The Vehicle collision warning system detects the vehicle ahead

Abstract

In the field of computer vision and deep learning, vehicle identification is an important subject, which can be applied to the traffic auxiliary system, which can greatly reduce the occurrence of traffic accidents. In this paper, a large number of positive and negative samples are collected, and the algorithm of AdaBoost cascade classifier is used to train the classifier based on Haar feature and LBP feature. Finally, the selected feature and classifier are used for testing. Experimental results show that both of the two eigenvalues have better characteristics and higher detection accuracy when detecting the target vehicle, but LBP features have a faster detection speed, while Haar features have a higher accuracy.