Performing complex image analysis with python / django

Question

I am working on a django project that parses images containing text and (1) if the image needs to be rotated, and (2) where the text areas are.

I am currently using PIL to more easily process these images, but I'm not quite sure how I can use PIL or other libraries to accomplish both tasks. I was wondering if someone had done this before, and if there are libraries / api that can help with development.

Answer 1

OpenCV is probably the popular open source image processing library. This is C / C ++, but there are python bindings:

http://opencv.willowgarage.com/wiki/

and python docs

http://opencv.willowgarage.com/documentation/python/index.html

I have never done OCR with him, but I'm sure he is capable

Answer 2

I agree with @pastylegs that OpenCV is your best initial bet. If you need OCR-specific material, you can also look at ocropus .