Open AccessFatigue detection based on facial feature correction and fusion
Author(s) -
Yongshuo Wang,
Biao Liu,
Hengyang Wang
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2183/1/012022
Subject(s) - computer science , feature extraction , feature (linguistics) , fusion , field (mathematics) , artificial intelligence , interference (communication) , noise (video) , sensor fusion , pattern recognition (psychology) , image (mathematics) , mathematics , computer network , philosophy , linguistics , channel (broadcasting) , pure mathematics
Fatigue driving has always been a major hidden danger that threatens traffic safety. The introduction of fatigue monitoring technology can greatly reduce traffic accidents caused by fatigue driving and produce huge social effects. At present, in the field of vision-based fatigue driving research, the extraction of the driver’s eye or mouth features is often used for judgment. However, the criterion is single and the post-processing is insufficient, and the interference of factors such as noise can easily cause misjudgement and reduce the recognition rate. In order to improve the accuracy of parameter extraction and solve the problem of low recognition rate of a single parameter, this paper proposes a feature extraction algorithm based on Euler angle correction, and adds a multi-feature fusion decision method to enhance the applicability of the algorithm. Experiments prove that the algorithm has higher accuracy after modification and fusion.