FFmpeg
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
vf_ocr.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2015 Paul B Mahol
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or
7  * modify it under the terms of the GNU Lesser General Public
8  * License as published by the Free Software Foundation; either
9  * version 2.1 of the License, or (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14  * Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with FFmpeg; if not, write to the Free Software
18  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19  */
20 
21 #include <tesseract/capi.h>
22 
23 #include "libavutil/opt.h"
24 #include "avfilter.h"
25 #include "formats.h"
26 #include "internal.h"
27 #include "video.h"
28 
29 typedef struct OCRContext {
30  const AVClass *class;
31 
32  char *datapath;
33  char *language;
34  char *whitelist;
35  char *blacklist;
36 
37  TessBaseAPI *tess;
38 } OCRContext;
39 
40 #define OFFSET(x) offsetof(OCRContext, x)
41 #define FLAGS AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
42 
43 static const AVOption ocr_options[] = {
44  { "datapath", "set datapath", OFFSET(datapath), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 0, FLAGS },
45  { "language", "set language", OFFSET(language), AV_OPT_TYPE_STRING, {.str="eng"}, 0, 0, FLAGS },
46  { "whitelist", "set character whitelist", OFFSET(whitelist), AV_OPT_TYPE_STRING, {.str="0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ.:;,-+_!?\"'[]{}()<>|/\\=*&%$#@!~"}, 0, 0, FLAGS },
47  { "blacklist", "set character blacklist", OFFSET(blacklist), AV_OPT_TYPE_STRING, {.str=""}, 0, 0, FLAGS },
48  { NULL }
49 };
50 
52 {
53  OCRContext *s = ctx->priv;
54 
55  s->tess = TessBaseAPICreate();
56  if (TessBaseAPIInit3(s->tess, s->datapath, s->language) == -1) {
57  av_log(ctx, AV_LOG_ERROR, "failed to init tesseract\n");
58  return AVERROR(EINVAL);
59  }
60 
61  if (!TessBaseAPISetVariable(s->tess, "tessedit_char_whitelist", s->whitelist)) {
62  av_log(ctx, AV_LOG_ERROR, "failed to set whitelist\n");
63  return AVERROR(EINVAL);
64  }
65 
66  if (!TessBaseAPISetVariable(s->tess, "tessedit_char_blacklist", s->blacklist)) {
67  av_log(ctx, AV_LOG_ERROR, "failed to set blacklist\n");
68  return AVERROR(EINVAL);
69  }
70 
71  av_log(ctx, AV_LOG_DEBUG, "Tesseract version: %s\n", TessVersion());
72 
73  return 0;
74 }
75 
77 {
78  static const enum AVPixelFormat pix_fmts[] = {
88  };
89 
90  AVFilterFormats *fmts_list = ff_make_format_list(pix_fmts);
91  if (!fmts_list)
92  return AVERROR(ENOMEM);
93  ff_set_common_formats(ctx, fmts_list);
94 
95  return 0;
96 }
97 
98 static int filter_frame(AVFilterLink *inlink, AVFrame *in)
99 {
100  AVDictionary **metadata = &in->metadata;
101  AVFilterContext *ctx = inlink->dst;
102  AVFilterLink *outlink = ctx->outputs[0];
103  OCRContext *s = ctx->priv;
104  char *result;
105 
106  result = TessBaseAPIRect(s->tess, in->data[0], 1,
107  in->linesize[0], 0, 0, in->width, in->height);
108  av_dict_set(metadata, "lavfi.ocr.text", result, 0);
109  TessDeleteText(result);
110 
111  return ff_filter_frame(outlink, in);
112 }
113 
115 {
116  OCRContext *s = ctx->priv;
117 
118  TessBaseAPIEnd(s->tess);
119  TessBaseAPIDelete(s->tess);
120 }
121 
123 
124 static const AVFilterPad ocr_inputs[] = {
125  {
126  .name = "default",
127  .type = AVMEDIA_TYPE_VIDEO,
128  .filter_frame = filter_frame,
129  },
130  { NULL }
131 };
132 
133 static const AVFilterPad ocr_outputs[] = {
134  {
135  .name = "default",
136  .type = AVMEDIA_TYPE_VIDEO,
137  },
138  { NULL }
139 };
140 
142  .name = "ocr",
143  .description = NULL_IF_CONFIG_SMALL("Optical Character Recognition."),
144  .priv_size = sizeof(OCRContext),
145  .priv_class = &ocr_class,
147  .init = init,
148  .uninit = uninit,
149  .inputs = ocr_inputs,
150  .outputs = ocr_outputs,
151 };
#define NULL
Definition: coverity.c:32
const char * s
Definition: avisynth_c.h:768
This structure describes decoded (raw) audio or video data.
Definition: frame.h:201
static const AVFilterPad ocr_outputs[]
Definition: vf_ocr.c:133
AVOption.
Definition: opt.h:246
planar YUV 4:4:4, 24bpp, (1 Cr & Cb sample per 1x1 Y samples)
Definition: pixfmt.h:67
Main libavfilter public API header.
AVFilter ff_vf_ocr
Definition: vf_ocr.c:141
AVFILTER_DEFINE_CLASS(ocr)
TessBaseAPI * tess
Definition: vf_ocr.c:37
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:283
static av_cold int init(AVFilterContext *ctx)
Definition: vf_ocr.c:51
const char * name
Pad name.
Definition: internal.h:60
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1151
planar YUV 4:2:0, 20bpp, (1 Cr & Cb sample per 2x2 Y & A samples)
Definition: pixfmt.h:102
#define FLAGS
Definition: vf_ocr.c:41
#define av_cold
Definition: attributes.h:82
static av_cold void uninit(AVFilterContext *ctx)
Definition: vf_ocr.c:114
AVOptions.
planar YUV 4:4:0 full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV440P and setting color_range...
Definition: pixfmt.h:101
planar YUV 4:2:2, 16bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV422P and setting col...
Definition: pixfmt.h:75
AVDictionary * metadata
metadata.
Definition: frame.h:488
static int query_formats(AVFilterContext *ctx)
Definition: vf_ocr.c:76
#define av_log(a,...)
A filter pad used for either input or output.
Definition: internal.h:54
planar YUV 4:2:2 24bpp, (1 Cr & Cb sample per 2x1 Y & A samples)
Definition: pixfmt.h:188
int width
Definition: frame.h:259
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:176
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:568
static const AVFilterPad ocr_inputs[]
Definition: vf_ocr.c:124
#define AVERROR(e)
Definition: error.h:43
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:179
void * priv
private data for use by the filter
Definition: avfilter.h:353
#define AV_LOG_DEBUG
Stuff which is only useful for libav* developers.
Definition: log.h:197
char * datapath
Definition: vf_ocr.c:32
planar YUV 4:2:2, 16bpp, (1 Cr & Cb sample per 2x1 Y samples)
Definition: pixfmt.h:66
char * blacklist
Definition: vf_ocr.c:35
planar YUV 4:2:0, 12bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV420P and setting col...
Definition: pixfmt.h:74
AVFormatContext * ctx
Definition: movenc.c:48
static const AVFilterPad outputs[]
Definition: af_afftfilt.c:389
static int filter_frame(AVFilterLink *inlink, AVFrame *in)
Definition: vf_ocr.c:98
static const AVFilterPad inputs[]
Definition: af_afftfilt.c:379
int linesize[AV_NUM_DATA_POINTERS]
For video, size in bytes of each picture line.
Definition: frame.h:232
planar YUV 4:4:4 32bpp, (1 Cr & Cb sample per 1x1 Y & A samples)
Definition: pixfmt.h:189
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(constint16_t *) pi >>8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(constint32_t *) pi >>24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(constfloat *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(constfloat *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(constfloat *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(constdouble *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(constdouble *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(constdouble *) pi *(1U<< 31))))#defineSET_CONV_FUNC_GROUP(ofmt, ifmt) staticvoidset_generic_function(AudioConvert *ac){}voidff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enumAVSampleFormatout_fmt, enumAVSampleFormatin_fmt, intchannels, intsample_rate, intapply_map){AudioConvert *ac;intin_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) returnNULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt)>2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);returnNULL;}returnac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}elseif(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;elseac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);returnac;}intff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){intuse_generic=1;intlen=in->nb_samples;intp;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%dsamples-audio_convert:%sto%s(dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));returnff_convert_dither(ac-> in
int av_dict_set(AVDictionary **pm, const char *key, const char *value, int flags)
Set the given entry in *pm, overwriting an existing entry.
Definition: dict.c:70
planar YUV 4:1:0, 9bpp, (1 Cr & Cb sample per 4x4 Y samples)
Definition: pixfmt.h:68
Describe the class of an AVClass context structure.
Definition: log.h:67
Filter definition.
Definition: avfilter.h:144
char * language
Definition: vf_ocr.c:33
const char * name
Filter name.
Definition: avfilter.h:148
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:350
static enum AVPixelFormat pix_fmts[]
Definition: libkvazaar.c:266
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:215
planar YUV 4:2:0, 12bpp, (1 Cr & Cb sample per 2x2 Y samples)
Definition: pixfmt.h:62
Y , 8bpp.
Definition: pixfmt.h:70
static const AVOption ocr_options[]
Definition: vf_ocr.c:43
planar YUV 4:4:4, 24bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV444P and setting col...
Definition: pixfmt.h:76
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples)
Definition: pixfmt.h:69
#define OFFSET(x)
Definition: vf_ocr.c:40
A list of supported formats for one end of a filter link.
Definition: formats.h:64
char * whitelist
Definition: vf_ocr.c:34
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples) full scale (JPEG), deprecated in favor ...
Definition: pixfmt.h:272
An instance of a filter.
Definition: avfilter.h:338
int height
Definition: frame.h:259
planar YUV 4:4:0 (1 Cr & Cb sample per 1x2 Y samples)
Definition: pixfmt.h:100
internal API functions
AVPixelFormat
Pixel format.
Definition: pixfmt.h:60