The image dataset of Indian coins: A machine learning approach for Indian currency
Yogesh Suryawanshi, Vidula Meshram, Kailas Patil, Michael Testani, Prawit Chumchu, Anmol Sharma
Abstract
In an increasingly digital world, the significance of creating a Comprehensive Image Dataset of Contemporary Indian Coins (CIDCIC) cannot be overstated. This research presents a dataset comprising 6,672 images of 53 different classes of Indian coins, including denominations of 25 Paisa, 50 Paisa, 1 Rupee, 2 Rupee, 5 Rupee, 10 Rupee, and 20 Rupee. The images of coins with various shapes and sizes are taken from obverse and reverse sides in various environments and different backgrounds. The core significance of this dataset unfolds in its potential to offer invaluable assistance to visually impaired individuals as they navigate their daily financial transactions. The dataset is a significant contribution to the domains of computer vision, artificial intelligence, and machine learning, specifically addressing the challenges related to coin detection, recognition, and monetary system integrity. These technologies can empower visually impaired individuals to independently and accurately recognize and distinguish between various coin denominations, thereby enhancing their participation in the financial realm. The dataset addresses limitations in existing dataset of having limited size, and scope. It addresses the limitations associated to the limited number of coins and the lack of diversity in images, encompassing various angles, environments, backgrounds, and directions of coins. The dataset provides a broader and more up-to-date representation of contemporary Indian coins.