An optimized object detection system using machine learning technique

Abstract

On average at least one person dies in a vehicle collision accident per minute globally. In addition, the accidents cause injuries to nearly ten million people every year and serious injuries to thirty percent of them. Vehicle congestions, accidents, and car robberies are occurred due to the increase in vehicles, which leads to serious issues. To solve these issues, traffic monitoring is very important. Many surveillance cameras are used for monitoring the traffic. Ordinary video surveillance systems require reviewing all video sequences in order to detect objects, in case of checking some abnormal event. Obtaining newlineestimated results is a time-consuming and effort-demanding task. However, intelligent newlinesystems, inspired by human beings, are established to reduce such wasted time and effort. Video-based detection mechanisms are fairly cost-efficient, simple and also provide more potential advantages including more flexibility compared to inductive loops, as well as larger detection areas. Thus, video-based traffic monitoring is applied to three major stages: detection, tracking, and data association. In the detection and tracking phase, the accurate detection of multiple vehicles in a complicated traffic environment is too difficult. This process is made more difficult if there are occlusions between vehicles. For this drawback, robust detection algorithms are required for vehicle detection and tracking. So, this research methodology proposed new methods for vehicle detection and tracking.