Read PDF with OCT don't always return the same text

So I’m trying to read a PDF and extract a name of a person, the name is ‘marijose’, with a scale of 2, I was getting ‘mac]’, with a scale of 15, I’m getting ‘marljose’ and sometime ‘manjose’

  • there multiple name, but so far, I only get false result with this 1.

Module: Uipath.PDF.Activities.ReadPDFWithOct
Engine: Tesseract OCR (UiPath.Core.Activities.GoogleOCR)

would there be a better ways to get the exact text?

Yes, I have the same issue as the Microsoft OCR and Google OCR are not efficient enough to read the handwritten text and screenshots.

I used python tesseract and I got the best results. I hope python tesseract is efficient enough to get the same text everytime…