Real-Time Object Detection and Localization System with Mild Artificial Intelligence

Abstract

This project is about the detection and localization of an object, implemented using MATLAB simulation software, in order to replace work that requires human vision and produce higher throughput and better efficiency. This project aims to develop a system that uses a built-in camera of a laptop as an input to detect an object, identify the class of the object and locate the position of the object in a real-time video. The system will go through three stages of development which are, training preparation, training process, and performance evaluation. A random number of images will be fed into the system to evaluate the performance of the system, and the result obtained will be classified into three classifications (TP, FP, and FN). More input data used to train a system will produce better accuracy and performance. The surface area of an object is another key factor that helps the system to increase its performance.