Description
I don't know of any decent open source pure Java OCR libraries, but there are command line OCR tools like Tesseract (http://code.google.com/p/tesseract-ocr/) that could be invoked by Tika to extract text content (where available) from image files.