Optical system to detect and calculate moving object speed

Abstract Detection objects demand increases in various applications within last decay. An optical system is designed to automatically detect moving objects in real time and calculate the speed of these objects based on the deep learning algorithm R-CNN. The data is compared with the Alex-Net method. The suggested algorithm is one of the machine learning algorithms, which depends on the principle of training and learning. Then it relied on the successive classifier. The study included several algorithms that were programmed and developed using MATLAB (R2021a) software. The suggested system consists of a phone camera, a red ball, and different environmental lightness (changing from one to eight lamps). There are two luxmeter used to check lightness intensity around the ball and the camera phone. There are four parameters used to assess the quality of the algorithms and the suggested setup: accuracy, average time, detection percentage, and speed. Results show high quality to detect and track the ball with almost 100%.