Morphological based Segmentation and Recognition of Indian Coins

Coins have very much importance in human’s day to day life, which are used in everyone’s daily routine like banks, super markets, vending machines etc. So, there is a basic need to automate the counting and sorting of coins. Coin recognition applications play an important role in industry and computer vision. Many approaches developed for the coin detection and calculate its corresponding values. This paper recognizes Indian coins of different denomination The recognition process consists of three steps, 1) we present a simple and fast method for coin segmentation, based on morphological thresholding technique to remove noise and to enhance the quality of coin image, 2) we applied some simple descriptors like mean intensity, area, perimeter to extract the regional features of the coins used for recognition and sorting and 3) we performed edge detection using statistical operators. Only after detecting the edges in the image the number can be recognized. In this paper we describe the pattern recognition method used for identification of coins in the new coin recognition and sorting system. Keywords— Edge detection, features extraction, segmentation, thresholding, top-hat transformation.