falkTX
/
FFmpeg
mirror of https://github.com/falkTX/FFmpeg.git
1
0
Fork 0
Code Issues 0 Releases 338 Wiki Activity
Browse Source

lavfi: port boxblur filter from libmpcodecs 

With the following additions:
* support to gray format
* support to yuva420p format
* parametric luma/chroma/alpha radius
* consistency check on the radius values, avoid crashes with invalid values

Signed-off-by: Anton Khirnov <anton@khirnov.net>
tags/n0.9
Stefano Sabatini Anton Khirnov 14 years ago
parent
171868e25a
commit
ce6b6ef645
7 changed files with 415 additions and 1 deletions
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. +1
    -0
      Changelog
  2. +1
    -0
      configure
  3. +60
    -0
      doc/filters.texi
  4. +1
    -0
      libavfilter/Makefile
  5. +1
    -0
      libavfilter/allfilters.c
  6. +1
    -1
      libavfilter/avfilter.h
  7. +350
    -0
      libavfilter/vf_boxblur.c

+ 1
- 0
Changelog View File

@@ -51,6 +51,7 @@ easier to use. The changes are:
- CELT in Ogg demuxing
- VC-1 interlaced decoding
- lut, lutrgb, and lutyuv filters
- boxblur filter


version 0.7:


+ 1
- 0
configure View File

@@ -1480,6 +1480,7 @@ udp_protocol_deps="network"

# filters
blackframe_filter_deps="gpl"
boxblur_filter_deps="gpl"
cropdetect_filter_deps="gpl"
drawtext_filter_deps="libfreetype"
frei0r_filter_deps="frei0r dlopen strtok_r"


+ 60
- 0
doc/filters.texi View File

@@ -183,6 +183,66 @@ threshold, and defaults to 98.
@var{threshold} is the threshold below which a pixel value is
considered black, and defaults to 32.

@section boxblur

Apply boxblur algorithm to the input video.

This filter accepts the parameters:
@var{luma_power}:@var{luma_radius}:@var{chroma_radius}:@var{chroma_power}:@var{alpha_radius}:@var{alpha_power}

Chroma and alpha parameters are optional, if not specified they default
to the corresponding values set for @var{luma_radius} and
@var{luma_power}.

@var{luma_radius}, @var{chroma_radius}, and @var{alpha_radius} represent
the radius in pixels of the box used for blurring the corresponding
input plane. They are expressions, and can contain the following
constants:
@table @option
@item w, h
the input width and heigth in pixels

@item cw, ch
the input chroma image width and height in pixels

@item hsub, vsub
horizontal and vertical chroma subsample values. For example for the
pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
@end table

The radius must be a non-negative number, and must not be greater than
the value of the expression @code{min(w,h)/2} for the luma and alpha planes,
and of @code{min(cw,ch)/2} for the chroma planes.

@var{luma_power}, @var{chroma_power}, and @var{alpha_power} represent
how many times the boxblur filter is applied to the corresponding
plane.

Some examples follow:

@itemize

@item
Apply a boxblur filter with luma, chroma, and alpha radius
set to 2:
@example
boxblur=2:1
@end example

@item
Set luma radius to 2, alpha and chroma radius to 0
@example
boxblur=2:1:0:0:0:0
@end example

@item
Set luma and chroma radius to a fraction of the video dimension
@example
boxblur=min(h\,w)/10:1:min(cw\,ch)/10:1
@end example

@end itemize

@section copy

Copy the input source unchanged to the output. Mainly useful for


+ 1
- 0
libavfilter/Makefile View File

@@ -20,6 +20,7 @@ OBJS-$(CONFIG_ANULLSRC_FILTER) += asrc_anullsrc.o
OBJS-$(CONFIG_ANULLSINK_FILTER) += asink_anullsink.o

OBJS-$(CONFIG_BLACKFRAME_FILTER) += vf_blackframe.o
OBJS-$(CONFIG_BOXBLUR_FILTER) += vf_boxblur.o
OBJS-$(CONFIG_COPY_FILTER) += vf_copy.o
OBJS-$(CONFIG_CROP_FILTER) += vf_crop.o
OBJS-$(CONFIG_CROPDETECT_FILTER) += vf_cropdetect.o


+ 1
- 0
libavfilter/allfilters.c View File

@@ -41,6 +41,7 @@ void avfilter_register_all(void)
REGISTER_FILTER (ANULLSINK, anullsink, asink);

REGISTER_FILTER (BLACKFRAME, blackframe, vf);
REGISTER_FILTER (BOXBLUR, boxblur, vf);
REGISTER_FILTER (COPY, copy, vf);
REGISTER_FILTER (CROP, crop, vf);
REGISTER_FILTER (CROPDETECT, cropdetect, vf);


+ 1
- 1
libavfilter/avfilter.h View File

@@ -29,7 +29,7 @@
#include "libavutil/rational.h"

#define LIBAVFILTER_VERSION_MAJOR 2
#define LIBAVFILTER_VERSION_MINOR 9
#define LIBAVFILTER_VERSION_MINOR 10
#define LIBAVFILTER_VERSION_MICRO 0

#define LIBAVFILTER_VERSION_INT AV_VERSION_INT(LIBAVFILTER_VERSION_MAJOR, \


+ 350
- 0
libavfilter/vf_boxblur.c View File

@@ -0,0 +1,350 @@
/*
* Copyright (c) 2002 Michael Niedermayer <michaelni@gmx.at>
* Copyright (c) 2011 Stefano Sabatini
*
* This file is part of Libav.
*
* Libav 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 2 of the License, or
* (at your option) any later version.
*
* Libav 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 Libav; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*/

/**
* @file
* Apply a boxblur filter to the input video.
* Ported from MPlayer libmpcodecs/vf_boxblur.c.
*/

#include "libavutil/avstring.h"
#include "libavutil/eval.h"
#include "libavutil/pixdesc.h"
#include "avfilter.h"

static const char *var_names[] = {
"w",
"h",
"cw",
"ch",
"hsub",
"vsub",
NULL
};

enum var_name {
VAR_W,
VAR_H,
VAR_CW,
VAR_CH,
VAR_HSUB,
VAR_VSUB,
VARS_NB
};

typedef struct {
int radius;
int power;
} FilterParam;

typedef struct {
FilterParam luma_param;
FilterParam chroma_param;
FilterParam alpha_param;
char luma_radius_expr [256];
char chroma_radius_expr[256];
char alpha_radius_expr [256];

int hsub, vsub;
int radius[4];
int power[4];
uint8_t *temp[2]; ///< temporary buffer used in blur_power()
} BoxBlurContext;

#define Y 0
#define U 1
#define V 2
#define A 3

static av_cold int init(AVFilterContext *ctx, const char *args, void *opaque)
{
BoxBlurContext *boxblur = ctx->priv;
int e;

if (!args) {
av_log(ctx, AV_LOG_ERROR,
"Filter expects 2 or 4 or 6 arguments, none provided\n");
return AVERROR(EINVAL);
}

e = sscanf(args, "%255[^:]:%d:%255[^:]:%d:%255[^:]:%d",
boxblur->luma_radius_expr, &boxblur->luma_param .power,
boxblur->chroma_radius_expr, &boxblur->chroma_param.power,
boxblur->alpha_radius_expr, &boxblur->alpha_param .power);

if (e != 2 && e != 4 && e != 6) {
av_log(ctx, AV_LOG_ERROR,
"Filter expects 2 or 4 or 6 params, provided %d\n", e);
return AVERROR(EINVAL);
}

if (e < 4) {
boxblur->chroma_param.power = boxblur->luma_param.power;
av_strlcpy(boxblur->chroma_radius_expr, boxblur->luma_radius_expr,
sizeof(boxblur->chroma_radius_expr));
}
if (e < 6) {
boxblur->alpha_param.power = boxblur->luma_param.power;
av_strlcpy(boxblur->alpha_radius_expr, boxblur->luma_radius_expr,
sizeof(boxblur->alpha_radius_expr));
}

return 0;
}

static av_cold void uninit(AVFilterContext *ctx)
{
BoxBlurContext *boxblur = ctx->priv;

av_freep(&boxblur->temp[0]);
av_freep(&boxblur->temp[1]);
}

static int query_formats(AVFilterContext *ctx)
{
enum PixelFormat pix_fmts[] = {
PIX_FMT_YUV444P, PIX_FMT_YUV422P, PIX_FMT_YUV420P,
PIX_FMT_YUV411P, PIX_FMT_YUV410P, PIX_FMT_YUVA420P,
PIX_FMT_YUV440P, PIX_FMT_GRAY8,
PIX_FMT_YUVJ444P, PIX_FMT_YUVJ422P, PIX_FMT_YUVJ420P,
PIX_FMT_YUVJ440P,
PIX_FMT_NONE
};

avfilter_set_common_formats(ctx, avfilter_make_format_list(pix_fmts));
return 0;
}

static int config_input(AVFilterLink *inlink)
{
const AVPixFmtDescriptor *desc = &av_pix_fmt_descriptors[inlink->format];
AVFilterContext *ctx = inlink->dst;
BoxBlurContext *boxblur = ctx->priv;
int w = inlink->w, h = inlink->h;
int cw, ch;
double var_values[VARS_NB], res;
char *expr;
int ret;

av_freep(&boxblur->temp[0]);
av_freep(&boxblur->temp[1]);
if (!(boxblur->temp[0] = av_malloc(FFMAX(w, h))))
return AVERROR(ENOMEM);
if (!(boxblur->temp[1] = av_malloc(FFMAX(w, h)))) {
av_freep(&boxblur->temp[0]);
return AVERROR(ENOMEM);
}

boxblur->hsub = desc->log2_chroma_w;
boxblur->vsub = desc->log2_chroma_h;

var_values[VAR_W] = inlink->w;
var_values[VAR_H] = inlink->h;
var_values[VAR_CW] = cw = w>>boxblur->hsub;
var_values[VAR_CH] = ch = h>>boxblur->vsub;
var_values[VAR_HSUB] = 1<<boxblur->hsub;
var_values[VAR_VSUB] = 1<<boxblur->vsub;

#define EVAL_RADIUS_EXPR(comp) \
expr = boxblur->comp##_radius_expr; \
ret = av_expr_parse_and_eval(&res, expr, var_names, var_values, \
NULL, NULL, NULL, NULL, NULL, 0, ctx); \
boxblur->comp##_param.radius = res; \
if (ret < 0) { \
av_log(NULL, AV_LOG_ERROR, \
"Error when evaluating " #comp " radius expression '%s'\n", expr); \
return ret; \
}
EVAL_RADIUS_EXPR(luma);
EVAL_RADIUS_EXPR(chroma);
EVAL_RADIUS_EXPR(alpha);

av_log(ctx, AV_LOG_DEBUG,
"luma_radius:%d luma_power:%d "
"chroma_radius:%d chroma_power:%d "
"alpha_radius:%d alpha_power:%d "
"w:%d chroma_w:%d h:%d chroma_h:%d\n",
boxblur->luma_param .radius, boxblur->luma_param .power,
boxblur->chroma_param.radius, boxblur->chroma_param.power,
boxblur->alpha_param .radius, boxblur->alpha_param .power,
w, cw, h, ch);

#define CHECK_RADIUS_VAL(w_, h_, comp) \
if (boxblur->comp##_param.radius < 0 || \
2*boxblur->comp##_param.radius > FFMIN(w_, h_)) { \
av_log(ctx, AV_LOG_ERROR, \
"Invalid " #comp " radius value %d, must be >= 0 and <= %d\n", \
boxblur->comp##_param.radius, FFMIN(w_, h_)/2); \
return AVERROR(EINVAL); \
}
CHECK_RADIUS_VAL(w, h, luma);
CHECK_RADIUS_VAL(cw, ch, chroma);
CHECK_RADIUS_VAL(w, h, alpha);

boxblur->radius[Y] = boxblur->luma_param.radius;
boxblur->radius[U] = boxblur->radius[V] = boxblur->chroma_param.radius;
boxblur->radius[A] = boxblur->alpha_param.radius;

boxblur->power[Y] = boxblur->luma_param.power;
boxblur->power[U] = boxblur->power[V] = boxblur->chroma_param.power;
boxblur->power[A] = boxblur->alpha_param.power;

return 0;
}

static inline void blur(uint8_t *dst, int dst_step, const uint8_t *src, int src_step,
int len, int radius)
{
/* Naive boxblur would sum source pixels from x-radius .. x+radius
* for destination pixel x. That would be O(radius*width).
* If you now look at what source pixels represent 2 consecutive
* output pixels, then you see they are almost identical and only
* differ by 2 pixels, like:
* src0 111111111
* dst0 1
* src1 111111111
* dst1 1
* src0-src1 1 -1
* so when you know one output pixel you can find the next by just adding
* and subtracting 1 input pixel.
* The following code adopts this faster variant.
*/
const int length = radius*2 + 1;
const int inv = ((1<<16) + length/2)/length;
int x, sum = 0;

for (x = 0; x < radius; x++)
sum += src[x*src_step]<<1;
sum += src[radius*src_step];

for (x = 0; x <= radius; x++) {
sum += src[(radius+x)*src_step] - src[(radius-x)*src_step];
dst[x*dst_step] = (sum*inv + (1<<15))>>16;
}

for (; x < len-radius; x++) {
sum += src[(radius+x)*src_step] - src[(x-radius-1)*src_step];
dst[x*dst_step] = (sum*inv + (1<<15))>>16;
}

for (; x < len; x++) {
sum += src[(2*len-radius-x-1)*src_step] - src[(x-radius-1)*src_step];
dst[x*dst_step] = (sum*inv + (1<<15))>>16;
}
}

static inline void blur_power(uint8_t *dst, int dst_step, const uint8_t *src, int src_step,
int len, int radius, int power, uint8_t *temp[2])
{
uint8_t *a = temp[0], *b = temp[1];

if (radius && power) {
blur(a, 1, src, src_step, len, radius);
for (; power > 2; power--) {
uint8_t *c;
blur(b, 1, a, 1, len, radius);
c = a; a = b; b = c;
}
if (power > 1) {
blur(dst, dst_step, a, 1, len, radius);
} else {
int i;
for (i = 0; i < len; i++)
dst[i*dst_step] = a[i];
}
} else {
int i;
for (i = 0; i < len; i++)
dst[i*dst_step] = src[i*src_step];
}
}

static void hblur(uint8_t *dst, int dst_linesize, const uint8_t *src, int src_linesize,
int w, int h, int radius, int power, uint8_t *temp[2])
{
int y;

if (radius == 0 && dst == src)
return;

for (y = 0; y < h; y++)
blur_power(dst + y*dst_linesize, 1, src + y*src_linesize, 1,
w, radius, power, temp);
}

static void vblur(uint8_t *dst, int dst_linesize, const uint8_t *src, int src_linesize,
int w, int h, int radius, int power, uint8_t *temp[2])
{
int x;

if (radius == 0 && dst == src)
return;

for (x = 0; x < w; x++)
blur_power(dst + x, dst_linesize, src + x, src_linesize,
h, radius, power, temp);
}

static void draw_slice(AVFilterLink *inlink, int y0, int h0, int slice_dir)
{
AVFilterContext *ctx = inlink->dst;
BoxBlurContext *boxblur = ctx->priv;
AVFilterLink *outlink = inlink->dst->outputs[0];
AVFilterBufferRef *inpicref = inlink ->cur_buf;
AVFilterBufferRef *outpicref = outlink->out_buf;
int plane;
int cw = inlink->w >> boxblur->hsub, ch = h0 >> boxblur->vsub;
int w[4] = { inlink->w, cw, cw, inlink->w };
int h[4] = { h0, ch, ch, h0 };

for (plane = 0; inpicref->data[plane] && plane < 4; plane++)
hblur(outpicref->data[plane], outpicref->linesize[plane],
inpicref ->data[plane], inpicref ->linesize[plane],
w[plane], h[plane], boxblur->radius[plane], boxblur->power[plane],
boxblur->temp);

for (plane = 0; inpicref->data[plane] && plane < 4; plane++)
vblur(outpicref->data[plane], outpicref->linesize[plane],
outpicref->data[plane], outpicref->linesize[plane],
w[plane], h[plane], boxblur->radius[plane], boxblur->power[plane],
boxblur->temp);

avfilter_draw_slice(outlink, y0, h0, slice_dir);
}

AVFilter avfilter_vf_boxblur = {
.name = "boxblur",
.description = NULL_IF_CONFIG_SMALL("Blur the input."),
.priv_size = sizeof(BoxBlurContext),
.init = init,
.uninit = uninit,
.query_formats = query_formats,

.inputs = (AVFilterPad[]) {{ .name = "default",
.type = AVMEDIA_TYPE_VIDEO,
.config_props = config_input,
.draw_slice = draw_slice,
.min_perms = AV_PERM_READ },
{ .name = NULL}},
.outputs = (AVFilterPad[]) {{ .name = "default",
.type = AVMEDIA_TYPE_VIDEO, },
{ .name = NULL}},
};

Write Preview
Loading…
Cancel
Save