MCPcopy Create free account
hub / github.com/creatale/node-dv / ReSegmentByClassification

Method ReSegmentByClassification

deps/tesseract/ccmain/applybox.cpp:509–531  ·  view source on GitHub ↗

Resegments the words by running the classifier in an attempt to find the correct segmentation that produces the required string.

Source from the content-addressed store, hash-verified

507/// Resegments the words by running the classifier in an attempt to find the
508/// correct segmentation that produces the required string.
509void Tesseract::ReSegmentByClassification(PAGE_RES* page_res) {
510 PAGE_RES_IT pr_it(page_res);
511 WERD_RES* word_res;
512 for (; (word_res = pr_it.word()) != NULL; pr_it.forward()) {
513 WERD* word = word_res->word;
514 if (word->text() == NULL || word->text()[0] == '\0')
515 continue; // Ignore words that have no text.
516 // Convert the correct text to a vector of UNICHAR_ID
517 GenericVector<UNICHAR_ID> target_text;
518 if (!ConvertStringToUnichars(word->text(), &target_text)) {
519 tprintf("APPLY_BOX: FAILURE: can't find class_id for '%s'\n",
520 word->text());
521 pr_it.DeleteCurrentWord();
522 continue;
523 }
524 if (!FindSegmentation(target_text, word_res)) {
525 tprintf("APPLY_BOX: FAILURE: can't find segmentation for '%s'\n",
526 word->text());
527 pr_it.DeleteCurrentWord();
528 continue;
529 }
530 }
531}
532
533/// Converts the space-delimited string of utf8 text to a vector of UNICHAR_ID.
534/// @return false if an invalid UNICHAR_ID is encountered.

Callers 1

AdaptToWordStrMethod · 0.80

Calls 4

textMethod · 0.80
DeleteCurrentWordMethod · 0.80
wordMethod · 0.45
forwardMethod · 0.45

Tested by

no test coverage detected