Intelligent Fatigue Detection Method in Industrial System Based on Facial Multi-feature Fusion

Abstract

With the development of science and technology, the complexity of industrial production system has been significantly improved, and higher requirements have been put forward for human safety. Aiming at the contradiction between the intrusiveness and the accuracy of fatigue behavior detection in human factor safety, a multi-feature fusion intelligent fatigue state detection method was proposed. Firstly, the image recognition algorithm was used to accurately obtain the information of facial expression feature points. By analyzing the dynamic characteristics of all the feature points, the facial fatigue features were extracted. Secondly, an experimental scheme was designed to collect and process facial expression feature data, and a dynamic marker model of fatigue characteristics was constructed to form a fatigue index that directly reflected the fatigue degree of the working process. Finally, the neural network regression model was established to fit the characteristic data. Through the comparative analysis of different schemes and various evaluation indexes, the rationality of the fitting results of this method was proved. The experimental results show that this method can reasonably quantify the fatigue index of people under different conditions, and realize the intelligent detection of facial fatigue state.