Method and device for establishing abnormal driving monitoring model as well as storage medium

Abstract

The invention discloses a method for establishing an abnormal driving monitoring model. The method comprises the following steps: acquiring a plurality of samples, wherein each sample comprises firstmodal data and objective results of the first modal data during a same driving time period collected by a plurality of collectors; inputting the first modal data of each sample into a first abnormal driving monitoring model for training, and outputting first merging data merging the single first modal data and second merging data merging all the first modal data of each sample; calculating a global loss function of the first abnormal driving monitoring model according to the first merging data, the second merging data and the objective results; and updating the weight of the first abnormal driving monitoring model according to the global loss function by using a back propagation mechanism and an Adam algorithm. The invention also discloses a device for establishing the abnormal driving monitoring model and a storage medium. By adopting the method, the device and the storage medium, multi-level accurate merging of multi-modal data can be realized, so that the abnormal driving monitoringmodel can accurately monitor various abnormal driving states.

Description

technical field [0001] The invention relates to the technical field of traffic analysis, in particular to a method, device and storage medium for establishing an abnormal driving monitoring model. Background technique [0002] In recent years, due to the impact of life pressure and physical and mental tension, traffic accidents caused by abnormal driving conditions such as fatigue driving, distraction, emotional driving, and sudden illness have gradually increased. Due to changes in age, physical or mental health, emotions, etc., even a good driver may not be able to maintain his original good driving state for a long time, but it is difficult for the driver himself to realize this gradual attenuation or subside. Therefore, monitoring the driver's driving behavior and giving an alert to abnormal driving behavior can improve the driver's driving ability and reduce his driving load, coordinate the relationship between the driver, the vehicle and the traffic environment, and e...

AI Technical Summary

Problems solved by technology

[0004] Existing abnormal driving monitoring methods currently have a single data source: they basically use a single modal modeling monitoring, such as driver monitoring video or mobile phone built-in motion sensors to identify driving status

Even if multi-modal monitoring is used, the fusion level of multi-modal data is simple and the effect is poor. Existing sensor data fusion methods often only use early fusion or decision-making fusion, and it is impossible to obtain the optimal fusion strategy integratedly.

Moreover, the abnormal state monitoring target is single. The existing abnormal driving states are generally divided into distraction, fatigue, etc., and the existing methods can only identify a single abnormal state, and cannot obtain multiple abnormal driving monitoring results.

Images