GPU based non-overlapping multi-camera vehicle tracking

Abstract

Vehicle tracking and surveillance is an area which is having a considerable attention in the context of security and safety. The detection and tacking of moving vehicles through multiple cameras is considered as a method of vehicle surveillance. This work addresses a problem of detecting and matching vehicles through multiple cameras. The power of GPUs are used to increase the number of video streams which can be processed using a single computer. In the detection process the Gabor filter is used as a directional filter and the SURF is used by the matcher to uniquely represent the vehicle.

A comprehensive review of vehicle detection using computer vision

“…both employed this method for vehicle detection, finding vehicles per frame to be 21 (frames/sec). In another study [70], a GPU based multiple camera was used to match unique representations of vehicles. The accuracy was found to be 94%, with limitations including rotation instability, illumination variations and occlusion.…”

Study on Method to Generate a Virtual Lane using V2V-based Clothoid Road Model

“…The driving trajectory was created from the equations from Clothoid road model, which uses the road curva-ture for its calculation. Equation (1) shows the formula to obtain the curvature. Here, K represents the curvature and R for the radius of curvature.…”

“…The active safety system such as LDWS (Lane departure warning system), LKAS (Lane keeping assist system), BSD (Blind spot detection), AEB (Autonomous emergency braking), FCW (Forward collision warning), or RCW (Rear-end collision warning) is mounted to commercial vehicles after considerable development. The currently existing active safety system detects the target using Lidar, Radar, or Camera sensors [1] [2]. The system using sensors not only has a limited range of detection on curved roads or intersections, but also cannot detect areas of blind spots.…”