0

For example: there is an image, and I have an article in the image. And how I can convert that article to a string? Can you tell me or give me a solution?

dda
  • 6,030
  • 2
  • 25
  • 34
qakmak
  • 1,287
  • 9
  • 31
  • 62
  • 10
    You are probably looking for some [OCR stuff](http://en.wikipedia.org/wiki/Optical_character_recognition). [Here is a list of libraries](http://en.wikipedia.org/wiki/List_of_optical_character_recognition_software), including some C# ones, too. – Uwe Keim Jan 10 '13 at 08:18
  • Try read about [Artificial neural network](http://en.wikipedia.org/wiki/Artificial_neural_network), [Applying Neural Networks to Character Recognition](http://www.ccs.neu.edu/home/feneric/charrecnn.html) – Alexander Balte Jan 10 '13 at 08:22
  • You should have a look at optical character recognition articles : [here][1] [1]: http://stackoverflow.com/questions/4913373/read-textdata-in-an-images-using-c-sharp – Melvin Jan 10 '13 at 08:22
  • 1
    see http://stackoverflow.com/questions/5933090/is-there-any-free-ocr-api-for-net/5933191 – Ria Jan 10 '13 at 08:22

2 Answers2

12

You would need an Optical Character Reader. Google have Tesseract.NET which should give you an OCR for .NET.

Note however that OCR's have limitations. Text with varied font and font size can cause incorrect results from the OCR.

Pre-processing usually helps, like for instance if you text will always be within a certain segment of image, you can just feed that sub-section to the OCR. Removing colours is usually also a good idea.

npinti
  • 51,780
  • 5
  • 72
  • 96
4

What you're looking for is highly likely Optical Character Recognition (OCR), which will allow you to transform characters in an image to their textual representation.

It seems that the most common way to solve your problem with C# would be to use Tessnet, which is a .NET wrapper around Google's native Tesseract OCR library.

For more information you can always reference these related questions:

OCR with the Tesseract interface

https://stackoverflow.com/questions/744966/any-open-source-c-sharp-ocr-library

Community
  • 1
  • 1
Rudi Visser
  • 21,350
  • 5
  • 71
  • 97