Falling Detection Method Based on Bi-modal Gated Feature Fusion
A single sensor is used to detect the method for human falling,which cannot adequately capture motion features,cam-eras can not obtain high-quality images under poor lighting conditions,and the point cloud sparsity of millimeter-wave radar reduces the effectiveness of remote target information.To solve these problems,a falling detection method based on bi-modal gated feature fusion is proposed.The method uses a radar and camera to synchronously detect the images.The radar branch obtains the fusion fea-ture based on the time-distance map and micro-Doppler map,and the visual branch extracts the optical feature of the target.The two features are sent to the gated fusion module,and the feature information is integrated according to the weight to realize the classifica-tion at the output layer.Experiments of the radar branch and the overall network are designed,the average accuracy of the radar branch fusion method is 91.7％,which is better than that of the single feature method.The accuracy of the gated fusion method in the overall network is 94.1％,which is 3.0％and 1.8％higher than that of the feature addition fusion and fore-tail fusion,respec-tively.It fully demonstrates that the method can effectively improve the performance of human falling detection.
万方数据
维普
