Create images of words from text in an image

Question

Does anyone know of any libraries (preferably Java, but I would look at anything) that would allow me to break apart text in an image and create smaller images for each word? I have tested GOCR and tesseract, but I'm not looking to transcribe the text and in many cases, such as handwritten notes or poorly functioning typewriters, the text has to be transcribed by a human anyways. The fact that most of the text is in Spanish also makes OCR more difficult. I've searched the stackoverflow archives and not found anything. The purpose of my question is at the university we have hundreds of thousands of really old handwritten notes, and if I could break them up word by word into smaller images then we could try transcribing and translating them.

Answer 1

http://sourceforge.net/projects/javaocr/

In particular, the screenshot below seems to be related to what you're after.