Ask Your Question
1

Template Matching for Coins

asked 2015-09-21 04:34:48 -0600

Leo_ gravatar image

updated 2015-09-26 03:28:28 -0600

I am undertaking a project that will automatically count values of coins from an input image. So far I have segmented the coins using some pre-processing with edge detection and using the Hough-Transform.

My question is how do I proceed from here? I need to do some template matching on the segmented images based on some previously stored features. How can I go about doing this.

I have also read about something called K-Nearest Neighbours and I feel it is something I should be using. But I am not too sure how to go about using it.

Research articles I have followed:

edit retag flag offensive close merge delete

3 answers

Sort by ยป oldest newest most voted
5

answered 2015-09-21 05:26:55 -0600

updated 2015-09-21 06:17:09 -0600

I suggest you to find more research articles about coin detection and coin recognition. After reading some articles, you can get an idea about how to detect and recognize coins. And finally, you can improve some of these algorithms or propose a novel one.

In order to benchmark your algorithm, you can use your own coins, or use a standard database. For example, see the Coin Images Seibersdorf (CIS) database [1]

[1] http://muscle.caa.tuwien.ac.at/coin_p...

Some articles:

Automated Coin Recognition System using ANN

Image Processing Based Systems and Techniques for the Recognition of Ancient and Modern Coins

COIN RECOGNITION SYSTEM WITH ROTATION INVARIANT USING ARTIFICIAL NEURAL NETWORK

A Robust Method of Image Based Coin Recognition

A Robust Rotation Invariant Coin Recognition System

REVIEW ON THE COIN RECOGNITION SYSTEM WITH ROTATION INVARIANT

edit flag offensive delete link more

Comments

Thank you for the articles. I will have a look at them. I have updated my question by adding a few images. Maybe to clarify in what I am asking.

Leo_ gravatar imageLeo_ ( 2015-09-21 07:30:14 -0600 )edit
0

answered 2015-09-21 07:43:18 -0600

JeanPaulBarddal gravatar image

I am working with a very similar problem. Basically, I'm working with SVM, BoVW, SIFT and some basic preprocessing. I've got some pretty good results so far. My biggest problem is with darker coins. Since it's a part time work, I am not evolving quickly, so the papers cited by @albertofernandez will be good for me too.

Cheers

edit flag offensive delete link more
0

answered 2015-09-21 10:04:33 -0600

User1 gravatar image

Start first by image with one coin only then if the matching system that u use is good then u can crop each coin and pass it to your matching system.

In your matching system you can use SIFT to detect key points and find descriptors then use any matcher such as FlANN matcher to find the matches.

Finally u can use RANSAC as criteria if there is good match or not . Hope this help.

edit flag offensive delete link more

Question Tools

2 followers

Stats

Asked: 2015-09-21 04:34:48 -0600

Seen: 2,487 times

Last updated: Sep 26 '15