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
init(AVFilterContext * ctx)51 static av_cold int init(AVFilterContext *ctx)
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
76 static const enum AVPixelFormat pix_fmts[] = {
77 AV_PIX_FMT_GRAY8,
78 AV_PIX_FMT_YUV410P, AV_PIX_FMT_YUV411P,
79 AV_PIX_FMT_YUV420P, AV_PIX_FMT_YUV422P,
80 AV_PIX_FMT_YUV440P, AV_PIX_FMT_YUV444P,
81 AV_PIX_FMT_YUVJ420P, AV_PIX_FMT_YUVJ422P,
82 AV_PIX_FMT_YUVJ440P, AV_PIX_FMT_YUVJ444P,
83 AV_PIX_FMT_YUVJ411P,
84 AV_PIX_FMT_YUVA444P, AV_PIX_FMT_YUVA422P, AV_PIX_FMT_YUVA420P,
85 AV_PIX_FMT_NONE
86 };
87
filter_frame(AVFilterLink * inlink,AVFrame * in)88 static int filter_frame(AVFilterLink *inlink, AVFrame *in)
89 {
90 AVDictionary **metadata = &in->metadata;
91 AVFilterContext *ctx = inlink->dst;
92 AVFilterLink *outlink = ctx->outputs[0];
93 OCRContext *s = ctx->priv;
94 char *result;
95 int *confs;
96
97 result = TessBaseAPIRect(s->tess, in->data[0], 1,
98 in->linesize[0], 0, 0, in->width, in->height);
99 confs = TessBaseAPIAllWordConfidences(s->tess);
100 av_dict_set(metadata, "lavfi.ocr.text", result, 0);
101 for (int i = 0; confs[i] != -1; i++) {
102 char number[256];
103
104 snprintf(number, sizeof(number), "%d ", confs[i]);
105 av_dict_set(metadata, "lavfi.ocr.confidence", number, AV_DICT_APPEND);
106 }
107
108 TessDeleteText(result);
109 TessDeleteIntArray(confs);
110
111 return ff_filter_frame(outlink, in);
112 }
113
uninit(AVFilterContext * ctx)114 static av_cold void uninit(AVFilterContext *ctx)
115 {
116 OCRContext *s = ctx->priv;
117
118 TessBaseAPIEnd(s->tess);
119 TessBaseAPIDelete(s->tess);
120 }
121
122 AVFILTER_DEFINE_CLASS(ocr);
123
124 static const AVFilterPad ocr_inputs[] = {
125 {
126 .name = "default",
127 .type = AVMEDIA_TYPE_VIDEO,
128 .filter_frame = filter_frame,
129 },
130 };
131
132 static const AVFilterPad ocr_outputs[] = {
133 {
134 .name = "default",
135 .type = AVMEDIA_TYPE_VIDEO,
136 },
137 };
138
139 const AVFilter ff_vf_ocr = {
140 .name = "ocr",
141 .description = NULL_IF_CONFIG_SMALL("Optical Character Recognition."),
142 .priv_size = sizeof(OCRContext),
143 .priv_class = &ocr_class,
144 .init = init,
145 .uninit = uninit,
146 .flags = AVFILTER_FLAG_METADATA_ONLY,
147 FILTER_INPUTS(ocr_inputs),
148 FILTER_OUTPUTS(ocr_outputs),
149 FILTER_PIXFMTS_ARRAY(pix_fmts),
150 };
151