Gait recognition method based on depth neural network

Abstract

The invention discloses a gait recognition method based on a depth neural network, and the method comprises the steps: carrying out the preprocessing of an original gait video image, extracting the gait Gaussian image of the original gait video image, and dividing a sample data set into a training set and a testing set at a proportion of 5: 1 according to a designed rule; building an eight-layer convolution neural network model consistent with an AlexNet structure, modifying the number of neurons of the last layer of the model, and enabling the model to adapt to a gait classification task; Carrying out the initialization of the parameters of a trained AlexNet model through the former seven layers of the model, and carrying out the random initialization of the last layer of the model; training the convolution neural network through the training set, and enabling the convolution neural network to effectively complete the gait recognition. The method is higher in robustness, can achieve the more effective recognition of the identity of a person under the condition of various types of concomitant variables, and can remarkably reduce the calculation resources and time of a training model.

Description

technical field [0001] The invention relates to computer vision and pattern recognition, in particular to a gait recognition method based on a deep neural network. Background technique [0002] In the existing gait recognition technology, the gait energy map as a gait feature descriptor has an ideal recognition effect and is simple to calculate. Some improvements have been made in the gait energy map and a Gaussian coefficient is added to it to obtain the gait Gauss The recognition effect of the graph is significantly better than that of the gait energy graph. Using the gait Gauss graph as a gait feature to select a suitable classifier can effectively identify the identity of a person. However, this kind of recognition is difficult to achieve the desired effect when the viewing angle span is large (more than 36 degrees) or the clothing and carrying conditions change significantly. [0003] Convolutional neural network is a multi-layer perceptron specially designed to recogn...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the recognition performance of the existing gait recognition technology is greatly affected by many covariates, the present invention proposes a gait recognition method based on a deep neural network. Complicated background conditions show good robustness, and the recognition rate is significantly improved under the influence of complex scenes and multiple covariates; the fine-tuning of the trained network is used to effectively solve the problem of insufficient data and save calculations volume and running time

Images