OSS-DocumentScanner
OSS-DocumentScanner copied to clipboard
OCR text positioning is not accurate
actually I think the blocks are correct. what ils more difficult is determining the font and the font size. and I think this is why you say it is not accurate. for now I dont see a better way to do it. and except with machine learning I dont think others are doing it better. I might be wrong though! dont hesitate to say if I am wrong
this app uses tesseract. you can look it up
To be honest I don't know anything about app development. I'm just putting the problems in the app what I see and you describe why the problem is there to me so I understand that.
Hope your project goes a long way this app has potential.
@code3AGL3 totally understand and i ll keep this issue open. I hope i can improve it