A capstone project report submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Engineering Civil and Infrastructure Engineering, School of Engineering and Technology

Capstone Project (B.Sc.)-Asian Institute of Technology, 2017

Lane change detection systems play vital role in nowadays. These play a prominent role to achieve the rising demand for optimum traffic management and safety improvements with rapid increase in vehicle usage. This research presents accuracy for lane change detection systems. There are five main lane change detection systems available for my research. Four companies are manufacturing these five systems and these have distinctly different interfaces. Initially data was collected, traffic camera footage was watched and then text files were read. Then identified what vehicles violated the rule, which vehicles were detected without violating rule. The system can detect vehicle type and license plate. The five data sets which were obtained for the study was analyzed using F1 score. The F1 score is a measure of a tests accuracy. A score more than 70% means system is performing optimally. Finally this research attempts to compare these systems and their advantages and disadvantages, errors and further improvements.