Neural Network Approach for the Identification System of the Type of Vehicle

This paper represents a framework for multi-class vehicle type identification based on several geometrical parameters. The system of identification of object must thus have a very great adaptability. We represent a system of identification of the type (model) of vehicles per vision. Several geometrical parameters (distance, surface, ratio ...) of decision, on bases of images taken in real conditions, were tested and analyzed. The details of preprocessing as well as the features represented above are described in this paper. According to these parameters, the rate of identification can reach 95% on a basis of images made up of 9 classes of the type of vehicles. Then artificial neural network (ANNE) was used to verify and to classify the different types of the vehicles, and a ratio of identification of about 97% was obtained. The details of the implementation and results of the simulation are discussed in this paper.