Java simple OCR algorithm

Question

I've been working on a simple OCR project for a couple days. The app is supposed to extract a text from an image. The solution I've come up with is: greyscaling, rotating, removing noise from the image and isolating every single character on the image. So I need some help with a simple algorithm that would let me recognise the character. I only need to recognise the letters A,B,C,D.

What sorts of ideas do you have about making the algorithm? Have you written anything to implement it so far? — APerson, Jul 01 '15 at 18:33
see [OCR and character similarity](http://stackoverflow.com/a/22879053/2521214) `Pixel approach` is a simple deterministic approach I develop for a commercial project few years back. It does not use neural networks so it is free of their issues ... which makes it ideal for a rookie to start with — Spektre, Jul 02 '15 at 05:36

score 0 · Answer 1 · answered Jul 01 '15 at 18:39

0

The method i'd use would be to make a mask of each letter, scaling it to the rectangle you found when cutting the picture and "score" each letter, by comparing RGB for example.

answered Jul 01 '15 at 18:39

Pâris Douady

833
8
18

Seems ok, but it will make it limited to recognising only some fonts. – Atanas Ivanov Jul 01 '15 at 18:46
Yeah sure, but i think by testing 2 or 3 fonts, you will cover 99% of all text. Also, different fonts looks like each other. – Pâris Douady Jul 01 '15 at 18:47
I would give it a try. – Atanas Ivanov Jul 01 '15 at 19:11

Java simple OCR algorithm

1 Answers1