Image processing tool in C#

Question

2.00/5 (2 votes)

See more:

I need to develop a software which can identify a specific number in an invoice, when that paticular image is scanned. So the process needs to identify the invoice number which is printed in the invoice image. This invoice number has a unique format like "BMAINV_xxxxxx". Any paticular way of doing this?

Posted 20-Sep-12 1:20am

WebMaster

Add a Solution

Comments

bbirajdar 2-Oct-12 12:59pm

Use barcode

3 solutions

Solution 1

I wonder, you've to use the OCR SDK. You'll find plenty of them here.

Posted 20-Sep-12 1:24am

Malli_S

Comments

[no name] 20-Sep-12 7:28am

no any ways to develop such a thing?

Malli_S 20-Sep-12 7:30am

There are, but why to re-invent the wheel? You can use the existing one. Try it, test it, and if feel satisfied, purchase it (if required).

[no name] 20-Sep-12 7:34am

I'm going to develop this for a company, so they won't let me use uncertified versions of any software & they won't be aggreed to spend like $5000 to buy this software, because this is just a small modification for our excisting system

Malli_S 20-Sep-12 7:38am

Then I suggest to use the SDK that give minimal features required and is cheaper in rate. You can use the trial version for development, it's not kind of un-certified version. Once your development is done, and you're satisfied with the features/functionality of the SDK, you can purchase then. Try out Pegasus SDK.

Solution 3

Take a look at this:Neural Network OCR[^]
and other articles from this author, it may give you an idea.
Also take a look at this: Handwriting Recognition using Kernel Discriminant Analysis[^]

Posted 2-Oct-12 6:17am

sjelen

Updated 2-Oct-12 6:19am

v2

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)

Joseph-Ali · Accepted Answer · 2012-10-02T05:27:00

The details depend on your requirements and how you want to implement your application but OCR alone will give you all the text on the image unless you apply the OCR to specific region. I'm NOT sure if all OCR engines support zoning or not but in our case, the engine we used did support zones.
If you know the co-ordinates of the invoice number, you can hardcode the values.
If you don't know the co-ordinates or they are not fixed in position, you have at least 3 options:
1) You can parse the whole text and check for "BMAINV" then take the next after it.
2) Use Forms recognition to search for a nearby mark that has a fixed position in relevance to the invoice number.
3) Use correlation (Search for an image inside a larger image) for example, if the image contains the words "Invoice number: XXXXX", crop a small region that contains "Invoice number" and search for it and then define a zone on the invoice number then OCR it.