A method for detecting human fatigue based on improved cascade convolution neural network

Abstract

The invention claims a method for detecting human body fatigue based on an improved cascade convolution neural network. The method comprises the following steps: skin color detection of an image witha face is combined with pre-training CNN classifier to identify a face region; A cascade neural network structure is designed to detect human eyes and their feature points. The first-order neural network is composed of gray-scale integral projection rough localization and multi-task convolutional neural network (G-RCNN) network is used to detect and locate human eyes, and two-level network (PCNN)is used to segment human eyes and predict feature points by using parallel sub-convolution system. Using the characteristic points of human eyes to calculate the degree of opening and closing of humaneyes to recognize the current state of human eyes; 4, jud that fatigue state of the human body according to the PERCLOS criterion; The invention can obtain high identification rate and has strong robustness to illumination and random noise.

Description

technical field [0001] The invention belongs to the field of image processing and pattern recognition, in particular to a human body fatigue detection method based on an improved cascaded convolutional neural network. Background technique [0002] Fatigue refers to the state where the body's labor efficiency tends to decline due to excessively long or overly intense physical or mental work under certain environmental conditions. Mental fatigue is the origin of many diseases. Fatigue not only endangers people's physical and mental health, but also brings major safety hazards to social production and life, especially in high-risk operations such as power industry, construction high-altitude operations, vehicle driving, aerospace, and large-scale complex industries. Distractions, sluggish reactions or poor coordination caused by fatigue can lead to serious production accidents. In terms of car driving, as the total number of cars in China has increased year by year, traffic ac...

AI Technical Summary

Problems solved by technology

Deng et al. used the skin color model combined with the layout of the three courts and five eyes of the face to locate the human eye, and used the size of the integral projection area of ​​the human eye to identify the state of the human eye. Although this method is simple in algorithm, the accuracy of positioning is less affected by the environment. Large, and because the proportion of the human eye area in the image is very small, the recognition accuracy of the human eye state using the integral projection is low

Li Xiang et al. use the moment feature of the image to calculate the similarity between the Zernike moment feature vector of the human eye template and the face area to be recognized, and select the area with the largest similarity as the human eye area. This method is similar to the template matching method, although it can reduce The impact of the small environment on the detection results, but the calculation is relatively large, and the results are affected by the selected human eye template

Images