A method, device and equipment for detecting fatigue driving in low light and strong light environments

By using the Enhancement MNet module for image correction and face key point detection in low light and strong light environments, the problem of low accuracy of the driver fatigue state detection algorithm based on computer vision under complex lighting conditions is solved, and higher detection accuracy and traffic safety guarantee are achieved.

CN119741682APending Publication Date: 2025-04-01XIAMEN INTRETECH +1
0 Cites 0 Cited by

Patent Information

Application Number
CN202411571874.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-06
Publication Date
2025-04-01

AI Technical Summary

Technical Problem

In low or strong light environments, the driver's fatigue state detection algorithm based on computer vision has low accuracy, making it difficult to identify the driver's fatigue state in a timely, accurate and fast manner.

Method used

The driver's face image is obtained in real time and the image correction model based on the Enhancement MNet module is processed to improve image quality. Then, the face detection and key point detection models are used to identify key information, determine the fatigue characteristics of the eyes, mouth and head, and perform fatigue driving detection.

Benefits of technology

It improves image quality and detection accuracy in low and strong light environments, and can more effectively identify the driver's fatigue state, including fatigue driving state, instantaneous fatigue driving state and long-term fatigue driving state, providing more comprehensive traffic safety guarantees.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention discloses a fatigue driving detection method, device and equipment in low light and strong light environment, which comprises: obtaining a driver's face image in real time, inputting the face image into a preset image correction model to obtain a corrected image, wherein the face image is an abnormal exposure image, and the image correction model is constructed based on the Enhancement MNet module; using a face detection model and a face key point detection model to identify the face area and face key points in the corrected image; determining eye fatigue features and mouth fatigue features according to face key points, and determining head fatigue features according to the face area; performing fatigue driving detection on the driver according to the eye fatigue features, mouth fatigue features and / or head fatigue features to obtain a fatigue driving detection result, wherein the fatigue driving detection result includes one of a fatigue driving state, an instantaneous fatigue driving state and a long-term fatigue driving state. The accuracy of detection is improved, and a strong guarantee is provided for traffic safety.
Need to check novelty before this filing date? Find Prior Art