The Use Of Gabor Filter And Back-Propagation Neural Network For The Automobile Types Recognition

Type of automobile is the general factor that makes automobile different from each other. But conventional sensor cannot detect the automobile and recognize its type. Because of those reasons, we made an experiment application that can count the number and recognize automobile based on its type. This application uses gabor filter for feature extraction and back-propagation neural network for training and recognizing type of automobile. The experiment was done using various parameters for back-propagation neural network and gabor filters. The experimental result shows that the best error rate of recognition results is 16%. It’s done with the brightness condition not too low or high.