Efficient Vehicle Tracking and Classification for an Automated Traffic Surveillance System

As digital cameras and powerful computers have become wide spread, the number of applications using vision techniques has increased enormously. One such application that has received significant attention from the computer vision community is traffic surveillance. We propose a new traffic surveillance system that works without prior, explicit camera calibration, and has the ability to perform surveillance tasks in real time. Camera intrinsic parameters and its position with respect to the ground plane were derived using geometric primitives common to any traffic scene. We use optical flow and knowledge of camera parameters to detect the pose of a vehicle in the 3D world. This information is used in a model-based vehicle detection and classification technique employed by our traffic surveillance application. The object (vehicle) classification uses two new techniques − color contour based matching and gradient based matching. We report good results for vehicle detection, tracking, and vehicle speed estimation. Vehicle classification results can still be improved, but the approach itself gives thoughtful insight and direction to future work that would result in a full fledged traffic surveillance system.