hello shree i am confused. can you explain precisely what i must to do? i perform a fine tune on fas.traineddata and after training fas.lstm, fas.lstm-number-dawg, fas.lstm-punc-dawg, fas.lstm-recoder and etc. are achieved. now when i copy fas.traineddata to tessdata folder an execute tesseract command, "Failed to load any lstm-specific dictionaries" appears and ocr output is not good. best regards
On Monday, 29 October 2018 04:59:23 UTC+3:30, shree wrote: > > The starter traineddata that you have used does not have any dawg files, > based on word list, numbers and punctuation, hence the report that > dictionaries are not found. > > > > On Fri, 26 Oct 2018, 14:38 Abu Anas, <abus...@gmail.com <javascript:>> > wrote: > >> I am also having similar problem. I have trained KB-JT-NEW from ben >> (continue from) and found the result: >> >> At iteration 127102/500000/500000, Mean rms=0.437%, delta=1.593%, char >> train=11.184%, word train=11.098%, skip ratio=0%, New worst char error = >> 11.184 wrote checkpoint. >> >> Finished! Error rate = 7.737 >> lstmtraining \ >> --stop_training \ >> --convert_to_int \ >> --continue_from data/checkpoints/KB-JT-NEW_checkpoint \ >> --traineddata data/KB-JT-NEW/KB-JT-NEW.traineddata \ >> --model_output data/KB-JT-NEW.traineddata >> Loaded file data/checkpoints/KB-JT-NEW_checkpoint, unpacking... >> >> But putting the .traineddata on /usr/local/share/tessdata/ and executing >> recognition gives bizarre result and showing: >> Failed to load any lstm-specific dictionaries for lang KB-JT-NEW!! >> >> >> >> >> On Tuesday, March 27, 2018 at 1:45:55 PM UTC+6, bha...@automot.us wrote: >>> >>> Hello, >>> >>> I am working on a project where I extract and the license plates and try >>> to get the plate number automatically. >>> >>> After applying some computer vision and image processing, I have come up >>> with the following result. >>> >>> <https://lh3.googleusercontent.com/-g6CxXgDj3ck/Wrl-DoT61lI/AAAAAAAAACs/hcE9t8zmQJk_almmkwFZJYXcSuB717OywCLcBGAs/s1600/topcrop1.jpg> >>> >>> >>> As it can be very obvious, the OCR generated with tesseract is: 6JZX97L >>> >>> Where as, actually, it is 6JZX974. >>> >>> I a, very new to the tesseract and it seemed like a very easy to use >>> library for my task, however, I do not have any idea on how to tackle a >>> scenario like this. If there is anyone who has worked on solving such a >>> problem, please share thoughts. >>> >>> Some other error prone numbers/letters: 0-O, 1-I, 2-Z, 5-S, 8-B... >>> >>> Thanks! >>> >> -- >> You received this message because you are subscribed to the Google Groups >> "tesseract-ocr" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to tesser...@googlegroups.com <javascript:>. >> To post to this group, send email to tesser...@googlegroups.com >> <javascript:>. >> Visit this group at https://groups.google.com/group/tesseract-ocr. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/tesseract-ocr/3977592d-2a2d-47dc-a48e-4dc98b41545b%40googlegroups.com >> >> <https://groups.google.com/d/msgid/tesseract-ocr/3977592d-2a2d-47dc-a48e-4dc98b41545b%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> For more options, visit https://groups.google.com/d/optout. >> > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscr...@googlegroups.com. To post to this group, send email to tesseract-ocr@googlegroups.com. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/9e5abfb0-f565-44bc-9acf-649f6114fbec%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
