Detection for Vehicle´s Overlap Based on Support Vector Machine

In the paper, support vector machine is presented to detection for vehicle´ s overlap, which has stronger generalization ability than the algorithm based on the empirical risk, such as artificial neural network. In the process of detection for vehicle´s overlap, principal component analysis is used to extract the features and reduce the dimension of features. Then, detection model for vehicle´s overlap based on SVM is constructed. We collected the 220 vehicle images including the overlapped vehicle images and non-overlapped vehicle images as the experimental data. The experimental results indicate that the accuracy of detection the vehicle´s overlap by SVM is higher than that of BP neural network.