mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2024-12-13 16:09:04 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

Don't create OCR result files when training data is created

Browse Source 
The configuration file lstm.train causes Tesseract to generate
training data for training of an LSTM line recognizer.

In this mode, no other files with OCR results should be written.
Without this patch, Tesseract writes a small text file.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
This commit is contained in:
Stefan Weil 2019-10-02 19:10:23 +02:00 committed by zdenop
parent 3dfd72721b
commit 58122ea313
1 changed files with 8 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
src/api/tesseractmain.cpp
View File

@ -565,6 +565,9 @@ static void PreloadRenderers(
api->GetBoolVariable("tessedit_create_txt", &b);
if (b || (!error && renderers->empty())) {
// Create text output if no other output was requested
// even if text output was not explicitly requested unless
// there was an error.
auto* renderer =
new tesseract::TessTextRenderer(outputbase);
if (renderer->happy()) {
@ -716,13 +719,15 @@ int main(int argc, char** argv) {
return ret_val;
}
// set in_training_mode to true when using one of these configs:
// ambigs.train, box.train, box.train.stderr, linebox, rebox
// Set in_training_mode to true when using one of these configs:
// ambigs.train, box.train, box.train.stderr, linebox, rebox, lstm.train.
// In this mode no other OCR result files are written.
bool b = false;
bool in_training_mode =
(api.GetBoolVariable("tessedit_ambigs_training", &b) && b) ||
(api.GetBoolVariable("tessedit_resegment_from_boxes", &b) && b) ||
(api.GetBoolVariable("tessedit_make_boxes_from_boxes", &b) && b);
(api.GetBoolVariable("tessedit_make_boxes_from_boxes", &b) && b) ||
(api.GetBoolVariable("tessedit_train_line_recognizer", &b) && b);
#ifdef DISABLED_LEGACY_ENGINE
auto cur_psm = api.GetPageSegMode();