Vehicle detection and tracking

Abstract

Real time trafficc monitoring is one of the most challenging problems in machine vision. This is one of the most sorted out research topic because of the wide spec-trum of promising applications in many areas such as smart surveillance, military applications, etc. We present a method of extracting moving targets from a real-time video stream. This approach detects and classifies vehicles in image sequences of trafficc scenes recorded by a stationary camera. Our method aims at segregating cars from non-cars and to track them through the video sequence. A classication criteria based on the features is applied to these targets to classify them into categories: cars and non-cars. Each vehicle can be described by its features. The template region is estimated by means of minimum distance approach with respect to centroid of the obtained blob of the target. Extraction of features from each frame ensures eefficiency of the tracking system.