Re: [tesseract-ocr] add new characters

2020-10-28 Thread Timo Struppi
New and fresh Linuxinstallation i ment :) On Wednesday, October 28, 2020 at 9:50:16 PM UTC+1 Timo Struppi wrote: > Hello, Problem solved. > > I just made a Linuxinstallation and the error was gone. > > Thanks again for you file and help! > > > On Wednesday, October 28, 2020 at 3:06:18 AM UTC+1 shr

Re: [tesseract-ocr] add new characters

2020-10-28 Thread Timo Struppi
Hello, Problem solved. I just made a Linuxinstallation and the error was gone. Thanks again for you file and help! On Wednesday, October 28, 2020 at 3:06:18 AM UTC+1 shree wrote: > Did you copy the traineddata file to > /usr/share/tesseract-ocr/4.00/tessdata? > What's the value of TESSDATA_PR

Re: [tesseract-ocr] add new characters

2020-10-27 Thread shree
Did you copy the traineddata file to /usr/share/tesseract-ocr/4.00/tessdata? What's the value of TESSDATA_PREFIX in your 'env' output? What's the output of? ls -l /usr/share/tesseract-ocr/4.00/tessdata/Sanskrit-1017-fast.traineddata combine_tessdata -d /usr/share/tesseract-ocr/4.00/tessdat

Re: [tesseract-ocr] add new characters

2020-10-27 Thread Timo Struppi
Help! I get following errorcode. What am i doing wrong? Error opening data file /usr/share/tesseract-ocr/4.00/tessdata/Sanskrit-1017-fast.traineddata Please make sure the TESSDATA_PREFIX environment variable is set to your "tessdata" directory. Failed loading language 'Sanskrit-1017-fast' Tesse

Re: [tesseract-ocr] add new characters

2020-10-24 Thread Timo Struppi
*perfect!* Thank you very much <3 Thats what i was looking for. International Alphabet of Sanskrit Transliteration Characters. Can tell me in which folder i must place the .traineddata? My configuration: tesseract 4.1.1 leptonica-1.79.0 libgif 5.1.4 : libjpeg 8d (libjpeg-turbo 2.0.3) : libp

Re: [tesseract-ocr] add new characters

2020-10-24 Thread Shree Devi Kumar
Ray has suggested using plus-minus type of training for adding a couple of characters to the traineddata. Did you try that? Please share the training data you used (box/tiff pairs or lstmf files). I have done replace a layer training for Sanskrit. It adds the two characters you want (in addition

[tesseract-ocr] add new characters

2020-10-24 Thread Timo Struppi
Hello, I dont want to invent the wheel new by creating a new language but how do i add the letters ṛ and ī to the OCR?? I tried a lot (vietOCR, Linux inteligent OCR solution, followed the few avaible tutorials etc) for several days but i am still not achieve to add a single letter. Many th