how to make Optical character recognition (ocr) with winform C#?

Question

1.00/5 (3 votes)

See more:

I am doing thesis but I do not know how to convert from image to text in textbox?
I used emgu CV and try to make winform C#, but not success.
please help me. Thanks.

additional information copied from comment below
Sorry, I don't know. My thesis is "Smart parking using process Images and RFID technology". I have done about 90%. I have recognized the vehicle number plate from photos, separating each character, read RFID tag.... I just do not know how to convert from character image to text characters to store in the database. Only this.
My major is telecom not IT, my C# skill is not good enough. Thanks!

Posted 31-Aug-14 8:09am

Member 10390715

Updated 31-Aug-14 22:16pm

Nelek

v2

Add a Solution

Comments

Sergey Alexandrovich Kryukov 31-Aug-14 15:22pm

May I ask you the topic of your thesis?
—SA

Member 10390715 1-Sep-14 3:49am

My thesis is "Smart parking using process Images and RFID technology". Thanks!

3 solutions

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

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

Nelek · Accepted Answer · 2014-08-31T08:38:00

Solution 1

Step 1) Do a little research. Google is a good start point
Step 2) Start coding
Step 3) Compile and use the debug to solve little issues
Step 4) When you get a problem you don't know how to solve but at least you tried it, then come back and ask for something concrete with a snippet of the code giving problems

Sorry if this is not the answer you were looking for. But your question is a bit too wide to be answered at the "Quick" Answers. It is better and you get faster help if you make 10 concrete questions about concrete problems, than a big question about a "how-to guide"

On the other hand... I strongly recommend you to read:
What have you tried?[^]
How to ask a question?[^]

Read those links carefully and learn from them. If you do, your next question will get much better answers.

addition
best way to post code snippets[^]

Posted 31-Aug-14 8:38am

Nelek

Updated 2-Sep-14 6:30am

v3

Comments

Sergey Alexandrovich Kryukov 31-Aug-14 15:26pm

5ed, but I must add that finding a good open-source code for OCR is extremely difficult. I know a number of bad solutions, and a number of those which look good but need a lot of extra work. I don't know anything I could use immediately; I only know that I would get a solution which could more or less satisfy me only with serious effort. Do you know any good open-source product?
—SA

Nelek 1-Sep-14 4:16am

No Sergey, I don't know any Open Source Solution that could be "ready-to-go" with garanties. The point of my answer (btw, thanks for the 5) was about the question itself. It is a tesis, so it is supposed a bit of investigation and it presuposes a merely level of inteligence. So the question could have been a bit more elaborated. Don't you think?

Sergey Alexandrovich Kryukov 1-Sep-14 11:32am

All right, thank you for answering. Of course, I agree with you, about the thesis. My students and I worked under strict requirements: 100% of purely original results, accurately referencing all the sources, everything we would rely upon or even the ideas we used, and so on... That's why I asked about the topic of the thesis. After OP answered, I understand that existing OCR algorithms should be leaned and can be used. CodeProject has a number of articles...
—SA

Sergey Alexandrovich Kryukov 1-Sep-14 11:47am

Actually, I think OP already done some decent effort. I referenced some works I tried out; and I think it's quite possible to work existing works into the solution OP needs, because the OCR on just the license plates should be easier.
Please see Solution 3.
—SA

Member 10390715 1-Sep-14 3:48am

Sorry, I don't know. My thesis is "Smart parking using process Images and RFID technology". I have done about 90%. I have recognized the vehicle number plate from photos, separating each character, read RFID tag.... I just do not know how to convert from character image to text characters to store in the database. Only this.
My major is telecom not IT, my C# skill is not good enough. Thanks!

Nelek 1-Sep-14 4:20am

I am sorry that you got stuck with your tesis. Don't get me wrong, I would gladly help you, if I could. What I wanted to point out is... How do you expect to get help just with what you are asking? You were not describing what you did (I copied your explanation here to the question itself). You were not giving any details where people here could start with.

As I said, read those links, they (specially the one on "what have you tried") explain very good what is to be here on the side of people wanting to help in the spare time and being asked such vague things. Read it, so you can get a picture of what I mean

Member 10390715 1-Sep-14 5:06am

ok, sorry and thanks, I will see it

Nelek 1-Sep-14 5:11am

Don't be sorry. I know it is difficult. CP helped me a lot during my tesis. I just wanted to tell you how you can get the help you need in an efficient way.

Member 10390715 2-Sep-14 11:58am

Thanks

Pikoh · Accepted Answer · 2014-08-31T23:28:00

Solution 2

Have a look at this (EMGU). Hope it helps.

EDIT
Sorry..didn't read you already tried EMGU. Maybe you could tell us what exact problems did you find with EMGU.

You may also try TESSERACT

Posted 31-Aug-14 23:28pm

Pikoh

Updated 1-Sep-14 0:47am

v2

Comments

Member 10390715 1-Sep-14 9:57am

Thanks, I will try!

Sergey Alexandrovich Kryukov · Accepted Answer · 2014-09-01T05:45:00

This is very good that you managed to separate the characters in your image recognition effort. You see, as you need to recognize only the license plate number, this should be simpler that the pages of text of unknown fonts, which typical OCR product could do. You have very few allowed characters and fixed font, but you cannot use a dictionary. In this case, even the relatively bad product can help, even Tesseract.

There are .NET wrappers for Tesseract:
http://code.google.com/p/tesseractdotnet[^],
https://github.com/charlesw/tesseract[^].

See also my past answer: Tesseract Wrapper in C#[^].

You can find a good number of articles on OCR technology on CodeProject. You can find them all and read; some products (in full source code, of course) are very good, but they need some work to produce a ready-to-use product. But as you already done some essential work, and your problem is relatively simple, I think you can easily adopt these codes and your derived work will do what you need. Just search them: http://www.codeproject.com/search.aspx?q=OCR+%28%22.NET%22+OR+%22C#%22%29&doctypeid=1&sort=createddesc[^].

I referenced several of my past answers; and these answers reference some CodeProject articles:
i need an ocr for a camera image capture[^].

Good luck,

—SA