1

I'm looking for a C# library that allows me to add the following to my .net application:

  • Scan image
  • Load image and pick extraction template
  • Use OCR or Machine Learning to extract text from the image based on my template
  • Save the results

Please recommend a library that meets this criteria

0

You may want to look at the Leadtools Forms SDK (https://www.leadtools.com/sdk/products/forms). It offers exactly that and is available for .NET or Java.

  • You can create a masterform template to designate predefined areas for barcodes, text, or OMR that you want to extract.
  • You can then scan in or load images to match against a set of masterforms and determine which template a form belongs to.
  • Once the masterform is identified, you can just extract those fields automatically for any kind of data analysis you’re doing.

This does not require lengthy machine learning training and the OCR has fairly good accuracy even before fine tuning, although you can also use their image processing libraries to clean up the document if there’s a lot of noise or low contrast.

They have this available through NuGet as well: https://www.nuget.org/packages/Leadtools.Document.Sdk/

And they have a short simple tutorial for this too: https://www.leadtools.com/help/sdk/v21/tutorials/forms/windows/dotnet/console/auto-recognize-and-process-a-form.html

DISCLAIMER: I work at LEAD Technologies who develops this toolset

Masterform creation:

Creating the Masterform template

Results:

Extracting the data from the filled form

2

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.