How can I implement OCR on a website using PHP?

Question

Are there any free OCR libraries that work with PHP or Python on a Linux server? The idea is to be able to load an image and pull out symbols from it, or allow users to “draw symbols” and analyze them from the specified image.

Answer 1

Since you're in a Linux box, I would highly recommend the Google open source project ocropus .

This is not PHP, but I think it will be your best option. Of course, you can call it from PHP via exec . Its mature and has many options. On the project website:

The OCRopus engine is based on two research projects: a high-performance handwriting recognition engine, developed in the mid-90s and deployed by the US Census Bureau, and new high-performance layout analysis methods.

There is another open source project, tesseract . I have used this in the past, and was pleased with the results. Includes training, limiting your alphabet, etc.

Answer 2

Have you seen the phpOCR lessons of Andrey Kucherenko? http://www.phpclasses.org/package/2874-PHP-Recognize-text-objects-in-graphical-images.html This is an old article, but may help you.