Abstract: We proposed a method for Indian coin detection under varying lighting conditions. The basic idea behind our method is to extract features of known coins and put in a database. And when a new coin is entered its features are compared with the images of the coin that stored in the database. We trained our system under different lighting conditions. We assumed different lighting intensities, which occurs due to various reasons such as power fluctuations. We tried to detect Indian coin value under these conditions. This work basically involved automatic detection and recognition of Indian coin using morphological operations of image processing. Image processing is a process in which an input image is transformed into digital image form and performs some operation on it, In order to get high image property and to extract some useful information from it. This work is a prototype simulation and a proof of concept. The application of this work ranges from simple vending machines to automatic mobile recharging. In this thesis, we use MATLAB based simulation to obtain our result.

Keywords: Edge detection, Segmentation, Thresholding, Edge enhancement, Features extraction, Patter matching.