Taking a photo of a physical (printed) book page often results in an image with curved text. This makes text recognition much more unreliable.
The goal of this project is to build a routine that can recognize the curvature of the book page image and apply the unwarping (eg. CIWarpKernel) based on that information. The output of the routine would be an unwarped image with straightened text lines.
6 freelanceria on tarjonnut keskimäärin $1079 tähän työhön
Hi I have read your description carefully. I have rich expereince with image processing and OCR. I think I can help you with my skill and expereince Hope your kind contacts. thanks.