Depth binary feature facial expression recognition method based on lightweight network

Abstract

The invention discloses a depth binary feature facial expression recognition method based on a lightweight network, and belongs to the technical field of mode recognition. The method mainly comprisesthe following steps: firstly, constructing a set of convolutional neural network framework for binarizing parameters, and implanting a binary convolution mode into each residual network layer to forma bidirectional decision network model; then, extracting LBP dynamic radius features of an image input into the network based on pixel gradient and creating an LBP weight map with Huffman weight and an LBP binary map with Huffman weight; taking an LBP weight map, an LBP binary map and an original image as multi-input features of a BRCNN network, and constructing depth binary features; and finally,cascading and classifying the depth binary features. The parameter quantity during network training is greatly reduced, and the calculation cost of the network is reduced; the expression ability of the features is enhanced, and the robustness and rate of the method in facial expression recognition are improved.

Description

technical field [0001] The invention belongs to the technical field of computer pattern recognition, in particular to a method for recognizing human facial expressions. Background technique [0002] In the history of computer vision development, deep learning has become one of the most popular means of solving computer vision problems; convolutional neural networks have also become one of the classic techniques of deep learning. The emergence of LeNet and AlexNet networks has promoted the development of deep learning, and then deeper and wider convolutional neural networks such as VGGNet, ResNet, and InceptionNet have developed the extracted image features in a more accurate and efficient direction. However, blindly expanding the depth and width of the neural network cannot directly and effectively improve the accuracy and efficiency of visual tasks, but will increase the burden on equipment operation. How to design a "fast and good" lightweight network has become the focus...

AI Technical Summary

Problems solved by technology

However, how to achieve the stage of fast training and high recognition accuracy in the facial expression recognition task combining traditional feature operator methods and deep network features is still a difficult point

Images