Parameters 从png/box创建traineddata文件时出现警告:未找到参数->;tessedit\u单匹配和il1\u自适应测试
我正在尝试从使用jTessBoxEditor创建的png/box集合中创建一个具有特定test_font.traineddata文件的新培训文件 执行该行时:Parameters 从png/box创建traineddata文件时出现警告:未找到参数->;tessedit\u单匹配和il1\u自适应测试,parameters,warnings,tesseract,Parameters,Warnings,Tesseract,我正在尝试从使用jTessBoxEditor创建的png/box集合中创建一个具有特定test_font.traineddata文件的新培训文件 执行该行时: tesseract %PATH%/test_font.test_font.exp0.png %PATH%/test_font.test_font.exp0 nobatch box.train 我收到警告信息: Warning: Parameter not found: tessedit_single_match Warning: Par
tesseract %PATH%/test_font.test_font.exp0.png %PATH%/test_font.test_font.exp0 nobatch box.train
我收到警告信息:
Warning: Parameter not found: tessedit_single_match
Warning: Parameter not found: il1_adaption_test
Tesseract Open Source OCR Engine v5.0.0-alpha.20191030 with Leptonica
APPLY_BOXES:
Boxes read from boxfile: 192
Found 192 good blobs.
Leaving 6 unlabelled blobs in 0 words.
Generated training data for 46 words
Warning: No shape table file present: shapetable
Reading %PATH%/test_font.test_font.exp0.tr ...
Flat shape table summary: Number of shapes = 60 max unichars = 1 number with multiple unichars = 0
Warning: no protos / configs for Joined in CreateIntTemplates ()
Warning: no protos / configs for | Broken | 0 | 1 in CreateIntTemplates ()
Done!
这似乎不是灾难性的,因为我仍然能够在之后生成unicharset文件。
但我想知道这些警告是从哪里来的,如果我能纠正它们,它们的影响是什么
之后,在执行命令期间:
mftraining -F font_properties -U unicharset -O %PATH%/test_font.unicharset %PATH%/test_font.test_font.exp0.tr
我得到的信息是:
Warning: Parameter not found: tessedit_single_match
Warning: Parameter not found: il1_adaption_test
Tesseract Open Source OCR Engine v5.0.0-alpha.20191030 with Leptonica
APPLY_BOXES:
Boxes read from boxfile: 192
Found 192 good blobs.
Leaving 6 unlabelled blobs in 0 words.
Generated training data for 46 words
Warning: No shape table file present: shapetable
Reading %PATH%/test_font.test_font.exp0.tr ...
Flat shape table summary: Number of shapes = 60 max unichars = 1 number with multiple unichars = 0
Warning: no protos / configs for Joined in CreateIntTemplates ()
Warning: no protos / configs for | Broken | 0 | 1 in CreateIntTemplates ()
Done!
您知道这些新警告的影响是什么,以及如何修复它们吗
提前非常感谢 我刚刚意识到这些参数Tesseract OCR\TesserData\configs\box.train和Tesseract OCR\TesserData\configs\lstm.train文件中的错误命令行无意中添加了Tesseract\u single\u match和il1\u adaption\u test 这些文件的正确结构似乎分别为:
disable_character_fragments T
file_type .bl
textord_fast_pitch_test T
tessedit_zero_rejection T
tessedit_minimal_rejection F
tessedit_write_rep_codes F
edges_children_fix F
edges_childarea 0.65
edges_boxarea 0.9
tessedit_resegment_from_boxes T
tessedit_train_from_boxes T
textord_no_rejects T
及
这似乎解决了第一个警告问题,但没有解决no protos/configs