Tools for Data Capturing from Scanned Images

940 Views Asked by At

What are best tools available for Data Capturing from Scanned Images? Doesn't matter they are free or paid. They should be able to convert scanned images to data (text, image, tables etc.). Most importantly they should have API to integrate with .NET 2.0 for automation.

Edit: The input data will generally be Text, Images, Tables places in the image. These images will basically be scanned pages from Catalogs containing data.

8

There are 8 best solutions below

0
On

What about IBM DataCap,EMC-Captiva inputaccel?

0
On

Strange, but most of the answers here are about OCR libraries, regardless question itself stated Data Capture as main task.

There is number of Data Capture application on the market, but when it comes to SDK, I can only recommend ABBYY FlexiCapture Engine: http://abbyy.com/flexicapture_engine/

It provides you with very powerful visual tool to define field extraction logic based on different rules, test it on batch and then use this descriptuion in SDK linked with your application. This field extraction technology is little bit complex at the beginning, but can cope with quite difficult things like tables split on several pages and so on. Disclaimer: I work for ABBYY

0
On

AForge has avlot of useful machine vision stuff.

0
On

I recommend that you take a look into ReadIris Pro, which claims to have Automated Document Processing. There's a good chance that it has some form of API, although I have not actually verified that.

I do know that it is happy with text, tables, and handprinting (handwriting in all caps), and as an added bonus can recognise barcodes.

It does cost, though, but it not prohibitively expensive.

0
On
0
On

You can use the people in Amazon's Mechanical Turk, you can program the people to do whatever you want, you can look at their API documentation for more info.

0
On

Here are some example products:

Aspire

Ocrtools

All these SourceForge .NET OCR's

0
On

ABBYY FineReader OCR API conforms to COM and can be used in C/C++, Visual Basic, .Net, Delphi, etc.