Hello, please I need help with this, I am using Microsoft OCR, scale 2 and scan as Profile. it recognize the other numbers but zero is assigned as an “O” letter, so I cannot work with that numeric number. What should I do? I would appreciate your help with this
- Increase the scale.
- Try another engine.
If you are only looking for numbers can try to allow only digits to be recognized from the options of one of the engines.
In Google OCR engine there is option to allow only specific characters.
Also i guess without AllowedCharacters itself you will be able to get the numbers correctly.
Nope, still not working even I followed all your suggestions
If you only need numeric data then replace “O” letters with zeros in OCR output.
Try with Google Tesseract OCR and follow below steps:
- Maximum correct information you’ll able to get within a scale of 2-4.
- You’ll be having options to restrict getOCRText method to various options like numbers only, alphabets only, custom also etc.
- Try with Screen OCR using scale between 2-4.
At last, if above points won’t work for you, then try to observe the pattern of wrong extracted data and make it correct by doing string manipulations. For example: if actual value is MSG1 and OCR is reading as MSGI, change it by using regex operations.
I’d done one project on OCR, I’ll also be sharing my findings with the community soon so that it helps others too.