FFmpeg
vf_cropdetect.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2002 A'rpi
3  * This file is part of FFmpeg.
4  *
5  * FFmpeg is free software; you can redistribute it and/or modify
6  * it under the terms of the GNU General Public License as published by
7  * the Free Software Foundation; either version 2 of the License, or
8  * (at your option) any later version.
9  *
10  * FFmpeg is distributed in the hope that it will be useful,
11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13  * GNU General Public License for more details.
14  *
15  * You should have received a copy of the GNU General Public License along
16  * with FFmpeg; if not, write to the Free Software Foundation, Inc.,
17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
18  */
19 
20 /**
21  * @file
22  * border detection filter
23  * Ported from MPlayer libmpcodecs/vf_cropdetect.c.
24  */
25 
26 #include "libavutil/imgutils.h"
27 #include "libavutil/internal.h"
28 #include "libavutil/opt.h"
29 
30 #include "avfilter.h"
31 #include "formats.h"
32 #include "internal.h"
33 #include "video.h"
34 
35 typedef struct CropDetectContext {
36  const AVClass *class;
37  int x1, y1, x2, y2;
38  float limit;
39  int round;
40  int skip;
42  int frame_nb;
43  int max_pixsteps[4];
46 
48 {
49  static const enum AVPixelFormat pix_fmts[] = {
64  };
65 
67  if (!fmts_list)
68  return AVERROR(ENOMEM);
69  return ff_set_common_formats(ctx, fmts_list);
70 }
71 
72 static int checkline(void *ctx, const unsigned char *src, int stride, int len, int bpp)
73 {
74  int total = 0;
75  int div = len;
76  const uint16_t *src16 = (const uint16_t *)src;
77 
78  switch (bpp) {
79  case 1:
80  while (len >= 8) {
81  total += src[ 0] + src[ stride] + src[2*stride] + src[3*stride]
82  + src[4*stride] + src[5*stride] + src[6*stride] + src[7*stride];
83  src += 8*stride;
84  len -= 8;
85  }
86  while (--len >= 0) {
87  total += src[0];
88  src += stride;
89  }
90  break;
91  case 2:
92  stride >>= 1;
93  while (len >= 8) {
94  total += src16[ 0] + src16[ stride] + src16[2*stride] + src16[3*stride]
95  + src16[4*stride] + src16[5*stride] + src16[6*stride] + src16[7*stride];
96  src16 += 8*stride;
97  len -= 8;
98  }
99  while (--len >= 0) {
100  total += src16[0];
101  src16 += stride;
102  }
103  break;
104  case 3:
105  case 4:
106  while (len >= 4) {
107  total += src[0] + src[1 ] + src[2 ]
108  + src[ stride] + src[1+ stride] + src[2+ stride]
109  + src[2*stride] + src[1+2*stride] + src[2+2*stride]
110  + src[3*stride] + src[1+3*stride] + src[2+3*stride];
111  src += 4*stride;
112  len -= 4;
113  }
114  while (--len >= 0) {
115  total += src[0] + src[1] + src[2];
116  src += stride;
117  }
118  div *= 3;
119  break;
120  }
121  total /= div;
122 
123  av_log(ctx, AV_LOG_DEBUG, "total:%d\n", total);
124  return total;
125 }
126 
128 {
129  CropDetectContext *s = ctx->priv;
130 
131  s->frame_nb = -1 * s->skip;
132 
133  av_log(ctx, AV_LOG_VERBOSE, "limit:%f round:%d skip:%d reset_count:%d\n",
134  s->limit, s->round, s->skip, s->reset_count);
135 
136  return 0;
137 }
138 
140 {
141  AVFilterContext *ctx = inlink->dst;
142  CropDetectContext *s = ctx->priv;
144 
145  av_image_fill_max_pixsteps(s->max_pixsteps, NULL, desc);
146 
147  if (s->limit < 1.0)
148  s->limit *= (1 << desc->comp[0].depth) - 1;
149 
150  s->x1 = inlink->w - 1;
151  s->y1 = inlink->h - 1;
152  s->x2 = 0;
153  s->y2 = 0;
154 
155  return 0;
156 }
157 
158 #define SET_META(key, value) \
159  av_dict_set_int(metadata, key, value, 0)
160 
162 {
163  AVFilterContext *ctx = inlink->dst;
164  CropDetectContext *s = ctx->priv;
165  int bpp = s->max_pixsteps[0];
166  int w, h, x, y, shrink_by;
167  AVDictionary **metadata;
168  int outliers, last_y;
169  int limit = lrint(s->limit);
170 
171  // ignore first s->skip frames
172  if (++s->frame_nb > 0) {
173  metadata = &frame->metadata;
174 
175  // Reset the crop area every reset_count frames, if reset_count is > 0
176  if (s->reset_count > 0 && s->frame_nb > s->reset_count) {
177  s->x1 = frame->width - 1;
178  s->y1 = frame->height - 1;
179  s->x2 = 0;
180  s->y2 = 0;
181  s->frame_nb = 1;
182  }
183 
184 #define FIND(DST, FROM, NOEND, INC, STEP0, STEP1, LEN) \
185  outliers = 0;\
186  for (last_y = y = FROM; NOEND; y = y INC) {\
187  if (checkline(ctx, frame->data[0] + STEP0 * y, STEP1, LEN, bpp) > limit) {\
188  if (++outliers > s->max_outliers) { \
189  DST = last_y;\
190  break;\
191  }\
192  } else\
193  last_y = y INC;\
194  }
195 
196  FIND(s->y1, 0, y < s->y1, +1, frame->linesize[0], bpp, frame->width);
197  FIND(s->y2, frame->height - 1, y > FFMAX(s->y2, s->y1), -1, frame->linesize[0], bpp, frame->width);
198  FIND(s->x1, 0, y < s->x1, +1, bpp, frame->linesize[0], frame->height);
199  FIND(s->x2, frame->width - 1, y > FFMAX(s->x2, s->x1), -1, bpp, frame->linesize[0], frame->height);
200 
201 
202  // round x and y (up), important for yuv colorspaces
203  // make sure they stay rounded!
204  x = (s->x1+1) & ~1;
205  y = (s->y1+1) & ~1;
206 
207  w = s->x2 - x + 1;
208  h = s->y2 - y + 1;
209 
210  // w and h must be divisible by 2 as well because of yuv
211  // colorspace problems.
212  if (s->round <= 1)
213  s->round = 16;
214  if (s->round % 2)
215  s->round *= 2;
216 
217  shrink_by = w % s->round;
218  w -= shrink_by;
219  x += (shrink_by/2 + 1) & ~1;
220 
221  shrink_by = h % s->round;
222  h -= shrink_by;
223  y += (shrink_by/2 + 1) & ~1;
224 
225  SET_META("lavfi.cropdetect.x1", s->x1);
226  SET_META("lavfi.cropdetect.x2", s->x2);
227  SET_META("lavfi.cropdetect.y1", s->y1);
228  SET_META("lavfi.cropdetect.y2", s->y2);
229  SET_META("lavfi.cropdetect.w", w);
230  SET_META("lavfi.cropdetect.h", h);
231  SET_META("lavfi.cropdetect.x", x);
232  SET_META("lavfi.cropdetect.y", y);
233 
235  "x1:%d x2:%d y1:%d y2:%d w:%d h:%d x:%d y:%d pts:%"PRId64" t:%f crop=%d:%d:%d:%d\n",
236  s->x1, s->x2, s->y1, s->y2, w, h, x, y, frame->pts,
237  frame->pts == AV_NOPTS_VALUE ? -1 : frame->pts * av_q2d(inlink->time_base),
238  w, h, x, y);
239  }
240 
241  return ff_filter_frame(inlink->dst->outputs[0], frame);
242 }
243 
244 #define OFFSET(x) offsetof(CropDetectContext, x)
245 #define FLAGS AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
246 
247 static const AVOption cropdetect_options[] = {
248  { "limit", "Threshold below which the pixel is considered black", OFFSET(limit), AV_OPT_TYPE_FLOAT, { .dbl = 24.0/255 }, 0, 65535, FLAGS },
249  { "round", "Value by which the width/height should be divisible", OFFSET(round), AV_OPT_TYPE_INT, { .i64 = 16 }, 0, INT_MAX, FLAGS },
250  { "reset", "Recalculate the crop area after this many frames", OFFSET(reset_count), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, FLAGS },
251  { "skip", "Number of initial frames to skip", OFFSET(skip), AV_OPT_TYPE_INT, { .i64 = 2 }, 0, INT_MAX, FLAGS },
252  { "reset_count", "Recalculate the crop area after this many frames",OFFSET(reset_count),AV_OPT_TYPE_INT,{ .i64 = 0 }, 0, INT_MAX, FLAGS },
253  { "max_outliers", "Threshold count of outliers", OFFSET(max_outliers),AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, FLAGS },
254  { NULL }
255 };
256 
257 AVFILTER_DEFINE_CLASS(cropdetect);
258 
260  {
261  .name = "default",
262  .type = AVMEDIA_TYPE_VIDEO,
263  .config_props = config_input,
264  .filter_frame = filter_frame,
265  },
266  { NULL }
267 };
268 
270  {
271  .name = "default",
272  .type = AVMEDIA_TYPE_VIDEO
273  },
274  { NULL }
275 };
276 
278  .name = "cropdetect",
279  .description = NULL_IF_CONFIG_SMALL("Auto-detect crop size."),
280  .priv_size = sizeof(CropDetectContext),
281  .priv_class = &cropdetect_class,
282  .init = init,
287 };
stride
int stride
Definition: mace.c:144
AVPixelFormat
AVPixelFormat
Pixel format.
Definition: pixfmt.h:64
init
static av_cold int init(AVFilterContext *ctx)
Definition: vf_cropdetect.c:127
AVERROR
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
opt.h
ff_make_format_list
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:286
ff_filter_frame
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:978
av_pix_fmt_desc_get
const AVPixFmtDescriptor * av_pix_fmt_desc_get(enum AVPixelFormat pix_fmt)
Definition: pixdesc.c:2541
inlink
The exact code depends on how similar the blocks are and how related they are to the and needs to apply these operations to the correct inlink or outlink if there are several Macros are available to factor that when no extra processing is inlink
Definition: filter_design.txt:212
SET_META
#define SET_META(key, value)
Definition: vf_cropdetect.c:158
AVFrame
This structure describes decoded (raw) audio or video data.
Definition: frame.h:303
w
uint8_t w
Definition: llviddspenc.c:39
AVOption
AVOption.
Definition: opt.h:248
AV_PIX_FMT_YUV420P10
#define AV_PIX_FMT_YUV420P10
Definition: pixfmt.h:389
AV_LOG_VERBOSE
#define AV_LOG_VERBOSE
Detailed information.
Definition: log.h:197
AV_PIX_FMT_BGR24
@ AV_PIX_FMT_BGR24
packed RGB 8:8:8, 24bpp, BGRBGR...
Definition: pixfmt.h:69
AV_PIX_FMT_BGRA
@ AV_PIX_FMT_BGRA
packed BGRA 8:8:8:8, 32bpp, BGRABGRA...
Definition: pixfmt.h:95
AV_PIX_FMT_YUV440P
@ AV_PIX_FMT_YUV440P
planar YUV 4:4:0 (1 Cr & Cb sample per 1x2 Y samples)
Definition: pixfmt.h:99
AVDictionary
Definition: dict.c:30
AVFilter::name
const char * name
Filter name.
Definition: avfilter.h:149
video.h
checkline
static int checkline(void *ctx, const unsigned char *src, int stride, int len, int bpp)
Definition: vf_cropdetect.c:72
AVFilterFormats
A list of supported formats for one end of a filter link.
Definition: formats.h:65
formats.h
AV_PIX_FMT_YUV422P9
#define AV_PIX_FMT_YUV422P9
Definition: pixfmt.h:387
AVFilterPad
A filter pad used for either input or output.
Definition: internal.h:54
AV_PIX_FMT_YUV444P10
#define AV_PIX_FMT_YUV444P10
Definition: pixfmt.h:392
lrint
#define lrint
Definition: tablegen.h:53
av_cold
#define av_cold
Definition: attributes.h:90
CropDetectContext::skip
int skip
Definition: vf_cropdetect.c:40
inputs
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several inputs
Definition: filter_design.txt:243
AV_PIX_FMT_YUV422P16
#define AV_PIX_FMT_YUV422P16
Definition: pixfmt.h:401
ff_set_common_formats
int ff_set_common_formats(AVFilterContext *ctx, AVFilterFormats *formats)
A helper for query_formats() which sets all links to the same list of formats.
Definition: formats.c:580
AV_PIX_FMT_YUVJ422P
@ AV_PIX_FMT_YUVJ422P
planar YUV 4:2:2, 16bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV422P and setting col...
Definition: pixfmt.h:79
CropDetectContext
Definition: vf_cropdetect.c:35
query_formats
static int query_formats(AVFilterContext *ctx)
Definition: vf_cropdetect.c:47
s
#define s(width, name)
Definition: cbs_vp9.c:257
AV_PIX_FMT_YUV444P16
#define AV_PIX_FMT_YUV444P16
Definition: pixfmt.h:402
av_q2d
static double av_q2d(AVRational a)
Convert an AVRational to a double.
Definition: rational.h:104
outputs
static const AVFilterPad outputs[]
Definition: af_acontrast.c:203
pix_fmts
static enum AVPixelFormat pix_fmts[]
Definition: libkvazaar.c:290
AV_PIX_FMT_YUV420P9
#define AV_PIX_FMT_YUV420P9
Definition: pixfmt.h:386
AV_LOG_DEBUG
#define AV_LOG_DEBUG
Stuff which is only useful for libav* developers.
Definition: log.h:202
AV_PIX_FMT_YUV420P16
#define AV_PIX_FMT_YUV420P16
Definition: pixfmt.h:400
ctx
AVFormatContext * ctx
Definition: movenc.c:48
AVFILTER_DEFINE_CLASS
AVFILTER_DEFINE_CLASS(cropdetect)
AV_PIX_FMT_YUV420P
@ AV_PIX_FMT_YUV420P
planar YUV 4:2:0, 12bpp, (1 Cr & Cb sample per 2x2 Y samples)
Definition: pixfmt.h:66
CropDetectContext::y1
int y1
Definition: vf_cropdetect.c:37
CropDetectContext::max_pixsteps
int max_pixsteps[4]
Definition: vf_cropdetect.c:43
AV_PIX_FMT_RGBA
@ AV_PIX_FMT_RGBA
packed RGBA 8:8:8:8, 32bpp, RGBARGBA...
Definition: pixfmt.h:93
AV_PIX_FMT_YUVJ444P
@ AV_PIX_FMT_YUVJ444P
planar YUV 4:4:4, 24bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV444P and setting col...
Definition: pixfmt.h:80
filter_frame
static int filter_frame(AVFilterLink *inlink, AVFrame *frame)
Definition: vf_cropdetect.c:161
AVClass
Describe the class of an AVClass context structure.
Definition: log.h:67
NULL
#define NULL
Definition: coverity.c:32
FLAGS
#define FLAGS
Definition: vf_cropdetect.c:245
AV_PIX_FMT_YUVJ420P
@ AV_PIX_FMT_YUVJ420P
planar YUV 4:2:0, 12bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV420P and setting col...
Definition: pixfmt.h:78
src
#define src
Definition: vp8dsp.c:255
CropDetectContext::y2
int y2
Definition: vf_cropdetect.c:37
cropdetect_options
static const AVOption cropdetect_options[]
Definition: vf_cropdetect.c:247
AV_PIX_FMT_YUV422P10
#define AV_PIX_FMT_YUV422P10
Definition: pixfmt.h:390
AV_PIX_FMT_GRAY8
@ AV_PIX_FMT_GRAY8
Y , 8bpp.
Definition: pixfmt.h:74
CropDetectContext::x1
int x1
Definition: vf_cropdetect.c:37
AV_PIX_FMT_RGB24
@ AV_PIX_FMT_RGB24
packed RGB 8:8:8, 24bpp, RGBRGB...
Definition: pixfmt.h:68
NULL_IF_CONFIG_SMALL
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
Definition: internal.h:117
FIND
#define FIND(DST, FROM, NOEND, INC, STEP0, STEP1, LEN)
FFMAX
#define FFMAX(a, b)
Definition: common.h:103
AV_PIX_FMT_YUV422P12
#define AV_PIX_FMT_YUV422P12
Definition: pixfmt.h:394
AV_NOPTS_VALUE
#define AV_NOPTS_VALUE
Undefined timestamp value.
Definition: avutil.h:248
AV_PIX_FMT_YUV444P12
#define AV_PIX_FMT_YUV444P12
Definition: pixfmt.h:396
OFFSET
#define OFFSET(x)
Definition: vf_cropdetect.c:244
CropDetectContext::max_outliers
int max_outliers
Definition: vf_cropdetect.c:44
CropDetectContext::reset_count
int reset_count
Definition: vf_cropdetect.c:41
CropDetectContext::limit
float limit
Definition: vf_cropdetect.c:38
CropDetectContext::round
int round
Definition: vf_cropdetect.c:39
config_input
static int config_input(AVFilterLink *inlink)
Definition: vf_cropdetect.c:139
AV_LOG_INFO
#define AV_LOG_INFO
Standard information.
Definition: log.h:192
CropDetectContext::frame_nb
int frame_nb
Definition: vf_cropdetect.c:42
internal.h
AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC
#define AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC
Some filters support a generic "enable" expression option that can be used to enable or disable a fil...
Definition: avfilter.h:126
AV_OPT_TYPE_FLOAT
@ AV_OPT_TYPE_FLOAT
Definition: opt.h:228
round
static av_always_inline av_const double round(double x)
Definition: libm.h:444
internal.h
AV_PIX_FMT_NV21
@ AV_PIX_FMT_NV21
as above, but U and V bytes are swapped
Definition: pixfmt.h:90
len
int len
Definition: vorbis_enc_data.h:426
AVFilterPad::name
const char * name
Pad name.
Definition: internal.h:60
AV_PIX_FMT_YUV444P9
#define AV_PIX_FMT_YUV444P9
Definition: pixfmt.h:388
AVFilter
Filter definition.
Definition: avfilter.h:145
AV_PIX_FMT_NV12
@ AV_PIX_FMT_NV12
planar YUV 4:2:0, 12bpp, 1 plane for Y and 1 plane for the UV components, which are interleaved (firs...
Definition: pixfmt.h:89
frame
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several the filter must be ready for frames arriving randomly on any input any filter with several inputs will most likely require some kind of queuing mechanism It is perfectly acceptable to have a limited queue and to drop frames when the inputs are too unbalanced request_frame For filters that do not use the this method is called when a frame is wanted on an output For a it should directly call filter_frame on the corresponding output For a if there are queued frames already one of these frames should be pushed If the filter should request a frame on one of its repeatedly until at least one frame has been pushed Return or at least make progress towards producing a frame
Definition: filter_design.txt:264
AV_PIX_FMT_YUV420P12
#define AV_PIX_FMT_YUV420P12
Definition: pixfmt.h:393
AV_PIX_FMT_YUV422P14
#define AV_PIX_FMT_YUV422P14
Definition: pixfmt.h:398
AV_PIX_FMT_NONE
@ AV_PIX_FMT_NONE
Definition: pixfmt.h:65
AV_OPT_TYPE_INT
@ AV_OPT_TYPE_INT
Definition: opt.h:225
avfilter.h
CropDetectContext::x2
int x2
Definition: vf_cropdetect.c:37
av_image_fill_max_pixsteps
void av_image_fill_max_pixsteps(int max_pixsteps[4], int max_pixstep_comps[4], const AVPixFmtDescriptor *pixdesc)
Compute the max pixel step for each plane of an image with a format described by pixdesc.
Definition: imgutils.c:35
AV_PIX_FMT_YUV444P
@ AV_PIX_FMT_YUV444P
planar YUV 4:4:4, 24bpp, (1 Cr & Cb sample per 1x1 Y samples)
Definition: pixfmt.h:71
AVFilterContext
An instance of a filter.
Definition: avfilter.h:333
desc
const char * desc
Definition: libsvtav1.c:79
AVMEDIA_TYPE_VIDEO
@ AVMEDIA_TYPE_VIDEO
Definition: avutil.h:201
AV_PIX_FMT_YUV422P
@ AV_PIX_FMT_YUV422P
planar YUV 4:2:2, 16bpp, (1 Cr & Cb sample per 2x1 Y samples)
Definition: pixfmt.h:70
avfilter_vf_cropdetect_inputs
static const AVFilterPad avfilter_vf_cropdetect_inputs[]
Definition: vf_cropdetect.c:259
avfilter_vf_cropdetect_outputs
static const AVFilterPad avfilter_vf_cropdetect_outputs[]
Definition: vf_cropdetect.c:269
AVPixFmtDescriptor
Descriptor that unambiguously describes how the bits of a pixel are stored in the up to 4 data planes...
Definition: pixdesc.h:70
AV_PIX_FMT_YUV411P
@ AV_PIX_FMT_YUV411P
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples)
Definition: pixfmt.h:73
imgutils.h
flags
#define flags(name, subs,...)
Definition: cbs_av1.c:561
AV_PIX_FMT_YUV410P
@ AV_PIX_FMT_YUV410P
planar YUV 4:1:0, 9bpp, (1 Cr & Cb sample per 4x4 Y samples)
Definition: pixfmt.h:72
av_log
#define av_log(a,...)
Definition: tableprint_vlc.h:28
h
h
Definition: vp9dsp_template.c:2038
AV_PIX_FMT_YUV444P14
#define AV_PIX_FMT_YUV444P14
Definition: pixfmt.h:399
AV_PIX_FMT_YUV420P14
#define AV_PIX_FMT_YUV420P14
Definition: pixfmt.h:397
ff_vf_cropdetect
const AVFilter ff_vf_cropdetect
Definition: vf_cropdetect.c:277