Need some advices to learn OCR related techniques

问题:

I'm working on an OCR project for iPhone using tesseract OCR engine. I'm planning to write the following modules:

• Capture image from iPhone camera
• Pre-process on the image to refine it, in order to improve the OCR output.
• Divide the OCR output into meaningful fields.
• Define some rules for the OCR engine in order to neglect any undefined characters. (e.g. if the the OCR output is 0226s5242 I want it to ignore the s character)

I want to begin learning the topics related to these modules, I'm not aware of the OCR related techniques, so any advice will be very helpful, Thanks.

回答1:

Then start with compiling tesseract for iphone.

This link may help you.

Also try below question:

How to cross compile tesseract ocr engine for iphone?

Then try some OCR demo : Try it here

And search in SOF, a lot of related links are here:

implement tesseract OCR in iphone

iPhone: How to use Tesseract

Tesseract Example for Iphone

etc are some of them..