Research on Facial Expression Recognition Method Based on Multiscale Binary Feature

Abstract:
In order to realize facial expression recognition and detection and improve the accuracy of deep learning for facial expression recognition, a facial expression recognition algorithm based on a multi-scale binary feature network model is proposed. In the traditional method of facial expression recognition by convolutional neural network model, the feature vector is extracted manually. This method is not only time-consuming, poor stability, but also insufficient information feature extraction of the image to be detected. The facial expression recognition algorithm based on the multi-scale binary feature network model uses feature vectors of different scales to extract features from the images to be detected. Then different features are input to the fully connected layer for feature fusion, and finally the fused features are sent to the convolutional neural network classifier for facial expression recognition. Through experimental verification on different data sets, different parameters, and different feature extraction methods, the experimental results show that the correct rate of facial expression recognition by this method is 99.58%. Facial expression recognition based on multi-scale binary feature learning achieves better facial expression recognition results than other methods.