Click here to Skip to main content
Rate this: bad
good
Please Sign up or sign in to vote.
Hi Everyone,
I want to process a image an recognize the text present in it. I kindly request you to suggest me how to do it?
When i searched Google, I saw post saying that it can be done using EmguCV and AForge.Net. Is it so? Is there any other framework or built in libraries for effectively doing it?
 
EDIT: I like to code in C#..
 
Regards,
Siva7891
Posted 10-Oct-11 5:32am
Edited 11-Oct-11 6:24am
v3

1 solution

Rate this: bad
good
Please Sign up or sign in to vote.

Solution 1

Try thes links as they are good articles to get you started:
 
Contour Analysis for Image Recognition in C#[^]
Image Recognition with Neural Networks[^]
 
As for EMGU download it from sourceforge.net Here.[^] Have a look in the extracted folder and there will be several examples "Emgu.CV.Example", look for LicensePlateRecognition. This will show you how to do Optical Character Recognition and will require less work on your side of things.
 
Use this article as reference to start your a new project or if you encounter errors : Creating Your First EMGU Image Processing Project[^]
 
Now "I want to process a image and recognize the text present in it" is a bit vague. You will notice that the text in the license plate example is located via pre-processing the image. This is fairly easy as the plates are rectangular and white or yellow in colour. You will have to be responsible for pre-processing the image and finding the rough area that the text is in.
 
You don't mention which programming language you would prefer if it is c# then EMGU and Aforge are the best options for you. Gather a few images that you wish to process have a look at the examples and see what results you can achieve. If you need help along the way come back and ask or try the EMGU forums as well.
 
Cheers
  Permalink  
v2
Comments
SAKryukov at 10-Oct-11 17:59pm
   
They look like interesting works, my 5.
--SA
C_Johnson at 11-Oct-11 5:24am
   
Hi, Hope you don't mind I updated you answer to make it a little more comprehensive cheers, Chris
Abhinav S at 11-Oct-11 8:33am
   
You might as well have entered a different answer.You would have had a 5 from me for that answer for sure.
:)
C_Johnson at 11-Oct-11 9:25am
   
lol cheers, but I always think it's better to have 1 comprehensive answer then several diluted ones. Just happy you didn't mind me editing yours and your links were to 2 great articles. :)

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

  Print Answers RSS
0 OriginalGriff 7,903
1 Sergey Alexandrovich Kryukov 7,142
2 DamithSL 5,604
3 Manas Bhardwaj 4,986
4 Maciej Los 4,820


Advertise | Privacy | Mobile
Web03 | 2.8.1411023.1 | Last Updated 11 Oct 2011
Copyright © CodeProject, 1999-2014
All Rights Reserved. Terms of Service
Layout: fixed | fluid

CodeProject, 503-250 Ferrand Drive Toronto Ontario, M3C 3G8 Canada +1 416-849-8900 x 100