an application to automatically extract text from comic books.
...It is based on the following 3 major algorithms
- Binarization of color images (Niblak and other methods)
- Connected components
- K-Means clustering
Apache Tesseract is used to perform Optical Character Recognition on the extracted text.
A subsequent version of the application will integrate with translation software in order to provide automated translation of comic book texts and re-inserion of translated texts
Java OCR is a suite of pure java libraries for image processing and character recognition. Small memory footprint and lack of external dependencies makes it suitable for android development. Provides modular structure for easier deployment
ContraCaptcha is a Java-based libary for solving (image-based) captchas. It provides classes for image processing and character recognition. It also includes solvers for some of the popular captcha-systems in use DISCLAIMER: only for learning purpose (;