A Highly Robust Vehicle Detection, Tracking and Speed Measurement Model for Intelligent Transport Systems

The high pace rise in vehicle counts, traffic density, and security concerns, a potential system for traffic surveillance, vehicle monitoring and control has become an inevitable need to facilitate intelligent transportation system (ITS). Although, numerous approaches have been proposed for moving vehicle detection and tracking, still the optimization needs can’t be ignored. In this paper, a robust image processing based vehicle detection, tracking and speed measurement model has been developed. The proposed model implements enhanced preprocessing, background subtraction, morphological operation as well as feature mapping processes for moving vehicle detection, tracking and speed estimation. To achieve optimal performance, a multi-directional filtering scheme has been develop for moving vehicle detection, which considers intensity, moving pixel orientation etc. for efficient candidate vehicle detection in traffic video. To enhance background subtraction, a novel multi-directional intensity strokes estimation approach has been introduced that plays significant role for distinguishing vehicle region from other background contents. In addition, the enhanced thinning and dilation based morphological process has been introduced that exhibits more precise and accurate vehicle detection. A novel feature clustering scheme with heuristic filtering based blob analysis and adaptive bounding box generation makes our proposed model more efficient for vehicle detection and tracking. Furthermore, a novel moving vehicle speed estimation approach has been developed that can be significant for efficient ITS systems.