CODEWITHSUNDEEP
opencvtesseractpython-tesseract
Fran
Fran

Reputation: 13

<lang>.traineddata not found ( even when it's in the correct folder)

I'm using tesseract to detect text in spanish in some screenshot of a game, I had some issues with the "spa.traineddata" so I started to train my own data called "spa1.traineddata" and I used the two files to make text detection more accurate, yesterday I make some tests and seemed to work well, but the file "spa1.traineddata" needed more training, so I decided to continue today, I added some new images to train my file "spa1.trainneddata" and then I wanted to test it and throws me the following error

pytesseract.pytesseract.TesseractError: (1, 'Error opening data file C:\\Program Files\\Tesseract-OCR/tessdata/-l spa.traineddata Please make sure the TESSDATA_PREFIX environment variable is set to your "tessdata" directory. Failed loading language \'-l spa\' Error opening data file /home/debian/src/github/tesseract-ocr/tesseract/bin/ndebug/x86_64-w64-mingw32-5.0.0-alpha.20200223/usr/x86_64-w64-mingw32/share/tessdata/spa1 --psm 
6.traineddata Please make sure the TESSDATA_PREFIX environment variable is set to your "tessdata" directory. Failed loading language \'spa1 --psm 6\' Tesseract couldn\'t load any languages! Could not initialize tesseract.')

and these are the lines of code that I use for testing 

custom_config = r'-l spa+spa1 --psm 6'
pytesseract.image_to_string(Image.open('imagenes/obv.png'), lang=custom_config)

I was searching and found that the error was because the lang.traideddata files were not in the tessdata folder, but mine are in the folder (That's why I was able to work with them yesterday)

I attach a screenshot of the tessdata folder, the last two files are the traineddata tessdata folder

also, it is useful I'm using vscode, python 3.7 and tesseract 4

I hope you can help me (sorry my bad english uwu)

Upvotes: 1

Views: 301

Answers (1)

Nikhil
Nikhil

Reputation: 1

I encountered same issue while working with Devanagari model. I realised later that the model I downloaded was wrong. Download your model from https://github.com/tesseract-ocr/tessdata/tree/main/script and put it in the tessdata folder. Worked fine for me after that.

Upvotes: 0

Related Questions