Driving behavior classification method and device based on noise reduction diffusion probability model and product
Through the method based on the noise reduction diffusion probability model, the problem of insufficient data volume of EEG signal is solved, the accuracy of driving behavior classification is improved, early detection of driver abnormal behavior is achieved, and the risk of traffic accidents is reduced.
Patent Information
- Application Number
- CN202411441083.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-16
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-10-16
AI Technical Summary
In the prior art, the use of electroencephalogram signals (EEG) for driving behavior analysis has the problem of insufficient data, which limits the accuracy of the driver's behavior model and makes it difficult to effectively predict abnormal behaviors of the driver, such as fatigue or distraction, resulting in The risk of traffic accidents increases.
The method based on the noise reduction diffusion probability model is adopted to enhance EEG characteristics by generating new samples and learning reconstruction errors, improve the accuracy of driving behavior classification, and solve the problem of insufficient sample capacity. The specific steps include: obtaining the sample EEG signal, coding and diffusion processing, inputting the noise reduction diffusion probability model for denoising and reconstruction processing, capturing the representation using the decoder, input classifier for classification, and updating the model parameters.
By enhancing the EEG characteristics, the accuracy of the identification of driving behavior models is improved, and the abnormal reactions of drivers can be detected earlier, reducing the risk of traffic accidents.
Smart Images

Figure CN119004264A_ABST
Abstract
Citation Information
Patent Citations
Method for detecting vigilance state of driver
CN105726046A
Method for recognizing road rage states of drivers on basis of electroencephalography and pulse information
CN109498041A
Driving ability determination method, electronic equipment and storage medium
CN116350190A
Driving intention recognition method based on function connection and graph neural network
CN117076980A
Model training method and device, electronic equipment, storage medium and program product
CN117216546A
Cited By
Automatic driving emergency processing method and system based on brain-computer interface
CN121388816A
