Method and device for detecting fatigue state based on audio and video
A fatigue state and fatigue detection technology, applied in neural learning methods, character and pattern recognition, instruments, etc., can solve problems such as rear-end collisions and inability to effectively remind drivers.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0099] Such as figure 1 As shown, the audio-video-based fatigue state detection method provided in Embodiment 2 of the present application includes:
[0100] S101: Driver identity recognition: confirm the identity of the target driver in the local driver file database through face recognition technology, if there is no corresponding personnel information, create a new driver file;
[0101] S102: Target driver data collection: during the driving process of the vehicle, collect the audio-video information and physiological information of the target driver during driving, and extract the data related to fatigue detection in the audio-video information and physiological information, and obtain Fatigue detection data; the audio and video information includes: facial video information, overall torso video information and audio information;
[0102] S103: Target driver fatigue state detection: analyze the current state of the target driver according to the fatigue detection data and...
Embodiment 2
[0172] Such as figure 1 As shown, the audio-video-based fatigue state detection method provided in Embodiment 2 of the present application includes:
[0173] S101: Driver identity recognition: confirm the identity of the target driver in the local driver file database through face recognition technology, if there is no corresponding personnel information, create a new driver file;
[0174] In some embodiments, the sample feature sequence is generated by analyzing and extracting the digital features of the driver's face image, and the sample feature sequence is compared with the feature sequence in the local driver file database to identify the driver's identity and assist subsequent based on Fatigue state detection of audio and video;
[0175] S102: Data collection of the target driver: During the driving process of the vehicle, the audio and video information and physiological information of the target driver during the driving process are collected through the face collecti...
Embodiment 3
[0294] Such as Figure 8 As shown, embodiment 3 provides a kind of fatigue state detection device based on audio and video, comprising:
[0295] Identity verification module, information collection module, fatigue analysis module, safety rating module and alarm takeover module; the identity verification module is connected to the information collection module, the information collection module is connected to the fatigue analysis module, and the fatigue analysis module The module is connected to the safety rating module, and the safety rating module is connected to the warning takeover module;
[0296] The identity verification module confirms the identity of the target driver in the local driver file database through face recognition technology, and if there is no corresponding personnel information, a new driver file is created;
[0297] In some embodiments, the device for collecting the face information of the driver can be a camera or other video collector, etc., and the ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


