World Coin Currency Detection Using Various Techniques and Their Comparative Analysis

The ability to automatically identify and classify coins is a key challenge in numismatics and currency recognition. Our paper presents an approach combining machine learning and deep learning to detect and classify coin currency, face value, and country. We collected a dataset of coin images from 32 different countries, which we preprocessed and labeled by face value and country. Then, to categorize the coins, we used a variety of machine learning algorithms, such as Convolutional Neural Networks (CNNs). We evaluated the performance of the models using metric accuracy. Our results show that the CNN models achieved high accuracy in detecting the currency, face value, and country of the coins. Specifically, our best performing model achieved an accuracy of 86.96%. We performed experiments to assess how well the models withstand variations in lighting conditions and changes in the orientations of the coins. Our results indicate that the suggested method is efficient in identifying and classifying coins from various countries, and has potential applications in automated coin counting, sorting, and authentication.