return one / (fabs(a - b) + eps);
}
// buffer is a temporary storage
// buffer2 must be preserved between dwt and idwt
template<typename T>
void dwt_wcdf(T * data, size_t pitch, T * buffer, size_t n, T * buffer2)
{
size_t n2 = n/2;
size_t n2a = (n + 1) / 2;
T * tmp = buffer + 2;
T * w = buffer2 + 2;
// copy data
for(size_t i = 0, j = 0; i < n; i++, j += pitch)
{
tmp[i] = data[j];
}
// extend mirror-like
tmp[-1] = tmp[1];
tmp[-2] = tmp[2];
tmp[n] = tmp[n-2];
tmp[n+1] = tmp[n-3];
// calculate weights
for(ptrdiff_t i = 0; i < (ptrdiff_t)n - 1; i++)
{
w[i] = wcdf_weight(tmp[i], tmp[i+1]);
//w[i] = 1;
}
w[-1] = w[-2] = (T)0.0;
w[n-1] = w[n] = w[n + 1] = (T)0.0;
// calculate coefficients
for(ptrdiff_t i = 1; i < (ptrdiff_t)n; i += 2)
{
tmp[i] = tmp[i] - (w[i-1]*tmp[i-1] + w[i]*tmp[i+1]) / (w[i-1] + w[i]);
}
for(ptrdiff_t i = 0; i < (ptrdiff_t)n; i += 2)
{
tmp[i] = tmp[i] + (T)0.5 * (w[i-1]*tmp[i-1] + w[i]*tmp[i+1]) / (w[i-1] + w[i]);
}
// copy with reordering
for(size_t i = 0, j = 0; i < n; i+=2, j += pitch)
{
data[j] = tmp[i];
}
for(size_t i = 1, j = n2a*pitch; i < n; i+=2, j += pitch)
{
data[j] = tmp[i];
}
}
template<typename T>
void idwt_wcdf(T * data, size_t pitch, T * buffer, size_t n, int alpha, T * buffer2)
{
size_t n2 = n/2;
size_t n2a = (n + 1) / 2;
T * tmp = buffer + 2;
T * w = buffer2 + 2;
// copy with reordering
for(size_t i = 0, j = 0; i < n; i+=2, j += pitch)
{
tmp[i] = data[j];
}
for(size_t i = 1, j = n2a*pitch; i < n; i+=2, j += pitch)
{
tmp[i] = (alpha * data[j]) / 1024;
}
// extend mirror-like
tmp[-1] = tmp[1];
tmp[-2] = tmp[2];
tmp[n] = tmp[n-2];
tmp[n+1] = tmp[n-3];
// calculate coefficients
for(ptrdiff_t i = 0; i < (ptrdiff_t)n; i += 2)
{
tmp[i] = tmp[i] - (T)0.5 * (w[i-1]*tmp[i-1] + w[i]*tmp[i+1]) / (w[i-1] + w[i]);
}
for(ptrdiff_t i = 1; i < (ptrdiff_t)n; i += 2)
{
tmp[i] = tmp[i] + (w[i-1]*tmp[i-1] + w[i]*tmp[i+1]) / (w[i-1] + w[i]);
}
// copy data
for(size_t i = 0, j = 0; i < n; i++, j += pitch)
{
data[j] = tmp[i];
}
}
//////////////////////////////////////////////////////////////////////////////
#endif