Click here to Skip to main content
15,891,316 members
Please Sign up or sign in to vote.
3.00/5 (1 vote)
See more:
I am working on project where
We verify scanned documents by using OCR tool.But for this every documents processed manually so it takes too much time.I want to do this by programmatically.
I looked for abbey cloud trial version tool but cant extract perfect data from documents in which handwritten data is not getting properly.So I want to know that
1. Which OCR tool(Free/Pad) is better which reads handwritten as well as other text from document images.
2. It should read table format data also.
3. All this should be controlled by programming so that process time can be reduced.
4. Document may contain images also.


Thanks in Advance
Posted
Comments
Sinisa Hajnal 4-Nov-15 2:42am    
You will never get 100% coverage, what you can is use the library to get most documents acceptable and flag some as unclear or simply enable some human checking via doc by doc imageview...The problem then is of course getting correct info on the scanned document.
sp_suresh 4-Nov-15 4:42am    
THANKS for reply.I know that i will never get 100% coverage but can we achieve 60-80% so that remaining percentage will be verified by human?If yes then we can reduce lot of time for this where documents in large scale verified .
Thanks
Sinisa Hajnal 4-Nov-15 6:10am    
I have some experience with IBM Datacap...it holds rules and templates, boasts web interface to get to the documents that need to be filled and keeps the data next to the image. But it is far from free...so apologies. My comment was targeted at "but cant extract perfect data from documents in which handwritten data"...you need to check several and find out which works acceptably well or has heuristic algorithm that will learn to recognize specific handwriting in time. Good luck
sp_suresh 4-Nov-15 6:16am    
Thanks a Lot Sinisa . Now i have clear picture.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)



CodeProject, 20 Bay Street, 11th Floor Toronto, Ontario, Canada M5J 2N8 +1 (416) 849-8900