remove obsolete opencv filter wrapper

This commit is contained in:
Paul B Mahol 2024-09-19 18:53:07 +02:00
parent 408b71ede4
commit a8b4e3681c
6 changed files with 1 additions and 542 deletions

View File

@ -18,6 +18,7 @@ version <next>:
- removed DEC Alpha DSP and support code
- perlin video source filter
- YUV colorspace negotiation for codecs and filters, obsoleting YUVJ
- removed ocv filter wrapper
version 7.0:
- DXV DXT1 encoder

8
configure vendored
View File

@ -251,7 +251,6 @@ External library support:
--enable-libmp3lame enable MP3 encoding via libmp3lame [no]
--enable-libopencore-amrnb enable AMR-NB de/encoding via libopencore-amrnb [no]
--enable-libopencore-amrwb enable AMR-WB decoding via libopencore-amrwb [no]
--enable-libopencv enable video filtering via libopencv [no]
--enable-libopenh264 enable H.264 encoding via OpenH264 [no]
--enable-libopenjpeg enable JPEG 2000 encoding via OpenJPEG [no]
--enable-libopenmpt enable decoding tracked files via libopenmpt [no]
@ -1931,7 +1930,6 @@ EXTERNAL_LIBRARY_LIST="
libmodplug
libmp3lame
libmysofa
libopencv
libopenh264
libopenjpeg
libopenmpt
@ -2287,7 +2285,6 @@ HEADERS_LIST="
machine_ioctl_bt848_h
machine_ioctl_meteor_h
malloc_h
opencv2_core_core_c_h
OpenGL_gl3_h
poll_h
pthread_np_h
@ -3893,7 +3890,6 @@ nlmeans_opencl_filter_deps="opencl"
nlmeans_vulkan_filter_deps="vulkan spirv_compiler"
nnedi_filter_deps="gpl"
ocr_filter_deps="libtesseract"
ocv_filter_deps="libopencv"
openclsrc_filter_deps="opencl"
overlay_opencl_filter_deps="opencl"
overlay_qsv_filter_deps="libmfx"
@ -6944,10 +6940,6 @@ enabled libnpp && { check_lib libnpp npp.h nppGetLibVersion -lnppig -
die "ERROR: libnpp not found"; }
enabled libopencore_amrnb && require libopencore_amrnb opencore-amrnb/interf_dec.h Decoder_Interface_init -lopencore-amrnb
enabled libopencore_amrwb && require libopencore_amrwb opencore-amrwb/dec_if.h D_IF_init -lopencore-amrwb
enabled libopencv && { check_headers opencv2/core/core_c.h &&
{ check_pkg_config libopencv opencv opencv2/core/core_c.h cvCreateImageHeader ||
require libopencv opencv2/core/core_c.h cvCreateImageHeader -lopencv_core -lopencv_imgproc; } ||
require_pkg_config libopencv opencv opencv/cxcore.h cvCreateImageHeader; }
enabled libopenh264 && require_pkg_config libopenh264 "openh264 >= 1.3.0" wels/codec_api.h WelsGetCodecVersion
enabled libopenjpeg && { check_pkg_config libopenjpeg "libopenjp2 >= 2.1.0" openjpeg.h opj_version ||
{ require_pkg_config libopenjpeg "libopenjp2 >= 2.1.0" openjpeg.h opj_version -DOPJ_STATIC && add_cppflags -DOPJ_STATIC; } }

View File

@ -18641,110 +18641,6 @@ Set character blacklist.
The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
The filter exports confidence of recognized words as the frame metadata @code{lavfi.ocr.confidence}.
@section ocv
Apply a video transform using libopencv.
To enable this filter, install the libopencv library and headers and
configure FFmpeg with @code{--enable-libopencv}.
It accepts the following parameters:
@table @option
@item filter_name
The name of the libopencv filter to apply.
@item filter_params
The parameters to pass to the libopencv filter. If not specified, the default
values are assumed.
@end table
Refer to the official libopencv documentation for more precise
information:
@url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
Several libopencv filters are supported; see the following subsections.
@anchor{dilate}
@subsection dilate
Dilate an image by using a specific structuring element.
It corresponds to the libopencv function @code{cvDilate}.
It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
@var{struct_el} represents a structuring element, and has the syntax:
@var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
@var{cols} and @var{rows} represent the number of columns and rows of
the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
point, and @var{shape} the shape for the structuring element. @var{shape}
must be "rect", "cross", "ellipse", or "custom".
If the value for @var{shape} is "custom", it must be followed by a
string of the form "=@var{filename}". The file with name
@var{filename} is assumed to represent a binary image, with each
printable character corresponding to a bright pixel. When a custom
@var{shape} is used, @var{cols} and @var{rows} are ignored, the number
or columns and rows of the read file are assumed instead.
The default value for @var{struct_el} is "3x3+0x0/rect".
@var{nb_iterations} specifies the number of times the transform is
applied to the image, and defaults to 1.
Some examples:
@example
# Use the default values
ocv=dilate
# Dilate using a structuring element with a 5x5 cross, iterating two times
ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
# Read the shape from the file diamond.shape, iterating two times.
# The file diamond.shape may contain a pattern of characters like this
# *
# ***
# *****
# ***
# *
# The specified columns and rows are ignored
# but the anchor point coordinates are not
ocv=dilate:0x0+2x2/custom=diamond.shape|2
@end example
@subsection erode
Erode an image by using a specific structuring element.
It corresponds to the libopencv function @code{cvErode}.
It accepts the parameters: @var{struct_el}:@var{nb_iterations},
with the same syntax and semantics as the @ref{dilate} filter.
@subsection smooth
Smooth the input video.
The filter takes the following parameters:
@var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
@var{type} is the type of smooth filter to apply, and must be one of
the following values: "blur", "blur_no_scale", "median", "gaussian",
or "bilateral". The default value is "gaussian".
The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
depends on the smooth type. @var{param1} and
@var{param2} accept integer positive values or 0. @var{param3} and
@var{param4} accept floating point values.
The default value for @var{param1} is 3. The default value for the
other parameters is 0.
These parameters correspond to the parameters assigned to the
libopencv function @code{cvSmooth}.
@section oscilloscope
2D Video Oscilloscope.

View File

@ -427,7 +427,6 @@ OBJS-$(CONFIG_NOISE_FILTER) += vf_noise.o
OBJS-$(CONFIG_NORMALIZE_FILTER) += vf_normalize.o
OBJS-$(CONFIG_NULL_FILTER) += vf_null.o
OBJS-$(CONFIG_OCR_FILTER) += vf_ocr.o
OBJS-$(CONFIG_OCV_FILTER) += vf_libopencv.o
OBJS-$(CONFIG_OSCILLOSCOPE_FILTER) += vf_datascope.o
OBJS-$(CONFIG_OVERLAY_FILTER) += vf_overlay.o framesync.o
OBJS-$(CONFIG_OVERLAY_CUDA_FILTER) += vf_overlay_cuda.o framesync.o vf_overlay_cuda.ptx.o \

View File

@ -404,7 +404,6 @@ extern const AVFilter ff_vf_noise;
extern const AVFilter ff_vf_normalize;
extern const AVFilter ff_vf_null;
extern const AVFilter ff_vf_ocr;
extern const AVFilter ff_vf_ocv;
extern const AVFilter ff_vf_oscilloscope;
extern const AVFilter ff_vf_overlay;
extern const AVFilter ff_vf_overlay_opencl;

View File

@ -1,428 +0,0 @@
/*
* Copyright (c) 2010 Stefano Sabatini
*
* This file is part of FFmpeg.
*
* FFmpeg 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.
*
* FFmpeg 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 FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/**
* @file
* libopencv wrapper functions
*/
#include "config.h"
#if HAVE_OPENCV2_CORE_CORE_C_H
#include <opencv2/core/core_c.h>
#include <opencv2/imgproc/imgproc_c.h>
#else
#include <opencv/cv.h>
#include <opencv/cxcore.h>
#endif
#include "libavutil/avstring.h"
#include "libavutil/common.h"
#include "libavutil/file.h"
#include "libavutil/mem.h"
#include "libavutil/opt.h"
#include "avfilter.h"
#include "filters.h"
#include "formats.h"
#include "video.h"
static void fill_iplimage_from_frame(IplImage *img, const AVFrame *frame, enum AVPixelFormat pixfmt)
{
IplImage *tmpimg;
int depth, channels_nb;
if (pixfmt == AV_PIX_FMT_GRAY8) { depth = IPL_DEPTH_8U; channels_nb = 1; }
else if (pixfmt == AV_PIX_FMT_BGRA) { depth = IPL_DEPTH_8U; channels_nb = 4; }
else if (pixfmt == AV_PIX_FMT_BGR24) { depth = IPL_DEPTH_8U; channels_nb = 3; }
else return;
tmpimg = cvCreateImageHeader((CvSize){frame->width, frame->height}, depth, channels_nb);
*img = *tmpimg;
img->imageData = img->imageDataOrigin = frame->data[0];
img->dataOrder = IPL_DATA_ORDER_PIXEL;
img->origin = IPL_ORIGIN_TL;
img->widthStep = frame->linesize[0];
}
static void fill_frame_from_iplimage(AVFrame *frame, const IplImage *img, enum AVPixelFormat pixfmt)
{
frame->linesize[0] = img->widthStep;
frame->data[0] = img->imageData;
}
typedef struct OCVContext {
const AVClass *class;
char *name;
char *params;
int (*init)(AVFilterContext *ctx, const char *args);
void (*uninit)(AVFilterContext *ctx);
void (*end_frame_filter)(AVFilterContext *ctx, IplImage *inimg, IplImage *outimg);
void *priv;
} OCVContext;
typedef struct SmoothContext {
int type;
int param1, param2;
double param3, param4;
} SmoothContext;
static av_cold int smooth_init(AVFilterContext *ctx, const char *args)
{
OCVContext *s = ctx->priv;
SmoothContext *smooth = s->priv;
char type_str[128] = "gaussian";
smooth->param1 = 3;
smooth->param2 = 0;
smooth->param3 = 0.0;
smooth->param4 = 0.0;
if (args)
sscanf(args, "%127[^|]|%d|%d|%lf|%lf", type_str, &smooth->param1, &smooth->param2, &smooth->param3, &smooth->param4);
if (!strcmp(type_str, "blur" )) smooth->type = CV_BLUR;
else if (!strcmp(type_str, "blur_no_scale")) smooth->type = CV_BLUR_NO_SCALE;
else if (!strcmp(type_str, "median" )) smooth->type = CV_MEDIAN;
else if (!strcmp(type_str, "gaussian" )) smooth->type = CV_GAUSSIAN;
else if (!strcmp(type_str, "bilateral" )) smooth->type = CV_BILATERAL;
else {
av_log(ctx, AV_LOG_ERROR, "Smoothing type '%s' unknown.\n", type_str);
return AVERROR(EINVAL);
}
if (smooth->param1 < 0 || !(smooth->param1%2)) {
av_log(ctx, AV_LOG_ERROR,
"Invalid value '%d' for param1, it has to be a positive odd number\n",
smooth->param1);
return AVERROR(EINVAL);
}
if ((smooth->type == CV_BLUR || smooth->type == CV_BLUR_NO_SCALE || smooth->type == CV_GAUSSIAN) &&
(smooth->param2 < 0 || (smooth->param2 && !(smooth->param2%2)))) {
av_log(ctx, AV_LOG_ERROR,
"Invalid value '%d' for param2, it has to be zero or a positive odd number\n",
smooth->param2);
return AVERROR(EINVAL);
}
av_log(ctx, AV_LOG_VERBOSE, "type:%s param1:%d param2:%d param3:%f param4:%f\n",
type_str, smooth->param1, smooth->param2, smooth->param3, smooth->param4);
return 0;
}
static void smooth_end_frame_filter(AVFilterContext *ctx, IplImage *inimg, IplImage *outimg)
{
OCVContext *s = ctx->priv;
SmoothContext *smooth = s->priv;
cvSmooth(inimg, outimg, smooth->type, smooth->param1, smooth->param2, smooth->param3, smooth->param4);
}
static int read_shape_from_file(int *cols, int *rows, int **values, const char *filename,
void *log_ctx)
{
uint8_t *buf, *p, *pend;
size_t size;
int ret, i, j, w;
if ((ret = av_file_map(filename, &buf, &size, 0, log_ctx)) < 0)
return ret;
/* prescan file to get the number of lines and the maximum width */
w = 0;
for (i = 0; i < size; i++) {
if (buf[i] == '\n') {
if (*rows == INT_MAX) {
av_log(log_ctx, AV_LOG_ERROR, "Overflow on the number of rows in the file\n");
ret = AVERROR_INVALIDDATA;
goto end;
}
++(*rows);
*cols = FFMAX(*cols, w);
w = 0;
} else if (w == INT_MAX) {
av_log(log_ctx, AV_LOG_ERROR, "Overflow on the number of columns in the file\n");
return AVERROR_INVALIDDATA;
}
w++;
}
if (*rows > (SIZE_MAX / sizeof(int) / *cols)) {
av_log(log_ctx, AV_LOG_ERROR, "File with size %dx%d is too big\n",
*rows, *cols);
ret = AVERROR_INVALIDDATA;
goto end;
}
if (!(*values = av_calloc(sizeof(int) * *rows, *cols))) {
ret = AVERROR(ENOMEM);
goto end;
}
/* fill *values */
p = buf;
pend = buf + size-1;
for (i = 0; i < *rows; i++) {
for (j = 0;; j++) {
if (p > pend || *p == '\n') {
p++;
break;
} else
(*values)[*cols*i + j] = !!av_isgraph(*(p++));
}
}
end:
av_file_unmap(buf, size);
#ifdef DEBUG
{
char *line;
if (!(line = av_malloc(*cols + 1)))
return AVERROR(ENOMEM);
for (i = 0; i < *rows; i++) {
for (j = 0; j < *cols; j++)
line[j] = (*values)[i * *cols + j] ? '@' : ' ';
line[j] = 0;
av_log(log_ctx, AV_LOG_DEBUG, "%3d: %s\n", i, line);
}
av_free(line);
}
#endif
return 0;
}
static int parse_iplconvkernel(IplConvKernel **kernel, char *buf, void *log_ctx)
{
char shape_filename[128] = "", shape_str[32] = "rect";
int cols = 0, rows = 0, anchor_x = 0, anchor_y = 0, shape = CV_SHAPE_RECT;
int *values = NULL, ret = 0;
sscanf(buf, "%dx%d+%dx%d/%32[^=]=%127s", &cols, &rows, &anchor_x, &anchor_y, shape_str, shape_filename);
if (!strcmp(shape_str, "rect" )) shape = CV_SHAPE_RECT;
else if (!strcmp(shape_str, "cross" )) shape = CV_SHAPE_CROSS;
else if (!strcmp(shape_str, "ellipse")) shape = CV_SHAPE_ELLIPSE;
else if (!strcmp(shape_str, "custom" )) {
shape = CV_SHAPE_CUSTOM;
if ((ret = read_shape_from_file(&cols, &rows, &values, shape_filename, log_ctx)) < 0)
return ret;
} else {
av_log(log_ctx, AV_LOG_ERROR,
"Shape unspecified or type '%s' unknown.\n", shape_str);
ret = AVERROR(EINVAL);
goto out;
}
if (rows <= 0 || cols <= 0) {
av_log(log_ctx, AV_LOG_ERROR,
"Invalid non-positive values for shape size %dx%d\n", cols, rows);
ret = AVERROR(EINVAL);
goto out;
}
if (anchor_x < 0 || anchor_y < 0 || anchor_x >= cols || anchor_y >= rows) {
av_log(log_ctx, AV_LOG_ERROR,
"Shape anchor %dx%d is not inside the rectangle with size %dx%d.\n",
anchor_x, anchor_y, cols, rows);
ret = AVERROR(EINVAL);
goto out;
}
*kernel = cvCreateStructuringElementEx(cols, rows, anchor_x, anchor_y, shape, values);
if (!*kernel) {
ret = AVERROR(ENOMEM);
goto out;
}
av_log(log_ctx, AV_LOG_VERBOSE, "Structuring element: w:%d h:%d x:%d y:%d shape:%s\n",
rows, cols, anchor_x, anchor_y, shape_str);
out:
av_freep(&values);
return ret;
}
typedef struct DilateContext {
int nb_iterations;
IplConvKernel *kernel;
} DilateContext;
static av_cold int dilate_init(AVFilterContext *ctx, const char *args)
{
OCVContext *s = ctx->priv;
DilateContext *dilate = s->priv;
char default_kernel_str[] = "3x3+0x0/rect";
char *kernel_str = NULL;
const char *buf = args;
int ret;
if (args) {
kernel_str = av_get_token(&buf, "|");
if (!kernel_str)
return AVERROR(ENOMEM);
}
ret = parse_iplconvkernel(&dilate->kernel,
(!kernel_str || !*kernel_str) ? default_kernel_str
: kernel_str,
ctx);
av_free(kernel_str);
if (ret < 0)
return ret;
if (!buf || sscanf(buf, "|%d", &dilate->nb_iterations) != 1)
dilate->nb_iterations = 1;
av_log(ctx, AV_LOG_VERBOSE, "iterations_nb:%d\n", dilate->nb_iterations);
if (dilate->nb_iterations <= 0) {
av_log(ctx, AV_LOG_ERROR, "Invalid non-positive value '%d' for nb_iterations\n",
dilate->nb_iterations);
return AVERROR(EINVAL);
}
return 0;
}
static av_cold void dilate_uninit(AVFilterContext *ctx)
{
OCVContext *s = ctx->priv;
DilateContext *dilate = s->priv;
cvReleaseStructuringElement(&dilate->kernel);
}
static void dilate_end_frame_filter(AVFilterContext *ctx, IplImage *inimg, IplImage *outimg)
{
OCVContext *s = ctx->priv;
DilateContext *dilate = s->priv;
cvDilate(inimg, outimg, dilate->kernel, dilate->nb_iterations);
}
static void erode_end_frame_filter(AVFilterContext *ctx, IplImage *inimg, IplImage *outimg)
{
OCVContext *s = ctx->priv;
DilateContext *dilate = s->priv;
cvErode(inimg, outimg, dilate->kernel, dilate->nb_iterations);
}
typedef struct OCVFilterEntry {
const char *name;
size_t priv_size;
int (*init)(AVFilterContext *ctx, const char *args);
void (*uninit)(AVFilterContext *ctx);
void (*end_frame_filter)(AVFilterContext *ctx, IplImage *inimg, IplImage *outimg);
} OCVFilterEntry;
static const OCVFilterEntry ocv_filter_entries[] = {
{ "dilate", sizeof(DilateContext), dilate_init, dilate_uninit, dilate_end_frame_filter },
{ "erode", sizeof(DilateContext), dilate_init, dilate_uninit, erode_end_frame_filter },
{ "smooth", sizeof(SmoothContext), smooth_init, NULL, smooth_end_frame_filter },
};
static av_cold int init(AVFilterContext *ctx)
{
OCVContext *s = ctx->priv;
int i;
if (!s->name) {
av_log(ctx, AV_LOG_ERROR, "No libopencv filter name specified\n");
return AVERROR(EINVAL);
}
for (i = 0; i < FF_ARRAY_ELEMS(ocv_filter_entries); i++) {
const OCVFilterEntry *entry = &ocv_filter_entries[i];
if (!strcmp(s->name, entry->name)) {
s->init = entry->init;
s->uninit = entry->uninit;
s->end_frame_filter = entry->end_frame_filter;
if (!(s->priv = av_mallocz(entry->priv_size)))
return AVERROR(ENOMEM);
return s->init(ctx, s->params);
}
}
av_log(ctx, AV_LOG_ERROR, "No libopencv filter named '%s'\n", s->name);
return AVERROR(EINVAL);
}
static av_cold void uninit(AVFilterContext *ctx)
{
OCVContext *s = ctx->priv;
if (s->uninit)
s->uninit(ctx);
av_freep(&s->priv);
}
static int filter_frame(AVFilterLink *inlink, AVFrame *in)
{
AVFilterContext *ctx = inlink->dst;
OCVContext *s = ctx->priv;
AVFilterLink *outlink= inlink->dst->outputs[0];
AVFrame *out;
IplImage inimg, outimg;
out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
if (!out) {
av_frame_free(&in);
return AVERROR(ENOMEM);
}
av_frame_copy_props(out, in);
fill_iplimage_from_frame(&inimg , in , inlink->format);
fill_iplimage_from_frame(&outimg, out, inlink->format);
s->end_frame_filter(ctx, &inimg, &outimg);
fill_frame_from_iplimage(out, &outimg, inlink->format);
av_frame_free(&in);
return ff_filter_frame(outlink, out);
}
#define OFFSET(x) offsetof(OCVContext, x)
#define FLAGS AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_FILTERING_PARAM
static const AVOption ocv_options[] = {
{ "filter_name", NULL, OFFSET(name), AV_OPT_TYPE_STRING, .flags = FLAGS },
{ "filter_params", NULL, OFFSET(params), AV_OPT_TYPE_STRING, .flags = FLAGS },
{ NULL }
};
AVFILTER_DEFINE_CLASS(ocv);
static const AVFilterPad ocv_inputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_VIDEO,
.filter_frame = filter_frame,
},
};
static const AVFilterPad ocv_outputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_VIDEO,
},
};
const AVFilter ff_vf_ocv = {
.name = "ocv",
.description = NULL_IF_CONFIG_SMALL("Apply transform using libopencv."),
.priv_size = sizeof(OCVContext),
.priv_class = &ocv_class,
.init = init,
.uninit = uninit,
FILTER_INPUTS(ocv_inputs),
FILTER_OUTPUTS(ocv_outputs),
FILTER_PIXFMTS(AV_PIX_FMT_BGR24, AV_PIX_FMT_BGRA, AV_PIX_FMT_GRAY8),
};