Recognition of Abnormal Vehicle Behaviors Using Combined Features

In order to improve the efficiency of traffic management in urban areas, a recognition method of abnormal vehicle behaviors in urban traffic scenes using combined features of histograms of oriented gradient (HOG) and local binary pattern (LBP) is proposed in this paper. Firstly, three feature extraction methods, such as histograms of oriented gradient, local binary pattern (LBP), and edge orientation histogram (EOH) are analyzed, comparatively. Secondly, the experiments of abnormal vehicle behaviors, such as normal driving, running a red light, pressing line, and illegal vehicle steering, are conducted using combined features and support vector machines. Results showed that the combined features of HOG and LBP offers the best classification performance, and the recognition rate is over 93.6%. With combined features of HOG and LBP, the classification accuracies of illegal vehicle steering are over 82.7% in the experiments.