ocr - Tesseract with limited words -
is possible recognize limited set of words in tesseract?
i need recognize set of words (around 200) , want tesseract correct words closest matching ones. in order that, i've updated language models words (eng.word-dawg , eng.freq-dawg) , increased sensitivity setting language_model_penalty_non_freq_dict_word , language_model_penalty_non_dict_word large numbers (tried 0.9 , 1.0). however, not have affect on output.
i have word (benzoate) tesseract recognize uenzoate. weird have benzoate in dictionary.
Comments
Post a Comment