OCR text exist for different colors


I’m using OCR text exist to recognise numbers in a .pdf file, which works most of the time but sometimes the number is in a different color (red in this case) but still clearly visible and it won’t recognise the number. I have tried playing around with the accuracy but with no succes. Any ideas?


Hi @koolenc,
You can also try to work with Tesseract OCR engine and property “invert” which can help with such cases.


What engine are you using currently?

I’m using Tesseract OCR and I tried invert but it’s still not working… Do you maybe have any other suggestions? Thanks!