Human activity recognition based on the inertial information and convolutional neural network / 生物医学工程学杂志

With the rapid improvement of the perception and computing capacity of mobile devices such as smart phones, human activity recognition using mobile devices as the carrier has been a new research hot-spot. The inertial information collected by the acceleration sensor in the smart mobile device is used for human activity recognition. Compared with the common computer vision recognition, it has the following advantages: convenience, low cost, and better reflection of the essence of human motion. Based on the WISDM data set collected by smart phones, the inertial navigation information and the deep learning algorithm-convolutional neural network (CNN) were adopted to build a human activity recognition model in this paper. The K nearest neighbor algorithm (KNN) and the random forest algorithm were compared with the CNN network in the recognition accuracy to evaluate the performance of the CNN network. The classification accuracy of CNN model reached 92.73%, which was much higher than KNN and random forest. Experimental results show that the CNN algorithm model can achieve more accurate human activity recognition and has broad application prospects in predicting and promoting human health.

Human Activity Recognition Based on Features Fusion / 医用生物力学

Objective To establish a human activity recognition （HAR）model based on human activity signals obtained by built-in sensors of the mobile phone, so as to support daily physical state assessment, special population monitoring and other biomedical researches. Methods The mobile signal was collected using the mobile phone built-in sensor, and the public data set UCI HAR and WISDM were used as experimental data. The HAR model was established by using the feature extraction method combined with convolutional neural network and autoregressive model. Results The models all achieved more than 90% recognition accuracy in the self-collected dataset, UCI HAR and WISDM. Conclusions The introduction of autoregressive model can avoid the manual design eigenvalues and effectively reduce the computational complexity of large-scale stacked convolutional layers. The research findings prove that the method based on feature fusion can effectively recognize human activity.