Automatic Coin Detection and Counting Based on Image Processing with OpenCV Python

Abstract

Image processing is an important field in object detection and calculation, especially in the field of computer vision. A problem that often arises in automatic coin counting is the inability of machines to accurately recognize and calculate coins quickly when there are variations in conditions, such as differences in size, color, lighting, and orientation of coins in the imagery. This research develops OpenCV and Python-based methods to detect and calculate the number of coins in the image. The stages of this method include pre-processing, segmentation, feature extraction, and object detection. Pre-processing improves image quality and reduces noise, while segmentation separates coin objects from the background. The extraction feature identifies the characteristics of the coin such as edges and colors, aiding in the object detection process. This method was tested on a dataset of coin images with various variations in size, exposure, and orientation, and the results showed an average accuracy rate of 98.08%. This research contributes to the development of automated systems for object counting in various contexts and can be the basis for further research in the field of image processing and computer vision