Cyber Times International Journal of Technology & Management

This paper presents an automatic vehicle counting and classification of vehicles which is a Fundamental task for video surveillance in urban traffic management. In Intelligent Transportation System (ITS), monitoring the traffic is one of the critical tasks. The proposed system can tackle the problem of vehicle occlusions caused by shadows, which generally lead to the failure of vehicle counting and classification. The system uses multicue background subtraction algorithm for the separation of moving vehicle from the captured video. The system tracks moving casted shadows on sunny days using a single standard camera. In this, with the combination of luminance and chromaticity is threshold between learned background and the current frame. Moving casted shadows are removed from the foreground objects using top hat transformations. For three dimensional vehicle tracking and classification, system considers image rectification and camera calibration, Linear tracking with 2 D Silhouette estimation, and 3D volume estimation.
According to these parameters moving objects are classified as light vehicle, heavy vehicle and two wheelers.