A Car Type Recognition Algorithm with Traffic Image based on Structure Features
Bo Liu, Jun Feng, Hongcai Chen, Changyou Zhang · 2017
Aiming at the problem of high recognition error, a method for car type recognition with traffic image is proposed based on component structure features in this paper. Firstly, structure features of multiple components are extracted. Secondly, the feature selection method based on Fisher criterion is used to reduce the feature dimension. Then the robust combination features are obtained by incremental learning. Finally, SVM multi-class classifier is adopted for car type recognition and accuracy is up to 100%, which is superior to the convolution neural network method.