Driver Sleepiness Detection Algorithm Based on Relevance Vector Machine
| Autor: | Tianliu Feng, Pengfei Zhao, Mingjun Liao, Lingxiang Wei |
|---|---|
| Přispěvatelé: | State Key Program of the National Natural Science Foundation of China |
| Rok vydání: | 2021 |
| Předmět: |
lcsh:TE1-450
Computer science 02 engineering and technology driver sleepiness driving simulator pattern recognition relevance vector machine (RVM) traffic safety Relevance vector machine lcsh:TG1-470 lcsh:Bridge engineering 0502 economics and business 0202 electrical engineering electronic engineering information engineering relevance vector machine (rvm) lcsh:Highway engineering. Roads and pavements Civil and Structural Engineering 050210 logistics & transportation Data collection Basis (linear algebra) 05 social sciences Work (physics) Driving simulator Building and Construction Identification (information) Pattern recognition (psychology) 020201 artificial intelligence & image processing State (computer science) Algorithm |
| Zdroj: | The Baltic Journal of Road and Bridge Engineering; Vol 16, No 1 (2021): The Baltic Journal of Road and Bridge Engineering; 118-139 The Baltic Journal of Road and Bridge Engineering, Vol 16, Iss 1, Pp 118-139 (2021) |
| ISSN: | 1822-4288 1822-427X |
| Popis: | Driver sleepiness is one of the most important causes of traffic accidents. Efficient and stable algorithms are crucial for distinguishing nonfatigue from fatigue state. Relevance vector machine (RVM) as a leading-edge detection approach allows meeting this requirement and represents a potential solution for fatigue state detection. To accurately and effectively identify the driver’s fatigue state and reduce the number of traffic accidents caused by driver sleepiness, this paper considers the degree of driver’s mouth opening and eye state as multi-source related variables and establishes classification of fatigue and non-fatigue states based on the related literature and investigation. On this basis, an RVM model for automatic detection of the fatigue state is proposed. Twenty male respondents participated in the data collection process and a total of 1000 datasets of driving status (half of non-fatigue and half of fatigue) were obtained. The results of fatigue state recognition were analysed by different RVM classifiers. The results show that the recognition accuracy of the RVM-driven state classifiers with different kernel functions was higher than 90%, which indicated that the mouth-opening degree and the eye state index used in this work were closely related to the fatigue state. Based on the obtained results, the proposed fatigue state identification method has the potential to improve the fatigue state detection accuracy. More importantly, it provides a scientific theoretical basis for the development of fatigue state warning methods. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|