You are on page 1of 7

Coin Recognition System

-Madhuraj Tripathi and Muheet Hajela

The Coin Recognition System


The objective of this system is to classify the Indian coins of different denomination present in an image whether placed correctly or upside down. The system based on the denominations of the recognized coins is able to provide their sum as well which opens up a new avenue for the application of this system. The radius of old coins and new coins of different denominations are considered for classification.

The Assumptions
The assumptions taken for the process are:

The coins belong to the Indian currency currently in circulation. The images of the coins are taken from a specified distance. The coins are complete and not broken of or worn out to a great
extent.

The Process
The steps involved in this whole process are: 1) The image is input and then if it is in RGB colour format it is converted to a grayscale image.

2) Then we use canny edge detection on this grayscale image with a threshold value of about 0.3.
3) Then with a structural element disk of about 2 pixel radius we dilate the image with edge detection.

The Process
4) Then we remove any horizontal or vertical lines in the image present as noise. 5) Then we remove all connected areas with less connected components than the pixels specified. 6) Then each of the areas are labelled and processed one after another. 7) For each area the min distance for each black pixel enclosed, the minimum distance from the nearest white pixel is determined and represented in the form of a circle and the distance is represented in the form of intensity of whiteness. Thus, the centre of each region identified is thus located and the radius id given by its very intensity.

The Process
8) For the purpose of presentation we have multiplied each of the regions identified separately.
9) The coins denomination is the determined on the basis of its radius.

Thank You

You might also like