Study Of Automatic Vehicle Detection And Classification Using Laser Sensor

Automatic Vehicle Classification becomes more so essential to the electric toll collection of highways and other related applications in intelligent transportation system (ITS). Traditional systems which utilize loop detectors or video cameras have deficiencies. Loop detectors are expensive to maintain and video-based systems are sensitive to environmental conditions and do not perform well in vehicle classification.The new generation of range laser sensors that are being developed offer the promise of sensors which are not sensitive to lighting conditions and provide information which should give better vehicle detection and classification percentages than current systems.This thesis paper presents an algorithm for automatic vehicle detection and classification (AVDC) system on highways. The classification algorithm proposed is based on representing vehicles in a parameter space including the vehicle height, length and width. In addition of this features, the vector of height profile features are extracted. Including the vector of height profile to the feature space is a sufficient condition for enhancing the distinguish ability of vehicles having similar lengths and heights. However, for certain vehicle types, using the height profile is not necessary.We show that if the laser sensor satisfies certain requirements, then a precise feature vector can be extracted, and simple deterministic algorithm can be applied to determine the vehicle class.