In order to solve the problem that the recognition rate is easily affected by the change of view Angle in the mainstream gait recognition process,the coupled network model is used to solve the problem that the minimum distance between classes is greater than the maximum distance within classes,and the comparison of gait samples is used to solve the recognition problem.The gait energy diagram is synthesized by binary image sequences,and the model is trained and optimized by logistic regression and contrast loss function.The gait recognition performance of the coupled neural network was verified by experiments and compared with the recognition results of the common convolutional neural network(CNNs).The recognition rate reached 73.7％and 60.5％in the case of backpack and overcoat occlusion,which is higher than CNNs,and the accuracy of gait recognition under occlusion is improved.
关键词
步态识别/步态能量图/视角变化/耦合神经网络模型
Key words
gait recognition/gait energy diagram/change of perspective/coupled neural network model
维普
万方数据