October 2011 - Gfxprim

[repo.or.cz] gfxprim.git branch generate updated: 631e1ca332e1b3bfccd1ebc4d83fa407eba13d51
by metan 01 Oct '11

01 Oct '11

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project gfxprim.git. The branch, generate has been updated via 631e1ca332e1b3bfccd1ebc4d83fa407eba13d51 (commit) from 4c68110cd82c373fe1d0c19ca01362edb452a0ad (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- http://repo.or.cz/w/gfxprim.git/commit/631e1ca332e1b3bfccd1ebc4d83fa407eba1… commit 631e1ca332e1b3bfccd1ebc4d83fa407eba13d51 Author: Cyril Hrubis <metan(a)ucw.cz> Date: Sat Oct 1 22:11:44 2011 +0200 Cleaned up the point pixel template. diff --git a/libs/filters/GP_Brightness.gen.c.t b/libs/filters/GP_Brightness.gen.c.t index 8262dc1..1ba6fb9 100644 --- a/libs/filters/GP_Brightness.gen.c.t +++ b/libs/filters/GP_Brightness.gen.c.t @@ -5,9 +5,8 @@ Brightness filters -- Increments all color channels by a fixed value. %% endblock %% block body -#include <GP_Context.h> -#include <GP_Pixel.h> -#include <GP_GetPutPixel.h> + +{{ filter_include() }} %% call(ps) filter_per_pixel_size('Brightness', 'int32_t inc') pix = pix + inc; @@ -19,6 +18,6 @@ pix = pix + inc; {{ filter_clamp_val(chan[0], chan[2]) }} %% endcall -{{ filter_functions('Brightness', 'int32_t inc', 'inc') }} +{{ filter_functions('Brightness', 'int32_t inc', 'inc', "inc=%i") }} %% endblock body diff --git a/libs/filters/GP_Contrast.gen.c.t b/libs/filters/GP_Contrast.gen.c.t index 5a78363..3398d1a 100644 --- a/libs/filters/GP_Contrast.gen.c.t +++ b/libs/filters/GP_Contrast.gen.c.t @@ -5,9 +5,8 @@ Contrast filters -- Multiply all color channels by a fixed value. %% endblock %% block body -#include <GP_Context.h> -#include <GP_Pixel.h> -#include <GP_GetPutPixel.h> + +{{ filter_include() }} %% call(ps) filter_per_pixel_size('Contrast', 'float mul') pix = 1.00 * pix * mul; @@ -19,6 +18,6 @@ pix = 1.00 * pix * mul; {{ filter_clamp_val(chan[0], chan[2]) }} %% endcall -{{ filter_functions('Contrast', 'float mul', 'mul') }} +{{ filter_functions('Contrast', 'float mul', 'mul', "mul=%2.3f") }} %% endblock body diff --git a/libs/filters/GP_Invert.gen.c.t b/libs/filters/GP_Invert.gen.c.t index f641fcf..1c11b30 100644 --- a/libs/filters/GP_Invert.gen.c.t +++ b/libs/filters/GP_Invert.gen.c.t @@ -5,9 +5,8 @@ Invert filters -- Invert image %% endblock %% block body -#include <GP_Context.h> -#include <GP_Pixel.h> -#include <GP_GetPutPixel.h> + +{{ filter_include() }} %% call(ps) filter_per_pixel_size('Invert') pix = {{ 2 ** ps.size - 1 }} - pix; diff --git a/pylib/templates/filter.point.c.t b/pylib/templates/filter.point.c.t index 93461d9..271418c 100644 --- a/pylib/templates/filter.point.c.t +++ b/pylib/templates/filter.point.c.t @@ -1,9 +1,12 @@ %% extends "base.c.t" -%% macro maybe_opts(opts) -%% if opts -,{{ opts }} -%% endif +{% macro maybe_opts(opts) %}{% if opts %}, {{ opts }}{% endif %}{% endmacro %} + +%% macro filter_include() +#include <GP_Context.h> +#include <GP_Pixel.h> +#include <GP_GetPutPixel.h> +#include <GP_Debug.h> %% endmacro /* @@ -54,6 +57,7 @@ void GP_Filter{{ name }}_{{ pt.name }}(const GP_Context *src, GP_Context *res{{ GP_PutPixel_Raw_{{ pt.pixelsize.suffix }}(res, x, y, pix); } } + %% endif %% endfor %% endmacro @@ -72,9 +76,11 @@ void GP_Filter{{ name }}_{{ pt.name }}(const GP_Context *src, GP_Context *res{{ /* * Switch per pixel sizes or pixel types. */ -%% macro filter_functions(name, opts="", params="") +%% macro filter_functions(name, opts="", params="", fmt="") void GP_Filter{{ name }}_Raw(const GP_Context *src, GP_Context *res{{ maybe_opts(opts) }}) { + GP_DEBUG(1, "Running filter {{ name }} {{ fmt }}"{{ maybe_opts(params) }}); + switch (src->pixel_type) { %% for pt in pixeltypes case GP_PIXEL_{{ pt.name }}: ----------------------------------------------------------------------- Summary of changes: libs/filters/GP_Brightness.gen.c.t | 7 +++---- libs/filters/GP_Contrast.gen.c.t | 7 +++---- libs/filters/GP_Invert.gen.c.t | 5 ++--- pylib/templates/filter.point.c.t | 16 +++++++++++----- 4 files changed, 19 insertions(+), 16 deletions(-) repo.or.cz automatic notification. Contact project admin jiri.bluebear.dluhos(a)gmail.com if you want to unsubscribe, or site admin admin(a)repo.or.cz if you receive no reply. -- gfxprim.git ("A simple 2D graphics library with emphasis on correctness and well-defined operation.")

1 0

[repo.or.cz] gfxprim.git branch generate updated: 4c68110cd82c373fe1d0c19ca01362edb452a0ad
by metan 01 Oct '11

01 Oct '11

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project gfxprim.git. The branch, generate has been updated via 4c68110cd82c373fe1d0c19ca01362edb452a0ad (commit) from 96956ac4fbd2c35d57c122ecd44a88c352153a5e (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- http://repo.or.cz/w/gfxprim.git/commit/4c68110cd82c373fe1d0c19ca01362edb452… commit 4c68110cd82c373fe1d0c19ca01362edb452a0ad Author: Cyril Hrubis <metan(a)ucw.cz> Date: Sat Oct 1 15:48:36 2011 +0200 Fix linear convolution corner cases at the corners. diff --git a/libs/filters/GP_Linear.c b/libs/filters/GP_Linear.c index bf41660..764d6b6 100644 --- a/libs/filters/GP_Linear.c +++ b/libs/filters/GP_Linear.c @@ -101,7 +101,7 @@ static void GP_FilterLinearConvolution(const GP_Context *src, GP_Context *dst, if (cy >= (int)src->h) cy = src->h - 1; - + pix = GP_GetPixel_Raw_24BPP(src, cx, cy); R[i][j] = GP_Pixel_GET_R_RGB888(pix); @@ -117,14 +117,18 @@ static void GP_FilterLinearConvolution(const GP_Context *src, GP_Context *dst, for (j = 0; j < kh; j++) { int cy = y + j - kh/2; - + int cx = x + kw/2; + if (cy < 0) cy = 0; - + if (cy >= (int)src->h) cy = src->h - 1; - pix = GP_GetPixel_Raw_24BPP(src, x, cy); + if (cx >= (int)src->w) + cx = src->w - 1; + + pix = GP_GetPixel_Raw_24BPP(src, cx, cy); R[idx][j] = GP_Pixel_GET_R_RGB888(pix); G[idx][j] = GP_Pixel_GET_G_RGB888(pix); @@ -133,10 +137,12 @@ static void GP_FilterLinearConvolution(const GP_Context *src, GP_Context *dst, /* count the pixel value from neighbours weighted by kernel */ for (i = 0; i < kw; i++) { - int k = i - idx; - - if (k < 0) - k += kw; + int k; + + if ((int)i < idx + 1) + k = kw - idx - 1 + i; + else + k = i - idx - 1; for (j = 0; j < kh; j++) { r += R[i][j] * kernel[k + j * kw]; ----------------------------------------------------------------------- Summary of changes: libs/filters/GP_Linear.c | 22 ++++++++++++++-------- 1 files changed, 14 insertions(+), 8 deletions(-) repo.or.cz automatic notification. Contact project admin jiri.bluebear.dluhos(a)gmail.com if you want to unsubscribe, or site admin admin(a)repo.or.cz if you receive no reply. -- gfxprim.git ("A simple 2D graphics library with emphasis on correctness and well-defined operation.")

1 0

[repo.or.cz] gfxprim.git branch generate updated: 96956ac4fbd2c35d57c122ecd44a88c352153a5e
by metan 01 Oct '11

01 Oct '11

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project gfxprim.git. The branch, generate has been updated via 96956ac4fbd2c35d57c122ecd44a88c352153a5e (commit) from 8de6ee02d1d25d761c590a1bdaeb142fd04191e3 (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- http://repo.or.cz/w/gfxprim.git/commit/96956ac4fbd2c35d57c122ecd44a88c35215… commit 96956ac4fbd2c35d57c122ecd44a88c352153a5e Author: Cyril Hrubis <metan(a)ucw.cz> Date: Sat Oct 1 13:58:00 2011 +0200 Implemented linear convolution and gaussian blur. diff --git a/demos/grinder/grinder.c b/demos/grinder/grinder.c index 803570d..82e7228 100644 --- a/demos/grinder/grinder.c +++ b/demos/grinder/grinder.c @@ -292,17 +292,25 @@ static GP_RetCode invert(GP_Context **c, const char *params) } static struct param blur_params[] = { + {"sigma", PARAM_FLOAT, "sigma parameter (eg. radii of blur)", NULL, NULL}, {NULL, 0, NULL, NULL, NULL} }; static GP_RetCode blur(GP_Context **c, const char *params) { - if (param_parse(params, blur_params, "blur", param_err)) + float sigma = 0; + + if (param_parse(params, blur_params, "blur", param_err, &sigma)) return GP_EINVAL; + if (sigma <= 0) { + print_error("blur: sigma parameter must be >= 0"); + return GP_EINVAL; + } + GP_Context *res = NULL; - res = GP_FilterBlur(*c); + res = GP_FilterGaussianBlur(*c, sigma); if (res == NULL) return GP_EINVAL; diff --git a/include/filters/GP_Linear.h b/include/filters/GP_Linear.h index 2ef5a0e..98e53f0 100644 --- a/include/filters/GP_Linear.h +++ b/include/filters/GP_Linear.h @@ -31,6 +31,6 @@ #include <GP_Context.h> -GP_Context *GP_FilterBlur(GP_Context *src); +GP_Context *GP_FilterGaussianBlur(GP_Context *src, float sigma); #endif /* GP_LINEAR_H */ diff --git a/libs/filters/GP_Linear.c b/libs/filters/GP_Linear.c index 66c0db6..bf41660 100644 --- a/libs/filters/GP_Linear.c +++ b/libs/filters/GP_Linear.c @@ -20,6 +20,8 @@ * * *****************************************************************************/ +#include <math.h> + #include <GP_Context.h> #include <GP_GetPutPixel.h> @@ -27,13 +29,13 @@ #include <GP_Linear.h> -static float HX[] = {0.17, 0.66, 0.17}; -static float HY[] = {0.17, 0.66, 0.17}; +static void GP_FilterLinearConvolution(const GP_Context *src, GP_Context *dst, + float kernel[], uint32_t kw, uint32_t kh); -GP_Context *GP_FilterBlur(GP_Context *src) +GP_Context *GP_FilterGaussianBlur(GP_Context *src, float sigma) { GP_Context *dst; - + if (src->pixel_type != GP_PIXEL_RGB888) return NULL; @@ -42,89 +44,113 @@ GP_Context *GP_FilterBlur(GP_Context *src) if (dst == NULL) return NULL; - GP_DEBUG(1, "Linear blur filter (3x3 kernel) %ux%u", src->w, src->h); - GP_Coord x, y; + /* compute kernel */ + int i, center = 3 * sigma; + int N = 2 * center + 1; + float kernel[N]; - for (y = 0; y < (GP_Coord)dst->h; y++) { - GP_Pixel pix; - uint32_t R1, R2, R3; - uint32_t G1, G2, G3; - uint32_t B1, B2, B3; - - pix = GP_GetPixel_Raw_24BPP(src, 0, y); - - R1 = GP_Pixel_GET_R_RGB888(pix); - G1 = GP_Pixel_GET_G_RGB888(pix); - B1 = GP_Pixel_GET_B_RGB888(pix); - - R2 = R1; - G2 = G1; - B2 = B1; + double sigma2 = sigma * sigma; - for (x = 0; x < (GP_Coord)dst->w; x++) { - float r, g, b; + for (i = 0; i < N; i++) { + double r = center - i; + + kernel[i] = exp(-0.5 * (r * r) / sigma2); + } - pix = GP_GetPixel_Raw_24BPP(src, x, y); - - R3 = GP_Pixel_GET_R_RGB888(pix); - G3 = GP_Pixel_GET_G_RGB888(pix); - B3 = GP_Pixel_GET_B_RGB888(pix); + GP_DEBUG(1, "Gaussian blur sigma=%2.3f kernel %ix%i image %ux%u", + sigma, N, N, src->w, src->h); - r = R1 * HX[0] + R2 * HX[1] + R3 * HX[2]; - g = G1 * HX[0] + G2 * HX[1] + G3 * HX[2]; - b = B1 * HX[0] + B2 * HX[1] + B3 * HX[2]; + /* apply in both directions */ + GP_FilterLinearConvolution(src, dst, kernel, N, 1); + GP_FilterLinearConvolution(dst, dst, kernel, 1, N); - if (r > 255) - r = 255; - if (g > 255) - g = 255; - if (b > 255) - b = 255; + return dst; +} - pix = GP_Pixel_CREATE_RGB888((uint32_t)r, (uint32_t)g, (uint32_t)b); +static void GP_FilterLinearConvolution(const GP_Context *src, GP_Context *dst, + float kernel[], uint32_t kw, uint32_t kh) +{ + float kernel_sum = 0; + GP_Coord x, y; + uint32_t i, j; - GP_PutPixel_Raw_24BPP(dst, x, y, pix); - - R1 = R2; - G1 = G2; - B1 = B2; - - R2 = R3; - G2 = G3; - B2 = B3; - } - } - - for (x = 0; x < (GP_Coord)dst->w; x++) { - GP_Pixel pix; - uint32_t R1, R2, R3; - uint32_t G1, G2, G3; - uint32_t B1, B2, B3; - - pix = GP_GetPixel_Raw_24BPP(src, x, 0); + GP_DEBUG(1, "Linear convolution kernel %ix%i image %ux%u", + kw, kh, src->w, src->h); - R1 = GP_Pixel_GET_R_RGB888(pix); - G1 = GP_Pixel_GET_G_RGB888(pix); - B1 = GP_Pixel_GET_B_RGB888(pix); - - R2 = R1; - G2 = G1; - B2 = B1; + /* count kernel sum for normalization */ + for (i = 0; i < kw * kh; i++) + kernel_sum += kernel[i]; - for (y = 0; y < (GP_Coord)dst->h; y++) { - float r, g, b; + /* do linear convolution */ + for (y = 0; y < (GP_Coord)dst->h; y++) { + GP_Pixel pix; + uint32_t R[kw][kh], G[kw][kh], B[kw][kh]; + + /* prefill the buffer on the start */ + for (j = 0; j < kh; j++) { + for (i = 0; i < kw - 1; i++) { + int cx = i - kw/2; + int cy = y + j - kh/2; + + if (cx < 0) + cx = 0; + + if (cy < 0) + cy = 0; + + if (cy >= (int)src->h) + cy = src->h - 1; + + pix = GP_GetPixel_Raw_24BPP(src, cx, cy); + + R[i][j] = GP_Pixel_GET_R_RGB888(pix); + G[i][j] = GP_Pixel_GET_G_RGB888(pix); + B[i][j] = GP_Pixel_GET_B_RGB888(pix); + } + } - pix = GP_GetPixel_Raw_24BPP(src, x, y); - - R3 = GP_Pixel_GET_R_RGB888(pix); - G3 = GP_Pixel_GET_G_RGB888(pix); - B3 = GP_Pixel_GET_B_RGB888(pix); + int idx = kw - 1; - r = R1 * HY[0] + R2 * HY[1] + R3 * HY[2]; - g = G1 * HY[0] + G2 * HY[1] + G3 * HY[2]; - b = B1 * HY[0] + B2 * HY[1] + B3 * HY[2]; + for (x = 0; x < (GP_Coord)dst->w; x++) { + float r = 0, g = 0, b = 0; + + for (j = 0; j < kh; j++) { + int cy = y + j - kh/2; + + if (cy < 0) + cy = 0; + + if (cy >= (int)src->h) + cy = src->h - 1; + + pix = GP_GetPixel_Raw_24BPP(src, x, cy); + + R[idx][j] = GP_Pixel_GET_R_RGB888(pix); + G[idx][j] = GP_Pixel_GET_G_RGB888(pix); + B[idx][j] = GP_Pixel_GET_B_RGB888(pix); + } + /* count the pixel value from neighbours weighted by kernel */ + for (i = 0; i < kw; i++) { + int k = i - idx; + + if (k < 0) + k += kw; + + for (j = 0; j < kh; j++) { + r += R[i][j] * kernel[k + j * kw]; + g += G[i][j] * kernel[k + j * kw]; + b += B[i][j] * kernel[k + j * kw]; + } + } + + /* normalize the result */ + r /= kernel_sum; + g /= kernel_sum; + b /= kernel_sum; + + /* and clamp just to be extra sure */ if (r > 255) r = 255; if (g > 255) @@ -136,15 +162,11 @@ GP_Context *GP_FilterBlur(GP_Context *src) GP_PutPixel_Raw_24BPP(dst, x, y, pix); - R1 = R2; - G1 = G2; - B1 = B2; - - R2 = R3; - G2 = G3; - B2 = B3; + idx++; + + if (idx >= (int)kw) + idx = 0; } } - - return dst; } + ----------------------------------------------------------------------- Summary of changes: demos/grinder/grinder.c | 12 +++- include/filters/GP_Linear.h | 2 +- libs/filters/GP_Linear.c | 188 ++++++++++++++++++++++++------------------- 3 files changed, 116 insertions(+), 86 deletions(-) repo.or.cz automatic notification. Contact project admin jiri.bluebear.dluhos(a)gmail.com if you want to unsubscribe, or site admin admin(a)repo.or.cz if you receive no reply. -- gfxprim.git ("A simple 2D graphics library with emphasis on correctness and well-defined operation.")

1 0

[repo.or.cz] gfxprim.git branch generate updated: 8de6ee02d1d25d761c590a1bdaeb142fd04191e3
by metan 01 Oct '11

01 Oct '11

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project gfxprim.git. The branch, generate has been updated via 8de6ee02d1d25d761c590a1bdaeb142fd04191e3 (commit) from 7765fb749d234f875480769baba4b8c3c4919a0d (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- http://repo.or.cz/w/gfxprim.git/commit/8de6ee02d1d25d761c590a1bdaeb142fd041… commit 8de6ee02d1d25d761c590a1bdaeb142fd04191e3 Author: Cyril Hrubis <metan(a)ucw.cz> Date: Sat Oct 1 02:06:24 2011 +0200 First simple linear blur. diff --git a/demos/grinder/grinder.c b/demos/grinder/grinder.c index 78d33d7..803570d 100644 --- a/demos/grinder/grinder.c +++ b/demos/grinder/grinder.c @@ -291,6 +291,28 @@ static GP_RetCode invert(GP_Context **c, const char *params) return GP_ESUCCESS; } +static struct param blur_params[] = { + {NULL, 0, NULL, NULL, NULL} +}; + +static GP_RetCode blur(GP_Context **c, const char *params) +{ + if (param_parse(params, blur_params, "blur", param_err)) + return GP_EINVAL; + + GP_Context *res = NULL; + + res = GP_FilterBlur(*c); + + if (res == NULL) + return GP_EINVAL; + + GP_ContextFree(*c); + *c = res; + + return GP_ESUCCESS; +} + /* filters */ struct filter { @@ -307,6 +329,7 @@ static struct filter filter_table[] = { {"bright", "alter image brightness", bright_params, bright}, {"contrast", "alter image contrast", contrast_params, contrast}, {"invert", "inverts image", invert_params, invert}, + {"blur", "gaussian blur", blur_params, blur}, {NULL, NULL, NULL, NULL} }; diff --git a/include/filters/GP_Filters.h b/include/filters/GP_Filters.h index 4c1304d..fc498b0 100644 --- a/include/filters/GP_Filters.h +++ b/include/filters/GP_Filters.h @@ -36,6 +36,7 @@ #include "filters/GP_Rotate.h" #include "filters/GP_Point.h" +#include "filters/GP_Linear.h" /* Image down and up scaling */ #include "filters/GP_Scale.h" diff --git a/include/filters/GP_Filters.h b/include/filters/GP_Linear.h similarity index 76% copy from include/filters/GP_Filters.h copy to include/filters/GP_Linear.h index 4c1304d..2ef5a0e 100644 --- a/include/filters/GP_Filters.h +++ b/include/filters/GP_Linear.h @@ -16,28 +16,21 @@ * Foundation, Inc., 51 Franklin Street, Fifth Floor, * * Boston, MA 02110-1301 USA * * * - * Copyright (C) 2009-2010 Jiri "BlueBear" Dluhos * - * <jiri.bluebear.dluhos(a)gmail.com> * - * * * Copyright (C) 2009-2011 Cyril Hrubis <metan(a)ucw.cz> * * * *****************************************************************************/ /* - GP_Context filters. + Linear filters. */ -#ifndef GP_FILTERS_H -#define GP_FILTERS_H - -/* Image rotations (90 180 270 grads) and mirroring */ -#include "filters/GP_Rotate.h" +#ifndef GP_LINEAR_H +#define GP_LINEAR_H -#include "filters/GP_Point.h" +#include <GP_Context.h> -/* Image down and up scaling */ -#include "filters/GP_Scale.h" +GP_Context *GP_FilterBlur(GP_Context *src); -#endif /* GP_FILTERS_H */ +#endif /* GP_LINEAR_H */ diff --git a/libs/filters/GP_Linear.c b/libs/filters/GP_Linear.c new file mode 100644 index 0000000..66c0db6 --- /dev/null +++ b/libs/filters/GP_Linear.c @@ -0,0 +1,150 @@ +/***************************************************************************** + * This file is part of gfxprim library. * + * * + * Gfxprim is free software; you can redistribute it and/or * + * modify it under the terms of the GNU Lesser General Public * + * License as published by the Free Software Foundation; either * + * version 2.1 of the License, or (at your option) any later version. * + * * + * Gfxprim 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 * + * Lesser General Public License for more details. * + * * + * You should have received a copy of the GNU Lesser General Public * + * License along with gfxprim; if not, write to the Free Software * + * Foundation, Inc., 51 Franklin Street, Fifth Floor, * + * Boston, MA 02110-1301 USA * + * * + * Copyright (C) 2009-2011 Cyril Hrubis <metan(a)ucw.cz> * + * * + *****************************************************************************/ + +#include <GP_Context.h> +#include <GP_GetPutPixel.h> + +#include <GP_Debug.h> + +#include <GP_Linear.h> + +static float HX[] = {0.17, 0.66, 0.17}; +static float HY[] = {0.17, 0.66, 0.17}; + +GP_Context *GP_FilterBlur(GP_Context *src) +{ + GP_Context *dst; + + if (src->pixel_type != GP_PIXEL_RGB888) + return NULL; + + dst = GP_ContextCopy(src, 0); + + if (dst == NULL) + return NULL; + + GP_DEBUG(1, "Linear blur filter (3x3 kernel) %ux%u", src->w, src->h); + + GP_Coord x, y; + + for (y = 0; y < (GP_Coord)dst->h; y++) { + GP_Pixel pix; + uint32_t R1, R2, R3; + uint32_t G1, G2, G3; + uint32_t B1, B2, B3; + + pix = GP_GetPixel_Raw_24BPP(src, 0, y); + + R1 = GP_Pixel_GET_R_RGB888(pix); + G1 = GP_Pixel_GET_G_RGB888(pix); + B1 = GP_Pixel_GET_B_RGB888(pix); + + R2 = R1; + G2 = G1; + B2 = B1; + + for (x = 0; x < (GP_Coord)dst->w; x++) { + float r, g, b; + + pix = GP_GetPixel_Raw_24BPP(src, x, y); + + R3 = GP_Pixel_GET_R_RGB888(pix); + G3 = GP_Pixel_GET_G_RGB888(pix); + B3 = GP_Pixel_GET_B_RGB888(pix); + + r = R1 * HX[0] + R2 * HX[1] + R3 * HX[2]; + g = G1 * HX[0] + G2 * HX[1] + G3 * HX[2]; + b = B1 * HX[0] + B2 * HX[1] + B3 * HX[2]; + + if (r > 255) + r = 255; + if (g > 255) + g = 255; + if (b > 255) + b = 255; + + pix = GP_Pixel_CREATE_RGB888((uint32_t)r, (uint32_t)g, (uint32_t)b); + + GP_PutPixel_Raw_24BPP(dst, x, y, pix); + + R1 = R2; + G1 = G2; + B1 = B2; + + R2 = R3; + G2 = G3; + B2 = B3; + } + } + + for (x = 0; x < (GP_Coord)dst->w; x++) { + GP_Pixel pix; + uint32_t R1, R2, R3; + uint32_t G1, G2, G3; + uint32_t B1, B2, B3; + + pix = GP_GetPixel_Raw_24BPP(src, x, 0); + + R1 = GP_Pixel_GET_R_RGB888(pix); + G1 = GP_Pixel_GET_G_RGB888(pix); + B1 = GP_Pixel_GET_B_RGB888(pix); + + R2 = R1; + G2 = G1; + B2 = B1; + + for (y = 0; y < (GP_Coord)dst->h; y++) { + float r, g, b; + + pix = GP_GetPixel_Raw_24BPP(src, x, y); + + R3 = GP_Pixel_GET_R_RGB888(pix); + G3 = GP_Pixel_GET_G_RGB888(pix); + B3 = GP_Pixel_GET_B_RGB888(pix); + + r = R1 * HY[0] + R2 * HY[1] + R3 * HY[2]; + g = G1 * HY[0] + G2 * HY[1] + G3 * HY[2]; + b = B1 * HY[0] + B2 * HY[1] + B3 * HY[2]; + + if (r > 255) + r = 255; + if (g > 255) + g = 255; + if (b > 255) + b = 255; + + pix = GP_Pixel_CREATE_RGB888((uint32_t)r, (uint32_t)g, (uint32_t)b); + + GP_PutPixel_Raw_24BPP(dst, x, y, pix); + + R1 = R2; + G1 = G2; + B1 = B2; + + R2 = R3; + G2 = G3; + B2 = B3; + } + } + + return dst; +} ----------------------------------------------------------------------- Summary of changes: demos/grinder/grinder.c | 23 +++ include/filters/GP_Filters.h | 1 + .../core/GP_Debug.c => include/filters/GP_Linear.h | 21 ++-- libs/filters/GP_Linear.c | 150 ++++++++++++++++++++ 4 files changed, 185 insertions(+), 10 deletions(-) copy libs/core/GP_Debug.c => include/filters/GP_Linear.h (88%) create mode 100644 libs/filters/GP_Linear.c repo.or.cz automatic notification. Contact project admin jiri.bluebear.dluhos(a)gmail.com if you want to unsubscribe, or site admin admin(a)repo.or.cz if you receive no reply. -- gfxprim.git ("A simple 2D graphics library with emphasis on correctness and well-defined operation.")

1 0