Abstract
Coins are integral part of our day to day life and now it becomes a basic need that coins can be sorted and counted automatically. This paper describes Automated Coin Recognition System for the recognition of Indian Coins with translation and rotation invariance. The paper presents a novel approach that uses the normalized local spatial features of the coin image to derive an abstract image. Decomposition of abstract image in to concentric circles is then used to extract a set of compact and effective features. The proposed method uses a single gallery image per coin for the recognition and show a robust and high recognition performance of 97.43% under variable conditions such as rotation and translation.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
