Human Posture Recognition Based On Convolutional Neural Network

Posture recognition is an important part of human behavior recognition, and also an important research content of human behavior recognition system. Its application value is increasingly extensive. In recent years, it has become a research hotspot in the field of computer vision.However, due to the high complexity of the human body, most of the images need to be preprocessed before feature extraction, which leads to difficulty in feature extraction and low recognition efficiency.Therefore, this paper proposes a research method of human posture recognition based on convolutional neural network.The model has 11 layers. Convolution and pooling operations are performed on the five human poses in the sampled data set, and finally enters the fully connected layer for classification to complete the training and recognition of the data set. The results show that, compared with traditional machine learning methods, this model allows the network to extract features for recognition and classification, which not only eliminates complex feature extraction methods, but also has better recognition performance and better results.