Coin Detection by an Improved Convolutional Neural Network Approach

The coin detection in vending machines is a crucial task for ensuring accurate and efficient transactions. In this work, an improved Convolutional Neural Network (CNN) approach for coin detection is introduced. The CNN model has been trained with the images of various types and sizes to recognize and classify these coins accurately. CNN architecture leverages its ability to extract relevant features from coin images, enabling it to detect coins regardless of orientation and lighting conditions. A promising result has been attained in terms of coin detection accuracy and speed, making this approach for real-time applications in vending machines. This research contributes to improving the reliability and efficiency of vending machine transactions, reducing errors, and enhancing user experiences. The CNN-based coin detection approach holds significant potential for various vending machine applications, from beverage dispensers to ticket kiosks, and can be adapted to accommodate different coin denominations and currencies.