Driver fatigue level recognition system based on bimodal feature fusion

Abstract

A driver fatigue level recognition system based on bimodal feature fusion comprises an image acquisition module, a face behavior recognition module and a driving recorder host, the image acquisition module and the driving recorder host are connected with the face behavior recognition module, and the driving recorder host is further connected with a CAN bus of a vehicle. The image acquisition module is used for acquiring a face dynamic image of a driver; the face behavior recognition module is used for recognizing eye closing features and mouth opening and closing features from the collected face dynamic images; the driving recorder host is used for obtaining and analyzing vehicle operation information from a vehicle CAN bus, calculating driving behavior characteristics through the vehicleoperation information, and analyzing the fatigue level of a driver through fusion of eye closing characteristics, mouth opening and closing characteristics and the driving behavior characteristics. According to the system, the limitation of a single information source is overcome, the correlation and complementarity of each information source are fully considered, and the fatigue grade analysis ismore accurate.

Description

technical field [0001] The invention relates to the field of fatigue driving prediction, in particular to a driver fatigue level recognition system based on dual-mode feature fusion. Background technique [0002] According to data from the National Bureau of Statistics, the number of traffic accidents in my country has exceeded 120,000 in the past five years, among which truck traffic accidents are particularly serious. People are injured, the accident rate of trucks is higher than that of ordinary motor vehicles, and the losses caused are also higher than the average level. Among them, traffic accidents caused by fatigue driving cause great loss to people's life and property safety every year. Various studies show that in all road accidents, about 20% are related to fatigue, and on some roads, it is as high as 50%. The results of a sample survey of freight vehicle drivers by relevant departments in our country show that: 84% of freight vehicle drivers drive for more than 8 ...

AI Technical Summary

Problems solved by technology

At present, the single fatigue driving detection method on the market has many disadvantages. For example, when analyzing driving behavior characteristics, there are many types of bus data and a large amount of data. It is extremely difficult to identify and extract the characteristic behavior of fatigue driving, and each driver has different driving habits. This leads to the inability to intelligently recognize; using the method of facial state recognition, it can effectively recognize the driver's actions such as closing eyes, yawning, and making a phone call, but it cannot work normally when the light environment in the cabin is poor or the camera is blocked

Images