1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
|
/**********************************************************************
* File: baseapi.cpp
* Description: Simple API for calling tesseract.
* Author: Ray Smith
*
* (C) Copyright 2006, Google Inc.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
** http://www.apache.org/licenses/LICENSE-2.0
** Unless required by applicable law or agreed to in writing, software
** distributed under the License is distributed on an "AS IS" BASIS,
** WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
** See the License for the specific language governing permissions and
** limitations under the License.
*
**********************************************************************/
#define _USE_MATH_DEFINES // for M_PI
// Include automatically generated configuration file if running autoconf.
#ifdef HAVE_CONFIG_H
#include "config_auto.h"
#endif
#ifndef DISABLED_LEGACY_ENGINE
#include "blobclass.h" // for ExtractFontName
#endif
#include "boxword.h" // for BoxWord
#include "coutln.h" // for C_OUTLINE_IT, C_OUTLINE_LIST
#include "dawg_cache.h" // for DawgCache
#include "dict.h" // for Dict
#include "edgblob.h" // for extract_edges
#include "elst.h" // for ELIST_ITERATOR, ELISTIZE, ELISTIZEH
#include "environ.h" // for l_uint8
#ifndef DISABLED_LEGACY_ENGINE
#include "equationdetect.h" // for EquationDetect
#endif
#include "errcode.h" // for ASSERT_HOST
#include "host.h" // for MAX_PATH
#include "imageio.h" // for IFF_TIFF_G4, IFF_TIFF, IFF_TIFF_G3, ...
#ifndef DISABLED_LEGACY_ENGINE
#include "intfx.h" // for INT_FX_RESULT_STRUCT
#endif
#include "mutableiterator.h" // for MutableIterator
#include "normalis.h" // for kBlnBaselineOffset, kBlnXHeight
#if defined(USE_OPENCL)
#include "openclwrapper.h" // for OpenclDevice
#endif
#include "pageres.h" // for PAGE_RES_IT, WERD_RES, PAGE_RES, CR_DE...
#include "paragraphs.h" // for DetectParagraphs
#include "params.h" // for BoolParam, IntParam, DoubleParam, Stri...
#include "pdblock.h" // for PDBLK
#include "points.h" // for FCOORD
#include "polyblk.h" // for POLY_BLOCK
#include "rect.h" // for TBOX
#include "stepblob.h" // for C_BLOB_IT, C_BLOB, C_BLOB_LIST
#include "tessdatamanager.h" // for TessdataManager, kTrainedDataSuffix
#include "tesseractclass.h" // for Tesseract
#include "tprintf.h" // for tprintf
#include "werd.h" // for WERD, WERD_IT, W_FUZZY_NON, W_FUZZY_SP
#include <tesseract/baseapi.h>
#include <tesseract/renderer.h> // for TessResultRenderer
#include <tesseract/resultiterator.h> // for ResultIterator
#include "strngs.h" // for STRING
#include <tesseract/thresholder.h> // for ImageThresholder
#include "helpers.h" // for IntCastRounded, chomp_string
#include <tesseract/ocrclass.h> // for ETEXT_DESC
#include <tesseract/osdetect.h> // for OSResults, OSBestResult, OrientationId...
#include <cmath> // for round, M_PI
#include <cstdint> // for int32_t
#include <cstring> // for strcmp, strcpy
#include <fstream> // for size_t
#include <iostream> // for std::cin
#include <locale> // for std::locale::classic
#include <memory> // for std::unique_ptr
#include <set> // for std::pair
#include <sstream> // for std::stringstream
#include <vector> // for std::vector
#include "allheaders.h" // for pixDestroy, boxCreate, boxaAddBox, box...
#ifdef HAVE_LIBCURL
#include <curl/curl.h>
#endif
#ifdef __linux__
#include <csignal> // for sigaction, SA_RESETHAND, SIGBUS, SIGFPE
#endif
#if defined(_WIN32)
#include <fcntl.h>
#include <io.h>
#else
#include <dirent.h> // for closedir, opendir, readdir, DIR, dirent
#include <libgen.h>
#include <sys/types.h>
#include <sys/stat.h> // for stat, S_IFDIR
#include <unistd.h>
#endif // _WIN32
namespace tesseract {
static BOOL_VAR(stream_filelist, false, "Stream a filelist from stdin");
static STRING_VAR(document_title, "", "Title of output document (used for hOCR and PDF output)");
/** Minimum sensible image size to be worth running tesseract. */
const int kMinRectSize = 10;
/** Character returned when Tesseract couldn't recognize as anything. */
const char kTesseractReject = '~';
/** Character used by UNLV error counter as a reject. */
const char kUNLVReject = '~';
/** Character used by UNLV as a suspect marker. */
const char kUNLVSuspect = '^';
/**
* Filename used for input image file, from which to derive a name to search
* for a possible UNLV zone file, if none is specified by SetInputName.
*/
static const char* kInputFile = "noname.tif";
/**
* Temp file used for storing current parameters before applying retry values.
*/
static const char* kOldVarsFile = "failed_vars.txt";
/** Max string length of an int. */
const int kMaxIntSize = 22;
/* Add all available languages recursively.
*/
static void addAvailableLanguages(const std::string& datadir, const std::string& base,
std::vector<std::string>* langs)
{
auto base2 = base;
if (!base2.empty())
base2 += "/";
const size_t extlen = sizeof(kTrainedDataSuffix);
#ifdef _WIN32
WIN32_FIND_DATA data;
HANDLE handle = FindFirstFile((datadir + base2 + "*").c_str(), &data);
if (handle != INVALID_HANDLE_VALUE) {
BOOL result = TRUE;
for (; result;) {
char *name = data.cFileName;
// Skip '.', '..', and hidden files
if (name[0] != '.') {
if ((data.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) ==
FILE_ATTRIBUTE_DIRECTORY) {
addAvailableLanguages(datadir, base2 + name, langs);
} else {
size_t len = strlen(name);
if (len > extlen && name[len - extlen] == '.' &&
strcmp(&name[len - extlen + 1], kTrainedDataSuffix) == 0) {
name[len - extlen] = '\0';
langs->push_back(base2 + name);
}
}
}
result = FindNextFile(handle, &data);
}
FindClose(handle);
}
#else // _WIN32
DIR* dir = opendir((datadir + base).c_str());
if (dir != nullptr) {
dirent *de;
while ((de = readdir(dir))) {
char *name = de->d_name;
// Skip '.', '..', and hidden files
if (name[0] != '.') {
struct stat st;
if (stat((datadir + base2 + name).c_str(), &st) == 0 &&
(st.st_mode & S_IFDIR) == S_IFDIR) {
addAvailableLanguages(datadir, base2 + name, langs);
} else {
size_t len = strlen(name);
if (len > extlen && name[len - extlen] == '.' &&
strcmp(&name[len - extlen + 1], kTrainedDataSuffix) == 0) {
name[len - extlen] = '\0';
langs->push_back(base2 + name);
}
}
}
}
closedir(dir);
}
#endif
}
TessBaseAPI::TessBaseAPI()
: tesseract_(nullptr),
osd_tesseract_(nullptr),
equ_detect_(nullptr),
reader_(nullptr),
// Thresholder is initialized to nullptr here, but will be set before use by:
// A constructor of a derived API, SetThresholder(), or
// created implicitly when used in InternalSetImage.
thresholder_(nullptr),
paragraph_models_(nullptr),
block_list_(nullptr),
page_res_(nullptr),
last_oem_requested_(OEM_DEFAULT),
recognition_done_(false),
rect_left_(0),
rect_top_(0),
rect_width_(0),
rect_height_(0),
image_width_(0),
image_height_(0) {
#if !defined(NDEBUG)
// The Tesseract executables would use the "C" locale by default,
// but other software which is linked against the Tesseract library
// typically uses the locale from the user's environment.
// Here the default is overridden to allow debugging of potential
// problems caused by the locale settings.
// Use the current locale if building debug code.
try {
std::locale::global(std::locale(""));
} catch (const std::runtime_error& ex) {
fprintf(stderr, "Warning: Could not set the current locale\n");
}
#endif
}
TessBaseAPI::~TessBaseAPI() {
End();
}
/**
* Returns the version identifier as a static string. Do not delete.
*/
const char* TessBaseAPI::Version() {
return TESSERACT_VERSION_STR;
}
/**
* If compiled with OpenCL AND an available OpenCL
* device is deemed faster than serial code, then
* "device" is populated with the cl_device_id
* and returns sizeof(cl_device_id)
* otherwise *device=nullptr and returns 0.
*/
size_t TessBaseAPI::getOpenCLDevice(void **data) {
#ifdef USE_OPENCL
ds_device device = OpenclDevice::getDeviceSelection();
if (device.type == DS_DEVICE_OPENCL_DEVICE) {
*data = new cl_device_id;
memcpy(*data, &device.oclDeviceID, sizeof(cl_device_id));
return sizeof(cl_device_id);
}
#endif
*data = nullptr;
return 0;
}
/**
* Set the name of the input file. Needed only for training and
* loading a UNLV zone file.
*/
void TessBaseAPI::SetInputName(const char* name) {
input_file_ = name ? name : "";
}
/** Set the name of the output files. Needed only for debugging. */
void TessBaseAPI::SetOutputName(const char* name) {
output_file_ = name ? name : "";
}
bool TessBaseAPI::SetVariable(const char* name, const char* value) {
if (tesseract_ == nullptr) tesseract_ = new Tesseract;
return ParamUtils::SetParam(name, value, SET_PARAM_CONSTRAINT_NON_INIT_ONLY,
tesseract_->params());
}
bool TessBaseAPI::SetDebugVariable(const char* name, const char* value) {
if (tesseract_ == nullptr) tesseract_ = new Tesseract;
return ParamUtils::SetParam(name, value, SET_PARAM_CONSTRAINT_DEBUG_ONLY,
tesseract_->params());
}
bool TessBaseAPI::GetIntVariable(const char *name, int *value) const {
auto *p = ParamUtils::FindParam<IntParam>(
name, GlobalParams()->int_params, tesseract_->params()->int_params);
if (p == nullptr) return false;
*value = (int32_t)(*p);
return true;
}
bool TessBaseAPI::GetBoolVariable(const char *name, bool *value) const {
auto *p = ParamUtils::FindParam<BoolParam>(
name, GlobalParams()->bool_params, tesseract_->params()->bool_params);
if (p == nullptr) return false;
*value = bool(*p);
return true;
}
const char *TessBaseAPI::GetStringVariable(const char *name) const {
auto *p = ParamUtils::FindParam<StringParam>(
name, GlobalParams()->string_params, tesseract_->params()->string_params);
return (p != nullptr) ? p->c_str() : nullptr;
}
bool TessBaseAPI::GetDoubleVariable(const char *name, double *value) const {
auto *p = ParamUtils::FindParam<DoubleParam>(
name, GlobalParams()->double_params, tesseract_->params()->double_params);
if (p == nullptr) return false;
*value = (double)(*p);
return true;
}
/** Get value of named variable as a string, if it exists. */
bool TessBaseAPI::GetVariableAsString(const char* name, std::string* val) {
return ParamUtils::GetParamAsString(name, tesseract_->params(), val);
}
/** Print Tesseract parameters to the given file. */
void TessBaseAPI::PrintVariables(FILE *fp) const {
ParamUtils::PrintParams(fp, tesseract_->params());
}
/**
* The datapath must be the name of the data directory or
* some other file in which the data directory resides (for instance argv[0].)
* The language is (usually) an ISO 639-3 string or nullptr will default to eng.
* If numeric_mode is true, then only digits and Roman numerals will
* be returned.
* @return: 0 on success and -1 on initialization failure.
*/
int TessBaseAPI::Init(const char* datapath, const char* language,
OcrEngineMode oem, char **configs, int configs_size,
const std::vector<std::string> *vars_vec,
const std::vector<std::string> *vars_values,
bool set_only_non_debug_params) {
return Init(datapath, 0, language, oem, configs, configs_size, vars_vec,
vars_values, set_only_non_debug_params, nullptr);
}
// In-memory version reads the traineddata file directly from the given
// data[data_size] array. Also implements the version with a datapath in data,
// flagged by data_size = 0.
int TessBaseAPI::Init(const char* data, int data_size, const char* language,
OcrEngineMode oem, char** configs, int configs_size,
const std::vector<std::string>* vars_vec,
const std::vector<std::string>* vars_values,
bool set_only_non_debug_params, FileReader reader) {
// Default language is "eng".
if (language == nullptr) language = "eng";
if (data == nullptr) {
data = "";
}
std::string datapath = data_size == 0 ? data : language;
// If the datapath, OcrEngineMode or the language have changed - start again.
// Note that the language_ field stores the last requested language that was
// initialized successfully, while tesseract_->lang stores the language
// actually used. They differ only if the requested language was nullptr, in
// which case tesseract_->lang is set to the Tesseract default ("eng").
if (tesseract_ != nullptr &&
(datapath_.empty() || language_.empty() || datapath_ != datapath ||
last_oem_requested_ != oem ||
(language_ != language && tesseract_->lang != language))) {
delete tesseract_;
tesseract_ = nullptr;
}
#ifdef USE_OPENCL
OpenclDevice od;
od.InitEnv();
#endif
bool reset_classifier = true;
if (tesseract_ == nullptr) {
reset_classifier = false;
tesseract_ = new Tesseract;
if (reader != nullptr) reader_ = reader;
TessdataManager mgr(reader_);
if (data_size != 0) {
mgr.LoadMemBuffer(language, data, data_size);
}
if (tesseract_->init_tesseract(
datapath.c_str(),
output_file_.c_str(),
language, oem, configs, configs_size, vars_vec, vars_values,
set_only_non_debug_params, &mgr) != 0) {
return -1;
}
}
// Update datapath and language requested for the last valid initialization.
datapath_ = datapath;
if ((strcmp(datapath_.c_str(), "") == 0) &&
(strcmp(tesseract_->datadir.c_str(), "") != 0))
datapath_ = tesseract_->datadir;
language_ = language;
last_oem_requested_ = oem;
#ifndef DISABLED_LEGACY_ENGINE
// For same language and datapath, just reset the adaptive classifier.
if (reset_classifier) {
tesseract_->ResetAdaptiveClassifier();
}
#endif // ndef DISABLED_LEGACY_ENGINE
return 0;
}
/**
* Returns the languages string used in the last valid initialization.
* If the last initialization specified "deu+hin" then that will be
* returned. If hin loaded eng automatically as well, then that will
* not be included in this list. To find the languages actually
* loaded use GetLoadedLanguagesAsVector.
* The returned string should NOT be deleted.
*/
const char* TessBaseAPI::GetInitLanguagesAsString() const {
return language_.c_str();
}
/**
* Returns the loaded languages in the vector of std::string.
* Includes all languages loaded by the last Init, including those loaded
* as dependencies of other loaded languages.
*/
void TessBaseAPI::GetLoadedLanguagesAsVector(
std::vector<std::string>* langs) const {
langs->clear();
if (tesseract_ != nullptr) {
langs->push_back(tesseract_->lang);
int num_subs = tesseract_->num_sub_langs();
for (int i = 0; i < num_subs; ++i)
langs->push_back(tesseract_->get_sub_lang(i)->lang);
}
}
/**
* Returns the available languages in the sorted vector of std::string.
*/
void TessBaseAPI::GetAvailableLanguagesAsVector(
std::vector<std::string>* langs) const {
langs->clear();
if (tesseract_ != nullptr) {
addAvailableLanguages(tesseract_->datadir, "", langs);
std::sort(langs->begin(), langs->end());
}
}
//TODO(amit): Adapt to lstm
#ifndef DISABLED_LEGACY_ENGINE
/**
* Init only the lang model component of Tesseract. The only functions
* that work after this init are SetVariable and IsValidWord.
* WARNING: temporary! This function will be removed from here and placed
* in a separate API at some future time.
*/
int TessBaseAPI::InitLangMod(const char* datapath, const char* language) {
if (tesseract_ == nullptr)
tesseract_ = new Tesseract;
else
ParamUtils::ResetToDefaults(tesseract_->params());
TessdataManager mgr;
return tesseract_->init_tesseract_lm(datapath, nullptr, language, &mgr);
}
#endif // ndef DISABLED_LEGACY_ENGINE
/**
* Init only for page layout analysis. Use only for calls to SetImage and
* AnalysePage. Calls that attempt recognition will generate an error.
*/
void TessBaseAPI::InitForAnalysePage() {
if (tesseract_ == nullptr) {
tesseract_ = new Tesseract;
#ifndef DISABLED_LEGACY_ENGINE
tesseract_->InitAdaptiveClassifier(nullptr);
#endif
}
}
/**
* Read a "config" file containing a set of parameter name, value pairs.
* Searches the standard places: tessdata/configs, tessdata/tessconfigs
* and also accepts a relative or absolute path name.
*/
void TessBaseAPI::ReadConfigFile(const char* filename) {
tesseract_->read_config_file(filename, SET_PARAM_CONSTRAINT_NON_INIT_ONLY);
}
/** Same as above, but only set debug params from the given config file. */
void TessBaseAPI::ReadDebugConfigFile(const char* filename) {
tesseract_->read_config_file(filename, SET_PARAM_CONSTRAINT_DEBUG_ONLY);
}
/**
* Set the current page segmentation mode. Defaults to PSM_AUTO.
* The mode is stored as an IntParam so it can also be modified by
* ReadConfigFile or SetVariable("tessedit_pageseg_mode", mode as string).
*/
void TessBaseAPI::SetPageSegMode(PageSegMode mode) {
if (tesseract_ == nullptr)
tesseract_ = new Tesseract;
tesseract_->tessedit_pageseg_mode.set_value(mode);
}
/** Return the current page segmentation mode. */
PageSegMode TessBaseAPI::GetPageSegMode() const {
if (tesseract_ == nullptr)
return PSM_SINGLE_BLOCK;
return static_cast<PageSegMode>(
static_cast<int>(tesseract_->tessedit_pageseg_mode));
}
/**
* Recognize a rectangle from an image and return the result as a string.
* May be called many times for a single Init.
* Currently has no error checking.
* Greyscale of 8 and color of 24 or 32 bits per pixel may be given.
* Palette color images will not work properly and must be converted to
* 24 bit.
* Binary images of 1 bit per pixel may also be given but they must be
* byte packed with the MSB of the first byte being the first pixel, and a
* one pixel is WHITE. For binary images set bytes_per_pixel=0.
* The recognized text is returned as a char* which is coded
* as UTF8 and must be freed with the delete [] operator.
*/
char* TessBaseAPI::TesseractRect(const unsigned char* imagedata,
int bytes_per_pixel,
int bytes_per_line,
int left, int top,
int width, int height) {
if (tesseract_ == nullptr || width < kMinRectSize || height < kMinRectSize)
return nullptr; // Nothing worth doing.
// Since this original api didn't give the exact size of the image,
// we have to invent a reasonable value.
int bits_per_pixel = bytes_per_pixel == 0 ? 1 : bytes_per_pixel * 8;
SetImage(imagedata, bytes_per_line * 8 / bits_per_pixel, height + top,
bytes_per_pixel, bytes_per_line);
SetRectangle(left, top, width, height);
return GetUTF8Text();
}
#ifndef DISABLED_LEGACY_ENGINE
/**
* Call between pages or documents etc to free up memory and forget
* adaptive data.
*/
void TessBaseAPI::ClearAdaptiveClassifier() {
if (tesseract_ == nullptr)
return;
tesseract_->ResetAdaptiveClassifier();
tesseract_->ResetDocumentDictionary();
}
#endif // ndef DISABLED_LEGACY_ENGINE
/**
* Provide an image for Tesseract to recognize. Format is as
* TesseractRect above. Copies the image buffer and converts to Pix.
* SetImage clears all recognition results, and sets the rectangle to the
* full image, so it may be followed immediately by a GetUTF8Text, and it
* will automatically perform recognition.
*/
void TessBaseAPI::SetImage(const unsigned char* imagedata,
int width, int height,
int bytes_per_pixel, int bytes_per_line) {
if (InternalSetImage()) {
thresholder_->SetImage(imagedata, width, height,
bytes_per_pixel, bytes_per_line);
SetInputImage(thresholder_->GetPixRect());
}
}
void TessBaseAPI::SetSourceResolution(int ppi) {
if (thresholder_)
thresholder_->SetSourceYResolution(ppi);
else
tprintf("Please call SetImage before SetSourceResolution.\n");
}
/**
* Provide an image for Tesseract to recognize. As with SetImage above,
* Tesseract takes its own copy of the image, so it need not persist until
* after Recognize.
* Pix vs raw, which to use?
* Use Pix where possible. Tesseract uses Pix as its internal representation
* and it is therefore more efficient to provide a Pix directly.
*/
void TessBaseAPI::SetImage(Pix* pix) {
if (InternalSetImage()) {
if (pixGetSpp(pix) == 4 && pixGetInputFormat(pix) == IFF_PNG) {
// remove alpha channel from png
Pix* p1 = pixRemoveAlpha(pix);
pixSetSpp(p1, 3);
(void)pixCopy(pix, p1);
pixDestroy(&p1);
}
thresholder_->SetImage(pix);
SetInputImage(thresholder_->GetPixRect());
}
}
/**
* Restrict recognition to a sub-rectangle of the image. Call after SetImage.
* Each SetRectangle clears the recogntion results so multiple rectangles
* can be recognized with the same image.
*/
void TessBaseAPI::SetRectangle(int left, int top, int width, int height) {
if (thresholder_ == nullptr)
return;
thresholder_->SetRectangle(left, top, width, height);
ClearResults();
}
/**
* ONLY available after SetImage if you have Leptonica installed.
* Get a copy of the internal thresholded image from Tesseract.
*/
Pix* TessBaseAPI::GetThresholdedImage() {
if (tesseract_ == nullptr || thresholder_ == nullptr) return nullptr;
if (tesseract_->pix_binary() == nullptr &&
!Threshold(tesseract_->mutable_pix_binary())) {
return nullptr;
}
return pixClone(tesseract_->pix_binary());
}
/**
* Get the result of page layout analysis as a leptonica-style
* Boxa, Pixa pair, in reading order.
* Can be called before or after Recognize.
*/
Boxa* TessBaseAPI::GetRegions(Pixa** pixa) {
return GetComponentImages(RIL_BLOCK, false, pixa, nullptr);
}
/**
* Get the textlines as a leptonica-style Boxa, Pixa pair, in reading order.
* Can be called before or after Recognize.
* If blockids is not nullptr, the block-id of each line is also returned as an
* array of one element per line. delete [] after use.
* If paraids is not nullptr, the paragraph-id of each line within its block is
* also returned as an array of one element per line. delete [] after use.
*/
Boxa* TessBaseAPI::GetTextlines(const bool raw_image, const int raw_padding,
Pixa** pixa, int** blockids, int** paraids) {
return GetComponentImages(RIL_TEXTLINE, true, raw_image, raw_padding,
pixa, blockids, paraids);
}
/**
* Get textlines and strips of image regions as a leptonica-style Boxa, Pixa
* pair, in reading order. Enables downstream handling of non-rectangular
* regions.
* Can be called before or after Recognize.
* If blockids is not nullptr, the block-id of each line is also returned as an
* array of one element per line. delete [] after use.
*/
Boxa* TessBaseAPI::GetStrips(Pixa** pixa, int** blockids) {
return GetComponentImages(RIL_TEXTLINE, false, pixa, blockids);
}
/**
* Get the words as a leptonica-style
* Boxa, Pixa pair, in reading order.
* Can be called before or after Recognize.
*/
Boxa* TessBaseAPI::GetWords(Pixa** pixa) {
return GetComponentImages(RIL_WORD, true, pixa, nullptr);
}
/**
* Gets the individual connected (text) components (created
* after pages segmentation step, but before recognition)
* as a leptonica-style Boxa, Pixa pair, in reading order.
* Can be called before or after Recognize.
*/
Boxa* TessBaseAPI::GetConnectedComponents(Pixa** pixa) {
return GetComponentImages(RIL_SYMBOL, true, pixa, nullptr);
}
/**
* Get the given level kind of components (block, textline, word etc.) as a
* leptonica-style Boxa, Pixa pair, in reading order.
* Can be called before or after Recognize.
* If blockids is not nullptr, the block-id of each component is also returned
* as an array of one element per component. delete [] after use.
* If text_only is true, then only text components are returned.
*/
Boxa* TessBaseAPI::GetComponentImages(PageIteratorLevel level,
bool text_only, bool raw_image,
const int raw_padding,
Pixa** pixa, int** blockids,
int** paraids) {
PageIterator* page_it = GetIterator();
if (page_it == nullptr)
page_it = AnalyseLayout();
if (page_it == nullptr)
return nullptr; // Failed.
// Count the components to get a size for the arrays.
int component_count = 0;
int left, top, right, bottom;
if (raw_image) {
// Get bounding box in original raw image with padding.
do {
if (page_it->BoundingBox(level, raw_padding,
&left, &top, &right, &bottom) &&
(!text_only || PTIsTextType(page_it->BlockType())))
++component_count;
} while (page_it->Next(level));
} else {
// Get bounding box from binarized imaged. Note that this could be
// differently scaled from the original image.
do {
if (page_it->BoundingBoxInternal(level, &left, &top, &right, &bottom) &&
(!text_only || PTIsTextType(page_it->BlockType())))
++component_count;
} while (page_it->Next(level));
}
Boxa* boxa = boxaCreate(component_count);
if (pixa != nullptr)
*pixa = pixaCreate(component_count);
if (blockids != nullptr)
*blockids = new int[component_count];
if (paraids != nullptr)
*paraids = new int[component_count];
int blockid = 0;
int paraid = 0;
int component_index = 0;
page_it->Begin();
do {
bool got_bounding_box;
if (raw_image) {
got_bounding_box =
page_it->BoundingBox(level, raw_padding, &left, &top, &right, &bottom);
} else {
got_bounding_box =
page_it->BoundingBoxInternal(level, &left, &top, &right, &bottom);
}
if (got_bounding_box &&
(!text_only || PTIsTextType(page_it->BlockType()))) {
Box* lbox = boxCreate(left, top, right - left, bottom - top);
boxaAddBox(boxa, lbox, L_INSERT);
if (pixa != nullptr) {
Pix* pix = nullptr;
if (raw_image) {
pix = page_it->GetImage(level, raw_padding, GetInputImage(), &left,
&top);
} else {
pix = page_it->GetBinaryImage(level);
}
pixaAddPix(*pixa, pix, L_INSERT);
pixaAddBox(*pixa, lbox, L_CLONE);
}
if (paraids != nullptr) {
(*paraids)[component_index] = paraid;
if (page_it->IsAtFinalElement(RIL_PARA, level))
++paraid;
}
if (blockids != nullptr) {
(*blockids)[component_index] = blockid;
if (page_it->IsAtFinalElement(RIL_BLOCK, level)) {
++blockid;
paraid = 0;
}
}
++component_index;
}
} while (page_it->Next(level));
delete page_it;
return boxa;
}
int TessBaseAPI::GetThresholdedImageScaleFactor() const {
if (thresholder_ == nullptr) {
return 0;
}
return thresholder_->GetScaleFactor();
}
/**
* Runs page layout analysis in the mode set by SetPageSegMode.
* May optionally be called prior to Recognize to get access to just
* the page layout results. Returns an iterator to the results.
* If merge_similar_words is true, words are combined where suitable for use
* with a line recognizer. Use if you want to use AnalyseLayout to find the
* textlines, and then want to process textline fragments with an external
* line recognizer.
* Returns nullptr on error or an empty page.
* The returned iterator must be deleted after use.
* WARNING! This class points to data held within the TessBaseAPI class, and
* therefore can only be used while the TessBaseAPI class still exists and
* has not been subjected to a call of Init, SetImage, Recognize, Clear, End
* DetectOS, or anything else that changes the internal PAGE_RES.
*/
PageIterator* TessBaseAPI::AnalyseLayout() { return AnalyseLayout(false); }
PageIterator* TessBaseAPI::AnalyseLayout(bool merge_similar_words) {
if (FindLines() == 0) {
if (block_list_->empty())
return nullptr; // The page was empty.
page_res_ = new PAGE_RES(merge_similar_words, block_list_, nullptr);
DetectParagraphs(false);
return new PageIterator(
page_res_, tesseract_, thresholder_->GetScaleFactor(),
thresholder_->GetScaledYResolution(),
rect_left_, rect_top_, rect_width_, rect_height_);
}
return nullptr;
}
/**
* Recognize the tesseract global image and return the result as Tesseract
* internal structures.
*/
int TessBaseAPI::Recognize(ETEXT_DESC* monitor) {
if (tesseract_ == nullptr)
return -1;
if (FindLines() != 0)
return -1;
delete page_res_;
if (block_list_->empty()) {
page_res_ = new PAGE_RES(false, block_list_,
&tesseract_->prev_word_best_choice_);
return 0; // Empty page.
}
tesseract_->SetBlackAndWhitelist();
recognition_done_ = true;
#ifndef DISABLED_LEGACY_ENGINE
if (tesseract_->tessedit_resegment_from_line_boxes) {
page_res_ = tesseract_->ApplyBoxes(input_file_.c_str(), true, block_list_);
} else if (tesseract_->tessedit_resegment_from_boxes) {
page_res_ = tesseract_->ApplyBoxes(input_file_.c_str(), false, block_list_);
} else
#endif // ndef DISABLED_LEGACY_ENGINE
{
page_res_ = new PAGE_RES(tesseract_->AnyLSTMLang(),
block_list_, &tesseract_->prev_word_best_choice_);
}
if (page_res_ == nullptr) {
return -1;
}
if (tesseract_->tessedit_train_line_recognizer) {
if (!tesseract_->TrainLineRecognizer(input_file_.c_str(), output_file_, block_list_)) {
return -1;
}
tesseract_->CorrectClassifyWords(page_res_);
return 0;
}
#ifndef DISABLED_LEGACY_ENGINE
if (tesseract_->tessedit_make_boxes_from_boxes) {
tesseract_->CorrectClassifyWords(page_res_);
return 0;
}
#endif // ndef DISABLED_LEGACY_ENGINE
int result = 0;
if (tesseract_->interactive_display_mode) {
#ifndef GRAPHICS_DISABLED
tesseract_->pgeditor_main(rect_width_, rect_height_, page_res_);
#endif // !GRAPHICS_DISABLED
// The page_res is invalid after an interactive session, so cleanup
// in a way that lets us continue to the next page without crashing.
delete page_res_;
page_res_ = nullptr;
return -1;
#ifndef DISABLED_LEGACY_ENGINE
} else if (tesseract_->tessedit_train_from_boxes) {
STRING fontname;
ExtractFontName(output_file_.c_str(), &fontname);
tesseract_->ApplyBoxTraining(fontname, page_res_);
} else if (tesseract_->tessedit_ambigs_training) {
FILE* training_output_file =
tesseract_->init_recog_training(input_file_.c_str());
// OCR the page segmented into words by tesseract.
tesseract_->recog_training_segmented(
input_file_.c_str(), page_res_, monitor, training_output_file);
fclose(training_output_file);
#endif // ndef DISABLED_LEGACY_ENGINE
} else {
// Now run the main recognition.
bool wait_for_text = true;
GetBoolVariable("paragraph_text_based", &wait_for_text);
if (!wait_for_text) DetectParagraphs(false);
if (tesseract_->recog_all_words(page_res_, monitor, nullptr, nullptr, 0)) {
if (wait_for_text) DetectParagraphs(true);
} else {
result = -1;
}
}
return result;
}
// Takes ownership of the input pix.
void TessBaseAPI::SetInputImage(Pix* pix) { tesseract_->set_pix_original(pix); }
Pix* TessBaseAPI::GetInputImage() { return tesseract_->pix_original(); }
const char * TessBaseAPI::GetInputName() {
if (!input_file_.empty()) {
return input_file_.c_str();
}
return nullptr;
}
const char * TessBaseAPI::GetDatapath() {
return tesseract_->datadir.c_str();
}
int TessBaseAPI::GetSourceYResolution() {
return thresholder_->GetSourceYResolution();
}
// If flist exists, get data from there. Otherwise get data from buf.
// Seems convoluted, but is the easiest way I know of to meet multiple
// goals. Support streaming from stdin, and also work on platforms
// lacking fmemopen.
// TODO: check different logic for flist/buf and simplify.
bool TessBaseAPI::ProcessPagesFileList(FILE *flist,
std::string *buf,
const char* retry_config,
int timeout_millisec,
TessResultRenderer* renderer,
int tessedit_page_number) {
if (!flist && !buf) return false;
int page = (tessedit_page_number >= 0) ? tessedit_page_number : 0;
char pagename[MAX_PATH];
std::vector<std::string> lines;
if (!flist) {
std::string line;
for (const auto ch : *buf) {
if (ch == '\n') {
lines.push_back(line);
line.clear();
} else {
line.push_back(ch);
}
}
if (lines.empty()) return false;
}
// Skip to the requested page number.
for (int i = 0; i < page; i++) {
if (flist) {
if (fgets(pagename, sizeof(pagename), flist) == nullptr) break;
}
}
// Begin producing output
if (renderer && !renderer->BeginDocument(document_title.c_str())) {
return false;
}
// Loop over all pages - or just the requested one
while (true) {
if (flist) {
if (fgets(pagename, sizeof(pagename), flist) == nullptr) break;
} else {
if (page >= lines.size()) break;
snprintf(pagename, sizeof(pagename), "%s", lines[page].c_str());
}
chomp_string(pagename);
Pix *pix = pixRead(pagename);
if (pix == nullptr) {
tprintf("Image file %s cannot be read!\n", pagename);
return false;
}
tprintf("Page %d : %s\n", page, pagename);
bool r = ProcessPage(pix, page, pagename, retry_config,
timeout_millisec, renderer);
pixDestroy(&pix);
if (!r) return false;
if (tessedit_page_number >= 0) break;
++page;
}
// Finish producing output
if (renderer && !renderer->EndDocument()) {
return false;
}
return true;
}
bool TessBaseAPI::ProcessPagesMultipageTiff(const l_uint8 *data,
size_t size,
const char* filename,
const char* retry_config,
int timeout_millisec,
TessResultRenderer* renderer,
int tessedit_page_number) {
Pix *pix = nullptr;
int page = (tessedit_page_number >= 0) ? tessedit_page_number : 0;
size_t offset = 0;
for (; ; ++page) {
if (tessedit_page_number >= 0) {
page = tessedit_page_number;
pix = (data) ? pixReadMemTiff(data, size, page)
: pixReadTiff(filename, page);
} else {
pix = (data) ? pixReadMemFromMultipageTiff(data, size, &offset)
: pixReadFromMultipageTiff(filename, &offset);
}
if (pix == nullptr) break;
tprintf("Page %d\n", page + 1);
char page_str[kMaxIntSize];
snprintf(page_str, kMaxIntSize - 1, "%d", page);
SetVariable("applybox_page", page_str);
bool r = ProcessPage(pix, page, filename, retry_config,
timeout_millisec, renderer);
pixDestroy(&pix);
if (!r) return false;
if (tessedit_page_number >= 0) break;
if (!offset) break;
}
return true;
}
// Master ProcessPages calls ProcessPagesInternal and then does any post-
// processing required due to being in a training mode.
bool TessBaseAPI::ProcessPages(const char* filename, const char* retry_config,
int timeout_millisec,
TessResultRenderer* renderer) {
bool result =
ProcessPagesInternal(filename, retry_config, timeout_millisec, renderer);
#ifndef DISABLED_LEGACY_ENGINE
if (result) {
if (tesseract_->tessedit_train_from_boxes &&
!tesseract_->WriteTRFile(output_file_.c_str())) {
tprintf("Write of TR file failed: %s\n", output_file_.c_str());
return false;
}
}
#endif // ndef DISABLED_LEGACY_ENGINE
return result;
}
static size_t
WriteMemoryCallback(void *contents, size_t size, size_t nmemb, void *userp)
{
size = size * nmemb;
std::string* buf = reinterpret_cast<std::string*>(userp);
buf->append(reinterpret_cast<const char*>(contents), size);
return size;
}
// In the ideal scenario, Tesseract will start working on data as soon
// as it can. For example, if you stream a filelist through stdin, we
// should start the OCR process as soon as the first filename is
// available. This is particularly useful when hooking Tesseract up to
// slow hardware such as a book scanning machine.
//
// Unfortunately there are tradeoffs. You can't seek on stdin. That
// makes automatic detection of datatype (TIFF? filelist? PNG?)
// impractical. So we support a command line flag to explicitly
// identify the scenario that really matters: filelists on
// stdin. We'll still do our best if the user likes pipes.
bool TessBaseAPI::ProcessPagesInternal(const char* filename,
const char* retry_config,
int timeout_millisec,
TessResultRenderer* renderer) {
bool stdInput = !strcmp(filename, "stdin") || !strcmp(filename, "-");
if (stdInput) {
#ifdef WIN32
if (_setmode(_fileno(stdin), _O_BINARY) == -1)
tprintf("ERROR: cin to binary: %s", strerror(errno));
#endif // WIN32
}
if (stream_filelist) {
return ProcessPagesFileList(stdin, nullptr, retry_config,
timeout_millisec, renderer,
tesseract_->tessedit_page_number);
}
// At this point we are officially in autodection territory.
// That means any data in stdin must be buffered, to make it
// seekable.
std::string buf;
const l_uint8 *data = nullptr;
if (stdInput) {
buf.assign((std::istreambuf_iterator<char>(std::cin)),
(std::istreambuf_iterator<char>()));
data = reinterpret_cast<const l_uint8 *>(buf.data());
} else if (strncmp(filename, "http:", 5) == 0 ||
strncmp(filename, "https:", 6) == 0 ) {
// Get image or image list by URL.
#ifdef HAVE_LIBCURL
CURL* curl = curl_easy_init();
if (curl == nullptr) {
fprintf(stderr, "Error, curl_easy_init failed\n");
return false;
} else {
CURLcode curlcode;
auto error = [curl, &curlcode](const char* function) {
fprintf(stderr, "Error, %s failed with error %s\n", function,
curl_easy_strerror(curlcode));
curl_easy_cleanup(curl);
return false;
};
curlcode = curl_easy_setopt(curl, CURLOPT_URL, filename);
if (curlcode != CURLE_OK) {
return error("curl_easy_setopt");
}
curlcode = curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, WriteMemoryCallback);
if (curlcode != CURLE_OK) {
return error("curl_easy_setopt");
}
curlcode = curl_easy_setopt(curl, CURLOPT_WRITEDATA, &buf);
if (curlcode != CURLE_OK) {
return error("curl_easy_setopt");
}
curlcode = curl_easy_perform(curl);
if (curlcode != CURLE_OK) {
return error("curl_easy_perform");
}
curl_easy_cleanup(curl);
data = reinterpret_cast<const l_uint8 *>(buf.data());
}
#else
fprintf(stderr, "Error, this tesseract has no URL support\n");
return false;
#endif
} else {
// Check whether the input file can be read.
if (FILE* file = fopen(filename, "rb")) {
fclose(file);
} else {
fprintf(stderr, "Error, cannot read input file %s: %s\n",
filename, strerror(errno));
return false;
}
}
// Here is our autodetection
int format;
int r = (data != nullptr) ?
findFileFormatBuffer(data, &format) :
findFileFormat(filename, &format);
// Maybe we have a filelist
if (r != 0 || format == IFF_UNKNOWN) {
std::string s;
if (data != nullptr) {
s = buf.c_str();
} else {
std::ifstream t(filename);
std::string u((std::istreambuf_iterator<char>(t)),
std::istreambuf_iterator<char>());
s = u.c_str();
}
return ProcessPagesFileList(nullptr, &s, retry_config,
timeout_millisec, renderer,
tesseract_->tessedit_page_number);
}
// Maybe we have a TIFF which is potentially multipage
bool tiff = (format == IFF_TIFF || format == IFF_TIFF_PACKBITS ||
format == IFF_TIFF_RLE || format == IFF_TIFF_G3 ||
format == IFF_TIFF_G4 || format == IFF_TIFF_LZW ||
#if LIBLEPT_MAJOR_VERSION > 1 || LIBLEPT_MINOR_VERSION > 76
format == IFF_TIFF_JPEG ||
#endif
format == IFF_TIFF_ZIP);
// Fail early if we can, before producing any output
Pix *pix = nullptr;
if (!tiff) {
pix = (data != nullptr) ? pixReadMem(data, buf.size()) : pixRead(filename);
if (pix == nullptr) {
return false;
}
}
// Begin the output
if (renderer && !renderer->BeginDocument(document_title.c_str())) {
pixDestroy(&pix);
return false;
}
// Produce output
r = (tiff) ?
ProcessPagesMultipageTiff(data, buf.size(), filename, retry_config,
timeout_millisec, renderer,
tesseract_->tessedit_page_number) :
ProcessPage(pix, 0, filename, retry_config,
timeout_millisec, renderer);
// Clean up memory as needed
pixDestroy(&pix);
// End the output
if (!r || (renderer && !renderer->EndDocument())) {
return false;
}
return true;
}
bool TessBaseAPI::ProcessPage(Pix* pix, int page_index, const char* filename,
const char* retry_config, int timeout_millisec,
TessResultRenderer* renderer) {
SetInputName(filename);
SetImage(pix);
bool failed = false;
if (tesseract_->tessedit_pageseg_mode == PSM_AUTO_ONLY) {
// Disabled character recognition
PageIterator* it = AnalyseLayout();
if (it == nullptr) {
failed = true;
} else {
delete it;
}
} else if (tesseract_->tessedit_pageseg_mode == PSM_OSD_ONLY) {
failed = FindLines() != 0;
} else if (timeout_millisec > 0) {
// Running with a timeout.
ETEXT_DESC monitor;
monitor.cancel = nullptr;
monitor.cancel_this = nullptr;
monitor.set_deadline_msecs(timeout_millisec);
// Now run the main recognition.
failed = Recognize(&monitor) < 0;
} else {
// Normal layout and character recognition with no timeout.
failed = Recognize(nullptr) < 0;
}
if (tesseract_->tessedit_write_images) {
Pix* page_pix = GetThresholdedImage();
pixWrite("tessinput.tif", page_pix, IFF_TIFF_G4);
}
if (failed && retry_config != nullptr && retry_config[0] != '\0') {
// Save current config variables before switching modes.
FILE* fp = fopen(kOldVarsFile, "wb");
if (fp == nullptr) {
tprintf("Error, failed to open file \"%s\"\n", kOldVarsFile);
} else {
PrintVariables(fp);
fclose(fp);
}
// Switch to alternate mode for retry.
ReadConfigFile(retry_config);
SetImage(pix);
Recognize(nullptr);
// Restore saved config variables.
ReadConfigFile(kOldVarsFile);
}
if (renderer && !failed) {
failed = !renderer->AddImage(this);
}
return !failed;
}
/**
* Get a left-to-right iterator to the results of LayoutAnalysis and/or
* Recognize. The returned iterator must be deleted after use.
*/
LTRResultIterator* TessBaseAPI::GetLTRIterator() {
if (tesseract_ == nullptr || page_res_ == nullptr)
return nullptr;
return new LTRResultIterator(
page_res_, tesseract_,
thresholder_->GetScaleFactor(), thresholder_->GetScaledYResolution(),
rect_left_, rect_top_, rect_width_, rect_height_);
}
/**
* Get a reading-order iterator to the results of LayoutAnalysis and/or
* Recognize. The returned iterator must be deleted after use.
* WARNING! This class points to data held within the TessBaseAPI class, and
* therefore can only be used while the TessBaseAPI class still exists and
* has not been subjected to a call of Init, SetImage, Recognize, Clear, End
* DetectOS, or anything else that changes the internal PAGE_RES.
*/
ResultIterator* TessBaseAPI::GetIterator() {
if (tesseract_ == nullptr || page_res_ == nullptr)
return nullptr;
return ResultIterator::StartOfParagraph(LTRResultIterator(
page_res_, tesseract_,
thresholder_->GetScaleFactor(), thresholder_->GetScaledYResolution(),
rect_left_, rect_top_, rect_width_, rect_height_));
}
/**
* Get a mutable iterator to the results of LayoutAnalysis and/or Recognize.
* The returned iterator must be deleted after use.
* WARNING! This class points to data held within the TessBaseAPI class, and
* therefore can only be used while the TessBaseAPI class still exists and
* has not been subjected to a call of Init, SetImage, Recognize, Clear, End
* DetectOS, or anything else that changes the internal PAGE_RES.
*/
MutableIterator* TessBaseAPI::GetMutableIterator() {
if (tesseract_ == nullptr || page_res_ == nullptr)
return nullptr;
return new MutableIterator(page_res_, tesseract_,
thresholder_->GetScaleFactor(),
thresholder_->GetScaledYResolution(),
rect_left_, rect_top_, rect_width_, rect_height_);
}
/** Make a text string from the internal data structures. */
char* TessBaseAPI::GetUTF8Text() {
if (tesseract_ == nullptr ||
(!recognition_done_ && Recognize(nullptr) < 0))
return nullptr;
std::string text("");
ResultIterator *it = GetIterator();
do {
if (it->Empty(RIL_PARA)) continue;
const std::unique_ptr<const char[]> para_text(it->GetUTF8Text(RIL_PARA));
text += para_text.get();
} while (it->Next(RIL_PARA));
char* result = new char[text.length() + 1];
strncpy(result, text.c_str(), text.length() + 1);
delete it;
return result;
}
static void AddBoxToTSV(const PageIterator* it, PageIteratorLevel level,
STRING* text) {
int left, top, right, bottom;
it->BoundingBox(level, &left, &top, &right, &bottom);
text->add_str_int("\t", left);
text->add_str_int("\t", top);
text->add_str_int("\t", right - left);
text->add_str_int("\t", bottom - top);
}
/**
* Make a TSV-formatted string from the internal data structures.
* page_number is 0-based but will appear in the output as 1-based.
* Returned string must be freed with the delete [] operator.
*/
char* TessBaseAPI::GetTSVText(int page_number) {
if (tesseract_ == nullptr || (page_res_ == nullptr && Recognize(nullptr) < 0))
return nullptr;
int lcnt = 1, bcnt = 1, pcnt = 1, wcnt = 1;
int page_id = page_number + 1; // we use 1-based page numbers.
STRING tsv_str("");
int page_num = page_id;
int block_num = 0;
int par_num = 0;
int line_num = 0;
int word_num = 0;
tsv_str.add_str_int("1\t", page_num); // level 1 - page
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
tsv_str.add_str_int("\t", rect_left_);
tsv_str.add_str_int("\t", rect_top_);
tsv_str.add_str_int("\t", rect_width_);
tsv_str.add_str_int("\t", rect_height_);
tsv_str += "\t-1\t\n";
ResultIterator* res_it = GetIterator();
while (!res_it->Empty(RIL_BLOCK)) {
if (res_it->Empty(RIL_WORD)) {
res_it->Next(RIL_WORD);
continue;
}
// Add rows for any new block/paragraph/textline.
if (res_it->IsAtBeginningOf(RIL_BLOCK)) {
block_num++;
par_num = 0;
line_num = 0;
word_num = 0;
tsv_str.add_str_int("2\t", page_num); // level 2 - block
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
AddBoxToTSV(res_it, RIL_BLOCK, &tsv_str);
tsv_str += "\t-1\t\n"; // end of row for block
}
if (res_it->IsAtBeginningOf(RIL_PARA)) {
par_num++;
line_num = 0;
word_num = 0;
tsv_str.add_str_int("3\t", page_num); // level 3 - paragraph
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
AddBoxToTSV(res_it, RIL_PARA, &tsv_str);
tsv_str += "\t-1\t\n"; // end of row for para
}
if (res_it->IsAtBeginningOf(RIL_TEXTLINE)) {
line_num++;
word_num = 0;
tsv_str.add_str_int("4\t", page_num); // level 4 - line
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
AddBoxToTSV(res_it, RIL_TEXTLINE, &tsv_str);
tsv_str += "\t-1\t\n"; // end of row for line
}
// Now, process the word...
int left, top, right, bottom;
res_it->BoundingBox(RIL_WORD, &left, &top, &right, &bottom);
word_num++;
tsv_str.add_str_int("5\t", page_num); // level 5 - word
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
tsv_str.add_str_int("\t", left);
tsv_str.add_str_int("\t", top);
tsv_str.add_str_int("\t", right - left);
tsv_str.add_str_int("\t", bottom - top);
tsv_str.add_str_int("\t", res_it->Confidence(RIL_WORD));
tsv_str += "\t";
// Increment counts if at end of block/paragraph/textline.
if (res_it->IsAtFinalElement(RIL_TEXTLINE, RIL_WORD)) lcnt++;
if (res_it->IsAtFinalElement(RIL_PARA, RIL_WORD)) pcnt++;
if (res_it->IsAtFinalElement(RIL_BLOCK, RIL_WORD)) bcnt++;
do {
tsv_str +=
std::unique_ptr<const char[]>(res_it->GetUTF8Text(RIL_SYMBOL)).get();
res_it->Next(RIL_SYMBOL);
} while (!res_it->Empty(RIL_BLOCK) && !res_it->IsAtBeginningOf(RIL_WORD));
tsv_str += "\n"; // end of row
wcnt++;
}
char* ret = new char[tsv_str.length() + 1];
strcpy(ret, tsv_str.c_str());
delete res_it;
return ret;
}
/** The 5 numbers output for each box (the usual 4 and a page number.) */
const int kNumbersPerBlob = 5;
/**
* The number of bytes taken by each number. Since we use int16_t for ICOORD,
* assume only 5 digits max.
*/
const int kBytesPerNumber = 5;
/**
* Multiplier for max expected textlength assumes (kBytesPerNumber + space)
* * kNumbersPerBlob plus the newline. Add to this the
* original UTF8 characters, and one kMaxBytesPerLine for safety.
*/
const int kBytesPerBoxFileLine = (kBytesPerNumber + 1) * kNumbersPerBlob + 1;
/** Max bytes in the decimal representation of int64_t. */
const int kBytesPer64BitNumber = 20;
/**
* A maximal single box could occupy kNumbersPerBlob numbers at
* kBytesPer64BitNumber digits (if someone sneaks in a 64 bit value) and a
* space plus the newline and the maximum length of a UNICHAR.
* Test against this on each iteration for safety.
*/
const int kMaxBytesPerLine = kNumbersPerBlob * (kBytesPer64BitNumber + 1) + 1 +
UNICHAR_LEN;
/**
* The recognized text is returned as a char* which is coded
* as a UTF8 box file.
* page_number is a 0-base page index that will appear in the box file.
* Returned string must be freed with the delete [] operator.
*/
char* TessBaseAPI::GetBoxText(int page_number) {
if (tesseract_ == nullptr ||
(!recognition_done_ && Recognize(nullptr) < 0))
return nullptr;
int blob_count;
int utf8_length = TextLength(&blob_count);
int total_length = blob_count * kBytesPerBoxFileLine + utf8_length +
kMaxBytesPerLine;
char* result = new char[total_length];
result[0] = '\0';
int output_length = 0;
LTRResultIterator* it = GetLTRIterator();
do {
int left, top, right, bottom;
if (it->BoundingBox(RIL_SYMBOL, &left, &top, &right, &bottom)) {
const std::unique_ptr</*non-const*/ char[]> text(
it->GetUTF8Text(RIL_SYMBOL));
// Tesseract uses space for recognition failure. Fix to a reject
// character, kTesseractReject so we don't create illegal box files.
for (int i = 0; text[i] != '\0'; ++i) {
if (text[i] == ' ')
text[i] = kTesseractReject;
}
snprintf(result + output_length, total_length - output_length,
"%s %d %d %d %d %d\n", text.get(), left, image_height_ - bottom,
right, image_height_ - top, page_number);
output_length += strlen(result + output_length);
// Just in case...
if (output_length + kMaxBytesPerLine > total_length)
break;
}
} while (it->Next(RIL_SYMBOL));
delete it;
return result;
}
/**
* Conversion table for non-latin characters.
* Maps characters out of the latin set into the latin set.
* TODO(rays) incorporate this translation into unicharset.
*/
const int kUniChs[] = {
0x20ac, 0x201c, 0x201d, 0x2018, 0x2019, 0x2022, 0x2014, 0
};
/** Latin chars corresponding to the unicode chars above. */
const int kLatinChs[] = {
0x00a2, 0x0022, 0x0022, 0x0027, 0x0027, 0x00b7, 0x002d, 0
};
/**
* The recognized text is returned as a char* which is coded
* as UNLV format Latin-1 with specific reject and suspect codes.
* Returned string must be freed with the delete [] operator.
*/
char* TessBaseAPI::GetUNLVText() {
if (tesseract_ == nullptr ||
(!recognition_done_ && Recognize(nullptr) < 0))
return nullptr;
bool tilde_crunch_written = false;
bool last_char_was_newline = true;
bool last_char_was_tilde = false;
int total_length = TextLength(nullptr);
PAGE_RES_IT page_res_it(page_res_);
char* result = new char[total_length];
char* ptr = result;
for (page_res_it.restart_page(); page_res_it.word () != nullptr;
page_res_it.forward()) {
WERD_RES *word = page_res_it.word();
// Process the current word.
if (word->unlv_crunch_mode != CR_NONE) {
if (word->unlv_crunch_mode != CR_DELETE &&
(!tilde_crunch_written ||
(word->unlv_crunch_mode == CR_KEEP_SPACE &&
word->word->space() > 0 &&
!word->word->flag(W_FUZZY_NON) &&
!word->word->flag(W_FUZZY_SP)))) {
if (!word->word->flag(W_BOL) &&
word->word->space() > 0 &&
!word->word->flag(W_FUZZY_NON) &&
!word->word->flag(W_FUZZY_SP)) {
/* Write a space to separate from preceding good text */
*ptr++ = ' ';
last_char_was_tilde = false;
}
if (!last_char_was_tilde) {
// Write a reject char.
last_char_was_tilde = true;
*ptr++ = kUNLVReject;
tilde_crunch_written = true;
last_char_was_newline = false;
}
}
} else {
// NORMAL PROCESSING of non tilde crunched words.
tilde_crunch_written = false;
tesseract_->set_unlv_suspects(word);
const char* wordstr = word->best_choice->unichar_string().c_str();
const STRING& lengths = word->best_choice->unichar_lengths();
int length = lengths.length();
int i = 0;
int offset = 0;
if (last_char_was_tilde &&
word->word->space() == 0 && wordstr[offset] == ' ') {
// Prevent adjacent tilde across words - we know that adjacent tildes
// within words have been removed.
// Skip the first character.
offset = lengths[i++];
}
if (i < length && wordstr[offset] != 0) {
if (!last_char_was_newline)
*ptr++ = ' ';
else
last_char_was_newline = false;
for (; i < length; offset += lengths[i++]) {
if (wordstr[offset] == ' ' ||
wordstr[offset] == kTesseractReject) {
*ptr++ = kUNLVReject;
last_char_was_tilde = true;
} else {
if (word->reject_map[i].rejected())
*ptr++ = kUNLVSuspect;
UNICHAR ch(wordstr + offset, lengths[i]);
int uni_ch = ch.first_uni();
for (int j = 0; kUniChs[j] != 0; ++j) {
if (kUniChs[j] == uni_ch) {
uni_ch = kLatinChs[j];
break;
}
}
if (uni_ch <= 0xff) {
*ptr++ = static_cast<char>(uni_ch);
last_char_was_tilde = false;
} else {
*ptr++ = kUNLVReject;
last_char_was_tilde = true;
}
}
}
}
}
if (word->word->flag(W_EOL) && !last_char_was_newline) {
/* Add a new line output */
*ptr++ = '\n';
tilde_crunch_written = false;
last_char_was_newline = true;
last_char_was_tilde = false;
}
}
*ptr++ = '\n';
*ptr = '\0';
return result;
}
#ifndef DISABLED_LEGACY_ENGINE
/**
* Detect the orientation of the input image and apparent script (alphabet).
* orient_deg is the detected clockwise rotation of the input image in degrees
* (0, 90, 180, 270)
* orient_conf is the confidence (15.0 is reasonably confident)
* script_name is an ASCII string, the name of the script, e.g. "Latin"
* script_conf is confidence level in the script
* Returns true on success and writes values to each parameter as an output
*/
bool TessBaseAPI::DetectOrientationScript(int* orient_deg, float* orient_conf,
const char** script_name,
float* script_conf) {
OSResults osr;
bool osd = DetectOS(&osr);
if (!osd) {
return false;
}
int orient_id = osr.best_result.orientation_id;
int script_id = osr.get_best_script(orient_id);
if (orient_conf) *orient_conf = osr.best_result.oconfidence;
if (orient_deg) *orient_deg = orient_id * 90; // convert quadrant to degrees
if (script_name) {
const char* script = osr.unicharset->get_script_from_script_id(script_id);
*script_name = script;
}
if (script_conf) *script_conf = osr.best_result.sconfidence;
return true;
}
/**
* The recognized text is returned as a char* which is coded
* as UTF8 and must be freed with the delete [] operator.
* page_number is a 0-based page index that will appear in the osd file.
*/
char* TessBaseAPI::GetOsdText(int page_number) {
int orient_deg;
float orient_conf;
const char* script_name;
float script_conf;
if (!DetectOrientationScript(&orient_deg, &orient_conf, &script_name,
&script_conf))
return nullptr;
// clockwise rotation needed to make the page upright
int rotate = OrientationIdToValue(orient_deg / 90);
std::stringstream stream;
// Use "C" locale (needed for float values orient_conf and script_conf).
stream.imbue(std::locale::classic());
// Use fixed notation with 2 digits after the decimal point for float values.
stream.precision(2);
stream
<< std::fixed
<< "Page number: " << page_number << "\n"
<< "Orientation in degrees: " << orient_deg << "\n"
<< "Rotate: " << rotate << "\n"
<< "Orientation confidence: " << orient_conf << "\n"
<< "Script: " << script_name << "\n"
<< "Script confidence: " << script_conf << "\n";
const std::string& text = stream.str();
char* result = new char[text.length() + 1];
strcpy(result, text.c_str());
return result;
}
#endif // ndef DISABLED_LEGACY_ENGINE
/** Returns the average word confidence for Tesseract page result. */
int TessBaseAPI::MeanTextConf() {
int* conf = AllWordConfidences();
if (!conf) return 0;
int sum = 0;
int *pt = conf;
while (*pt >= 0) sum += *pt++;
if (pt != conf) sum /= pt - conf;
delete [] conf;
return sum;
}
/** Returns an array of all word confidences, terminated by -1. */
int* TessBaseAPI::AllWordConfidences() {
if (tesseract_ == nullptr ||
(!recognition_done_ && Recognize(nullptr) < 0))
return nullptr;
int n_word = 0;
PAGE_RES_IT res_it(page_res_);
for (res_it.restart_page(); res_it.word() != nullptr; res_it.forward())
n_word++;
int* conf = new int[n_word+1];
n_word = 0;
for (res_it.restart_page(); res_it.word() != nullptr; res_it.forward()) {
WERD_RES *word = res_it.word();
WERD_CHOICE* choice = word->best_choice;
int w_conf = static_cast<int>(100 + 5 * choice->certainty());
// This is the eq for converting Tesseract confidence to 1..100
if (w_conf < 0) w_conf = 0;
if (w_conf > 100) w_conf = 100;
conf[n_word++] = w_conf;
}
conf[n_word] = -1;
return conf;
}
#ifndef DISABLED_LEGACY_ENGINE
/**
* Applies the given word to the adaptive classifier if possible.
* The word must be SPACE-DELIMITED UTF-8 - l i k e t h i s , so it can
* tell the boundaries of the graphemes.
* Assumes that SetImage/SetRectangle have been used to set the image
* to the given word. The mode arg should be PSM_SINGLE_WORD or
* PSM_CIRCLE_WORD, as that will be used to control layout analysis.
* The currently set PageSegMode is preserved.
* Returns false if adaption was not possible for some reason.
*/
bool TessBaseAPI::AdaptToWordStr(PageSegMode mode, const char* wordstr) {
int debug = 0;
GetIntVariable("applybox_debug", &debug);
bool success = true;
PageSegMode current_psm = GetPageSegMode();
SetPageSegMode(mode);
SetVariable("classify_enable_learning", "0");
const std::unique_ptr<const char[]> text(GetUTF8Text());
if (debug) {
tprintf("Trying to adapt \"%s\" to \"%s\"\n", text.get(), wordstr);
}
if (text != nullptr) {
PAGE_RES_IT it(page_res_);
WERD_RES* word_res = it.word();
if (word_res != nullptr) {
word_res->word->set_text(wordstr);
// Check to see if text matches wordstr.
int w = 0;
int t;
for (t = 0; text[t] != '\0'; ++t) {
if (text[t] == '\n' || text[t] == ' ')
continue;
while (wordstr[w] == ' ') ++w;
if (text[t] != wordstr[w])
break;
++w;
}
if (text[t] != '\0' || wordstr[w] != '\0') {
// No match.
delete page_res_;
std::vector<TBOX> boxes;
page_res_ = tesseract_->SetupApplyBoxes(boxes, block_list_);
tesseract_->ReSegmentByClassification(page_res_);
tesseract_->TidyUp(page_res_);
PAGE_RES_IT pr_it(page_res_);
if (pr_it.word() == nullptr)
success = false;
else
word_res = pr_it.word();
} else {
word_res->BestChoiceToCorrectText();
}
if (success) {
tesseract_->EnableLearning = true;
tesseract_->LearnWord(nullptr, word_res);
}
} else {
success = false;
}
} else {
success = false;
}
SetPageSegMode(current_psm);
return success;
}
#endif // ndef DISABLED_LEGACY_ENGINE
/**
* Free up recognition results and any stored image data, without actually
* freeing any recognition data that would be time-consuming to reload.
* Afterwards, you must call SetImage or TesseractRect before doing
* any Recognize or Get* operation.
*/
void TessBaseAPI::Clear() {
if (thresholder_ != nullptr)
thresholder_->Clear();
ClearResults();
if (tesseract_ != nullptr) SetInputImage(nullptr);
}
/**
* Close down tesseract and free up all memory. End() is equivalent to
* destructing and reconstructing your TessBaseAPI.
* Once End() has been used, none of the other API functions may be used
* other than Init and anything declared above it in the class definition.
*/
void TessBaseAPI::End() {
Clear();
delete thresholder_;
thresholder_ = nullptr;
delete page_res_;
page_res_ = nullptr;
delete block_list_;
block_list_ = nullptr;
if (paragraph_models_ != nullptr) {
for (auto model : *paragraph_models_) {
delete model;
}
delete paragraph_models_;
paragraph_models_ = nullptr;
}
if (osd_tesseract_ == tesseract_) osd_tesseract_ = nullptr;
delete tesseract_;
tesseract_ = nullptr;
delete osd_tesseract_;
osd_tesseract_ = nullptr;
delete equ_detect_;
equ_detect_ = nullptr;
input_file_.clear();
output_file_.clear();
datapath_.clear();
language_.clear();
}
// Clear any library-level memory caches.
// There are a variety of expensive-to-load constant data structures (mostly
// language dictionaries) that are cached globally -- surviving the Init()
// and End() of individual TessBaseAPI's. This function allows the clearing
// of these caches.
void TessBaseAPI::ClearPersistentCache() {
Dict::GlobalDawgCache()->DeleteUnusedDawgs();
}
/**
* Check whether a word is valid according to Tesseract's language model
* returns 0 if the word is invalid, non-zero if valid
*/
int TessBaseAPI::IsValidWord(const char *word) {
return tesseract_->getDict().valid_word(word);
}
// Returns true if utf8_character is defined in the UniCharset.
bool TessBaseAPI::IsValidCharacter(const char *utf8_character) {
return tesseract_->unicharset.contains_unichar(utf8_character);
}
// TODO(rays) Obsolete this function and replace with a more aptly named
// function that returns image coordinates rather than tesseract coordinates.
bool TessBaseAPI::GetTextDirection(int* out_offset, float* out_slope) {
PageIterator* it = AnalyseLayout();
if (it == nullptr) {
return false;
}
int x1, x2, y1, y2;
it->Baseline(RIL_TEXTLINE, &x1, &y1, &x2, &y2);
// Calculate offset and slope (NOTE: Kind of ugly)
if (x2 <= x1) x2 = x1 + 1;
// Convert the point pair to slope/offset of the baseline (in image coords.)
*out_slope = static_cast<float>(y2 - y1) / (x2 - x1);
*out_offset = static_cast<int>(y1 - *out_slope * x1);
// Get the y-coord of the baseline at the left and right edges of the
// textline's bounding box.
int left, top, right, bottom;
if (!it->BoundingBox(RIL_TEXTLINE, &left, &top, &right, &bottom)) {
delete it;
return false;
}
int left_y = IntCastRounded(*out_slope * left + *out_offset);
int right_y = IntCastRounded(*out_slope * right + *out_offset);
// Shift the baseline down so it passes through the nearest bottom-corner
// of the textline's bounding box. This is the difference between the y
// at the lowest (max) edge of the box and the actual box bottom.
*out_offset += bottom - std::max(left_y, right_y);
// Switch back to bottom-up tesseract coordinates. Requires negation of
// the slope and height - offset for the offset.
*out_slope = -*out_slope;
*out_offset = rect_height_ - *out_offset;
delete it;
return true;
}
/** Sets Dict::letter_is_okay_ function to point to the given function. */
void TessBaseAPI::SetDictFunc(DictFunc f) {
if (tesseract_ != nullptr) {
tesseract_->getDict().letter_is_okay_ = f;
}
}
/**
* Sets Dict::probability_in_context_ function to point to the given
* function.
*
* @param f A single function that returns the probability of the current
* "character" (in general a utf-8 string), given the context of a previous
* utf-8 string.
*/
void TessBaseAPI::SetProbabilityInContextFunc(ProbabilityInContextFunc f) {
if (tesseract_ != nullptr) {
tesseract_->getDict().probability_in_context_ = f;
// Set it for the sublangs too.
int num_subs = tesseract_->num_sub_langs();
for (int i = 0; i < num_subs; ++i) {
tesseract_->get_sub_lang(i)->getDict().probability_in_context_ = f;
}
}
}
/** Common code for setting the image. */
bool TessBaseAPI::InternalSetImage() {
if (tesseract_ == nullptr) {
tprintf("Please call Init before attempting to set an image.\n");
return false;
}
if (thresholder_ == nullptr)
thresholder_ = new ImageThresholder;
ClearResults();
return true;
}
/**
* Run the thresholder to make the thresholded image, returned in pix,
* which must not be nullptr. *pix must be initialized to nullptr, or point
* to an existing pixDestroyable Pix.
* The usual argument to Threshold is Tesseract::mutable_pix_binary().
*/
bool TessBaseAPI::Threshold(Pix** pix) {
ASSERT_HOST(pix != nullptr);
if (*pix != nullptr)
pixDestroy(pix);
// Zero resolution messes up the algorithms, so make sure it is credible.
int user_dpi = 0;
GetIntVariable("user_defined_dpi", &user_dpi);
int y_res = thresholder_->GetScaledYResolution();
if (user_dpi && (user_dpi < kMinCredibleResolution ||
user_dpi > kMaxCredibleResolution)) {
tprintf("Warning: User defined image dpi is outside of expected range "
"(%d - %d)!\n",
kMinCredibleResolution, kMaxCredibleResolution);
}
// Always use user defined dpi
if (user_dpi) {
thresholder_->SetSourceYResolution(user_dpi);
} else if (y_res < kMinCredibleResolution ||
y_res > kMaxCredibleResolution) {
tprintf("Warning: Invalid resolution %d dpi. Using %d instead.\n",
y_res, kMinCredibleResolution);
thresholder_->SetSourceYResolution(kMinCredibleResolution);
}
auto pageseg_mode =
static_cast<PageSegMode>(
static_cast<int>(tesseract_->tessedit_pageseg_mode));
if (!thresholder_->ThresholdToPix(pageseg_mode, pix)) return false;
thresholder_->GetImageSizes(&rect_left_, &rect_top_,
&rect_width_, &rect_height_,
&image_width_, &image_height_);
if (!thresholder_->IsBinary()) {
tesseract_->set_pix_thresholds(thresholder_->GetPixRectThresholds());
tesseract_->set_pix_grey(thresholder_->GetPixRectGrey());
} else {
tesseract_->set_pix_thresholds(nullptr);
tesseract_->set_pix_grey(nullptr);
}
// Set the internal resolution that is used for layout parameters from the
// estimated resolution, rather than the image resolution, which may be
// fabricated, but we will use the image resolution, if there is one, to
// report output point sizes.
int estimated_res = ClipToRange(thresholder_->GetScaledEstimatedResolution(),
kMinCredibleResolution,
kMaxCredibleResolution);
if (estimated_res != thresholder_->GetScaledEstimatedResolution()) {
tprintf("Estimated internal resolution %d out of range! "
"Corrected to %d.\n",
thresholder_->GetScaledEstimatedResolution(), estimated_res);
}
tesseract_->set_source_resolution(estimated_res);
return true;
}
/** Find lines from the image making the BLOCK_LIST. */
int TessBaseAPI::FindLines() {
if (thresholder_ == nullptr || thresholder_->IsEmpty()) {
tprintf("Please call SetImage before attempting recognition.\n");
return -1;
}
if (recognition_done_)
ClearResults();
if (!block_list_->empty()) {
return 0;
}
if (tesseract_ == nullptr) {
tesseract_ = new Tesseract;
#ifndef DISABLED_LEGACY_ENGINE
tesseract_->InitAdaptiveClassifier(nullptr);
#endif
}
if (tesseract_->pix_binary() == nullptr &&
!Threshold(tesseract_->mutable_pix_binary())) {
return -1;
}
tesseract_->PrepareForPageseg();
#ifndef DISABLED_LEGACY_ENGINE
if (tesseract_->textord_equation_detect) {
if (equ_detect_ == nullptr && !datapath_.empty()) {
equ_detect_ = new EquationDetect(datapath_.c_str(), nullptr);
}
if (equ_detect_ == nullptr) {
tprintf("Warning: Could not set equation detector\n");
} else {
tesseract_->SetEquationDetect(equ_detect_);
}
}
#endif // ndef DISABLED_LEGACY_ENGINE
Tesseract* osd_tess = osd_tesseract_;
OSResults osr;
if (PSM_OSD_ENABLED(tesseract_->tessedit_pageseg_mode) &&
osd_tess == nullptr) {
if (strcmp(language_.c_str(), "osd") == 0) {
osd_tess = tesseract_;
} else {
osd_tesseract_ = new Tesseract;
TessdataManager mgr(reader_);
if (datapath_.empty()) {
tprintf("Warning: Auto orientation and script detection requested,"
" but data path is undefined\n");
delete osd_tesseract_;
osd_tesseract_ = nullptr;
} else if (osd_tesseract_->init_tesseract(datapath_.c_str(), nullptr,
"osd", OEM_TESSERACT_ONLY,
nullptr, 0, nullptr, nullptr,
false, &mgr) == 0) {
osd_tess = osd_tesseract_;
osd_tesseract_->set_source_resolution(
thresholder_->GetSourceYResolution());
} else {
tprintf("Warning: Auto orientation and script detection requested,"
" but osd language failed to load\n");
delete osd_tesseract_;
osd_tesseract_ = nullptr;
}
}
}
if (tesseract_->SegmentPage(input_file_.c_str(), block_list_, osd_tess, &osr) < 0)
return -1;
// If Devanagari is being recognized, we use different images for page seg
// and for OCR.
tesseract_->PrepareForTessOCR(block_list_, osd_tess, &osr);
return 0;
}
/** Delete the pageres and clear the block list ready for a new page. */
void TessBaseAPI::ClearResults() {
if (tesseract_ != nullptr) {
tesseract_->Clear();
}
delete page_res_;
page_res_ = nullptr;
recognition_done_ = false;
if (block_list_ == nullptr)
block_list_ = new BLOCK_LIST;
else
block_list_->clear();
if (paragraph_models_ != nullptr) {
for (auto model : *paragraph_models_) {
delete model;
}
delete paragraph_models_;
paragraph_models_ = nullptr;
}
}
/**
* Return the length of the output text string, as UTF8, assuming
* liberally two spacing marks after each word (as paragraphs end with two
* newlines), and assuming a single character reject marker for each rejected
* character.
* Also return the number of recognized blobs in blob_count.
*/
int TessBaseAPI::TextLength(int* blob_count) {
if (tesseract_ == nullptr || page_res_ == nullptr)
return 0;
PAGE_RES_IT page_res_it(page_res_);
int total_length = 2;
int total_blobs = 0;
// Iterate over the data structures to extract the recognition result.
for (page_res_it.restart_page(); page_res_it.word () != nullptr;
page_res_it.forward()) {
WERD_RES *word = page_res_it.word();
WERD_CHOICE* choice = word->best_choice;
if (choice != nullptr) {
total_blobs += choice->length() + 2;
total_length += choice->unichar_string().length() + 2;
for (int i = 0; i < word->reject_map.length(); ++i) {
if (word->reject_map[i].rejected())
++total_length;
}
}
}
if (blob_count != nullptr)
*blob_count = total_blobs;
return total_length;
}
#ifndef DISABLED_LEGACY_ENGINE
/**
* Estimates the Orientation And Script of the image.
* Returns true if the image was processed successfully.
*/
bool TessBaseAPI::DetectOS(OSResults* osr) {
if (tesseract_ == nullptr)
return false;
ClearResults();
if (tesseract_->pix_binary() == nullptr &&
!Threshold(tesseract_->mutable_pix_binary())) {
return false;
}
if (input_file_.empty()) {
input_file_ = kInputFile;
}
return orientation_and_script_detection(input_file_.c_str(), osr, tesseract_) > 0;
}
#endif // ndef DISABLED_LEGACY_ENGINE
void TessBaseAPI::set_min_orientation_margin(double margin) {
tesseract_->min_orientation_margin.set_value(margin);
}
/**
* Return text orientation of each block as determined in an earlier page layout
* analysis operation. Orientation is returned as the number of ccw 90-degree
* rotations (in [0..3]) required to make the text in the block upright
* (readable). Note that this may not necessary be the block orientation
* preferred for recognition (such as the case of vertical CJK text).
*
* Also returns whether the text in the block is believed to have vertical
* writing direction (when in an upright page orientation).
*
* The returned array is of length equal to the number of text blocks, which may
* be less than the total number of blocks. The ordering is intended to be
* consistent with GetTextLines().
*/
void TessBaseAPI::GetBlockTextOrientations(int** block_orientation,
bool** vertical_writing) {
delete[] *block_orientation;
*block_orientation = nullptr;
delete[] *vertical_writing;
*vertical_writing = nullptr;
BLOCK_IT block_it(block_list_);
block_it.move_to_first();
int num_blocks = 0;
for (block_it.mark_cycle_pt(); !block_it.cycled_list(); block_it.forward()) {
if (!block_it.data()->pdblk.poly_block()->IsText()) {
continue;
}
++num_blocks;
}
if (!num_blocks) {
tprintf("WARNING: Found no blocks\n");
return;
}
*block_orientation = new int[num_blocks];
*vertical_writing = new bool[num_blocks];
block_it.move_to_first();
int i = 0;
for (block_it.mark_cycle_pt(); !block_it.cycled_list();
block_it.forward()) {
if (!block_it.data()->pdblk.poly_block()->IsText()) {
continue;
}
FCOORD re_rotation = block_it.data()->re_rotation();
float re_theta = re_rotation.angle();
FCOORD classify_rotation = block_it.data()->classify_rotation();
float classify_theta = classify_rotation.angle();
double rot_theta = - (re_theta - classify_theta) * 2.0 / M_PI;
if (rot_theta < 0) rot_theta += 4;
int num_rotations = static_cast<int>(rot_theta + 0.5);
(*block_orientation)[i] = num_rotations;
// The classify_rotation is non-zero only if the text has vertical
// writing direction.
(*vertical_writing)[i] = classify_rotation.y() != 0.0f;
++i;
}
}
void TessBaseAPI::DetectParagraphs(bool after_text_recognition) {
int debug_level = 0;
GetIntVariable("paragraph_debug_level", &debug_level);
if (paragraph_models_ == nullptr)
paragraph_models_ = new std::vector<ParagraphModel*>;
MutableIterator *result_it = GetMutableIterator();
do { // Detect paragraphs for this block
std::vector<ParagraphModel *> models;
::tesseract::DetectParagraphs(debug_level, after_text_recognition,
result_it, &models);
paragraph_models_->insert(paragraph_models_->end(), models.begin(), models.end());
} while (result_it->Next(RIL_BLOCK));
delete result_it;
}
/** This method returns the string form of the specified unichar. */
const char* TessBaseAPI::GetUnichar(int unichar_id) {
return tesseract_->unicharset.id_to_unichar(unichar_id);
}
/** Return the pointer to the i-th dawg loaded into tesseract_ object. */
const Dawg *TessBaseAPI::GetDawg(int i) const {
if (tesseract_ == nullptr || i >= NumDawgs()) return nullptr;
return tesseract_->getDict().GetDawg(i);
}
/** Return the number of dawgs loaded into tesseract_ object. */
int TessBaseAPI::NumDawgs() const {
return tesseract_ == nullptr ? 0 : tesseract_->getDict().NumDawgs();
}
/** Escape a char string - remove <>&"' with HTML codes. */
std::string HOcrEscape(const char* text) {
std::string ret;
const char *ptr;
for (ptr = text; *ptr; ptr++) {
switch (*ptr) {
case '<': ret += "<"; break;
case '>': ret += ">"; break;
case '&': ret += "&"; break;
case '"': ret += """; break;
case '\'': ret += "'"; break;
default: ret += *ptr;
}
}
return ret;
}
} // namespace tesseract.
|