Road traffic monitoring based on adaptive sensing of motion flows

• Main Author: Lin, ShiKai.
• Other Authors: Ang Yew Hock
• Format: Final Year Project
• Language: English
• Published: 2009
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/17930
• Institution: Nanyang Technological University

Road Traffic Monitoring Based on Adaptive Sensing of Motion Flows‟ is a final year project that aims to develop a real time computer vision system that can analyze traffic data, detecting traffic incidents and alert the user hence immediate actions can be taken. In this report, theoretical aspects of computer vision techniques used in the development of the program are explained in details. These techniques include grayscaling and median filtering being applied before image differencing is used for analysis of the traffic flow. The technical aspects of the program are also discussed and evaluated. The proposed system is able to load a video clip or using live streaming for analysis. The system is built primarily on image differencing methods and it is able to recognize the region of motion, the foreground objects as well as detecting static objects on the road which can be a road hazard. The evaluation of the system showed that the program works extremely well during the day time but the reliability of the program drops when the weather changes or due to unstable lighting conditions. Future recommendations discuss about the areas where improvements are needed.