The Study on Real-Time Vehicle-Flow Analyzing and Counting System Based on Video Processing

• Main Authors: Yu-Feng Lin, 林裕峰
• Other Authors: Thou-Ho Chen
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/51754661332142203139

碩士 === 國立高雄應用科技大學 === 電子與資訊工程研究所碩士班 === 95 === In recent year, the traffic density is rapidly increasing in most countries, making the traffic problem become one of major issues today. Therefore, a large number of researches are presented. The Intelligent Transportation System (ITS) is one of the best ways to solve the traffic problem. An ITS combines telecommunications, electronics, computer application, control and sensor technologies to improve traffic safety and efficiency. In this thesis, a vehicle counting method is proposed to measure traffic flow based on video processing. The motion detection and objection segmentation are used to grab the vehicle mask. By analyzing the mask of the vehicles and their minimal boundary boxes, the meaningful and obvious features of vehicles are extracted. Tracking moving targets is achieved by comparing the extracted features and measuring the minimal distance between two temporal images. The temporal classifier with statistical method is presented to increase the accuracy of vehicle classification. Finally, the velocity of the vehicles and the vehicle flow through the field of vision can be extracted from the image sequences by analyzing the data of vehicles. The experiment results show that the accuracy is about 90%, and this method can meet the requirement of real-time.