A deep learning-based fatigue detection method, system and computer equipment

Abstract

The invention discloses a fatigue detection method, system and computer equipment based on deep learning. The method includes: acquiring a current frame video stream picture; extracting the face of the current frame video stream picture through a face tracking algorithm based on a first deep learning model Position information and human eye position information; the human face position information includes the position information of the human face frame; the human eye position information includes the feature point position information of the left and right corners of each eye; based on the human eye position information, The eye state recognition algorithm based on the second deep learning model extracts the position information of the landmarks on the upper and lower edges of the eyes; calculates the degree of eye opening according to the position information of the landmarks on the upper and lower edges of the eyes, and judges the fatigue state. The invention can quickly track the position of the human face and extract the feature points of the upper and lower edges of the eyes, and has higher recognition accuracy, so that the fatigue state can be detected quickly and accurately.

Description

technical field [0001] The invention relates to the field of automobile driving assistance terminals, in particular to a fatigue detection method, system and computer equipment based on deep learning. Background technique [0002] With the rapid development of society, traffic vehicles on the road have shown a blowout growth, and traffic accidents have also occurred frequently. The most important reason for traffic accidents is that the driver is tired of driving and reacts too slowly in the process of driving. Therefore, it is very important to detect the driver's fatigue state when driving in real time, and reminding the driver in time can often greatly reduce the occurrence of traffic accidents. At present, there are various methods of fatigue detection. In terms of visual recognition methods, there are the following: [0003] (1) Fatigue detection algorithm processed by pure image algorithm: Adaboost face detection algorithm detects faces; horizontal projection and ver...

AI Technical Summary

Problems solved by technology

Advantages: fast speed, better recognition effect under good lighting conditions; Disadvantages: unable to adapt to complex lighting conditions, so the application scene is single

Advantages: The speed is also faster, and the robustness of the model is enhanced; Disadvantages: It is impossible to distinguish the degree of eye opening, and all small eyes are considered closed eyes

The advantages are: accurate face positioning, accurate fatigue classification; disadvantages: MTCNN face detection speed is relatively slow, and the classification results cannot reflect the state of eye opening in real time

Images