Vision Based Motorcycle Monitoring at Intersection of Nepal Roads
Published in NaSCoIT, 2018
Recommended citation: H. Acharya, Basanta Joshi, “ Vision Based Motorcycle Monitoring at Intersection of Nepal Roads”, Proceeding of 9th National Students’ Conference on Information Technology (NaSCoIT 2018), December 29, Kathmandu, Nepal
Abstract
Computer vision plays important role in Intelligent Transportation System (ITS) for traffic management and surveillance. This paper implements existing vision-based detection and tracking algorithms to detect and track motorcycles. While few research has been carried out for vehicle detection, but no research has been carried as far known for tracking vehicle in Nepal roads at intersections.GMM and Haar Cascade Classifier method are used for detection.Results show that contextual combination in bike detection gets 89% for sensitivity, 60 % for precision and 0.72 for F1-score. Low precision is due to high false positive in detection of every frame in video. The optical flow tracking with Haar detections rejects false positive which was high detected in detection process. This tracking improves all performance metrics: Sensitivity, precision, F1-score and accuracy. While tracking with optical flow gets 86.96% for sensitivity, 95.23 % for precision, 83.3 % for accuracy.