The Research of Applying Image Processing Technology to Detect and Track Vehicles on Highway

• Main Author: 趙冠維
• Other Authors: 鍾翼能
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/67979685638275407794

碩士 === 國立彰化師範大學 === 電機工程學系 === 101 === This thesis proposes the application of image processing technology in the vehicle detection system, which can automatically retrieve traffic information from the monitor on highway. The process can be divided into vehicle detection and vehicle tracking. Its main purpose is to collect traffic parameters, and to achieve the traffic management. In the vehicle detection, the system uses a series of image processing methods to obtain vehicle targets. The shadow will be removed based on the threshold in the HSV color space. All detection processes are set in the region of interest, so that computational complexity can be reduced to improve execution speed. In the vehicle tacking procedure, we use feature extraction method to obtain the bounding box of the vehicle. This thesis uses the bounding box boundary length and area to complete the vehicle size classification, and uses the virtual detection line to calculate the vehicle counting. We can get vehicle speed by calculating the frames it takes when vehicles pass the sensing area. As shown in the experiments, the proposed method can provide accurate traffic information and be helpful to the traffic management.