Rotation Invariable Method for Currency Coin Detection

This paper is to detect denominations of Indian coins. Counting the coins manually, collected in large amount, such as coins collected at Indian temples is difficult. This can be made easy with the proposed method of coin detection, where scanning of the coin from both the sides is considered while passing the coin through conveyer belt. By selecting one of the two images acquired, we proceed further with application of segmentation technique. Traversing row wise and column wise through the image acquired, centre and the radius of the coin is obtained. Database image is then selected for comparison based on the radius matching. The methodology proposed here used reduction technique that is the input image is reduced by the database image repeatedly, by rotating it with a fixed angle each time. Next, by plotting the graph of resultant values gives decision whether the coin match found or not, by comparing the minima to the fixed threshold. Denomination of the coin is verified by comparing coin from both the sides. Thus, the method proposed here is rotation invariable, also, by using two way scanning and comparison of coin, method determine the denomination accurately, even if the database is having different coins with same radius. KeywordsSegmentation, Radius Calculation, Reduction.