Handwritten scanned Doc to .txt File?

问题:

Are there any JAVA APIs or tools that can convert Handwritten Scanned Doc to txt files?

I have tried google tesseract and few other tools , but I am not getting satisfactory results for hand written scanned docs.

回答1:

Strange that other answers here are pointing out to OCR tools while question clearly states handwriting recongition.

Handwriting is even more difficult area than OCR and number of technologies available is very narrow. I don't think you will be able to find any open source tool for that, while there are few commertial vendors:

• http://www.a2ia.com
• http://www.parascript.com/

I don't know if they have Java API, but it is better to start researching from contacting them.

回答2:

You can try the Java OCR Project. I think that you might do the writing to a text file section yourself though.

Also, hand writing tends to vary from one individual to another, so I guess you will need to select some good training data to get good results.

回答3:

Have a look at these :

1. Java OCR

   • Java OCR is a suite of pure java libraries for image processing and character recognition. Provides modular structure for easier deployment .

2. GOCR

   • GOCR is an OCR program, developed under the GNU Public License. It converts scanned images of text back to text files.