Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data
Drowsy driving is a major threat to the safety of drivers and road traffic. Accurate and reliable drowsy driving detection technology can reduce accidents caused by drowsy driving. In this study, we present a new method to detect drowsy driving with vehicle sensor data obtained from the steering whe...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/7/2372 |
| id |
doaj-e9a19babeeb84d1ca7e26e4159ab54d9 |
|---|---|
| record_format |
Article |
| spelling |
doaj-e9a19babeeb84d1ca7e26e4159ab54d92021-03-29T23:05:15ZengMDPI AGSensors1424-82202021-03-01212372237210.3390/s21072372Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor DataYongsu Jeon0Beomjun Kim1Yunju Baek2Department of Information Convergence Engineering, Pusan National University, Busan 46241, KoreaDepartment of Smart Software, Yonam Institute of Technology, Jinju 52821, KoreaDepartment of Information Convergence Engineering, Pusan National University, Busan 46241, KoreaDrowsy driving is a major threat to the safety of drivers and road traffic. Accurate and reliable drowsy driving detection technology can reduce accidents caused by drowsy driving. In this study, we present a new method to detect drowsy driving with vehicle sensor data obtained from the steering wheel and pedal pressure. From our empirical study, we categorized drowsy driving into long-duration drowsy driving and short-duration drowsy driving. Furthermore, we propose an ensemble network model composed of convolution neural networks that can detect each type of drowsy driving. Each subnetwork is specialized to detect long- or short-duration drowsy driving using a fusion of features, obtained through time series analysis. To efficiently train the proposed network, we propose an imbalanced data-handling method that adjusts the ratio of normal driving data and drowsy driving data in the dataset by partially removing normal driving data. A dataset comprising 198.3 h of in-vehicle sensor data was acquired through a driving simulation that includes a variety of road environments such as urban environments and highways. The performance of the proposed model was evaluated with a dataset. This study achieved the detection of drowsy driving with an accuracy of up to 94.2%.https://www.mdpi.com/1424-8220/21/7/2372intelligent vehiclesafety systemdriver status monitoringdrowsy driving detectionensemble CNN |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Yongsu Jeon Beomjun Kim Yunju Baek |
| spellingShingle |
Yongsu Jeon Beomjun Kim Yunju Baek Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data Sensors intelligent vehicle safety system driver status monitoring drowsy driving detection ensemble CNN |
| author_facet |
Yongsu Jeon Beomjun Kim Yunju Baek |
| author_sort |
Yongsu Jeon |
| title |
Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data |
| title_short |
Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data |
| title_full |
Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data |
| title_fullStr |
Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data |
| title_full_unstemmed |
Ensemble CNN to Detect Drowsy Driving with In-Vehicle Sensor Data |
| title_sort |
ensemble cnn to detect drowsy driving with in-vehicle sensor data |
| publisher |
MDPI AG |
| series |
Sensors |
| issn |
1424-8220 |
| publishDate |
2021-03-01 |
| description |
Drowsy driving is a major threat to the safety of drivers and road traffic. Accurate and reliable drowsy driving detection technology can reduce accidents caused by drowsy driving. In this study, we present a new method to detect drowsy driving with vehicle sensor data obtained from the steering wheel and pedal pressure. From our empirical study, we categorized drowsy driving into long-duration drowsy driving and short-duration drowsy driving. Furthermore, we propose an ensemble network model composed of convolution neural networks that can detect each type of drowsy driving. Each subnetwork is specialized to detect long- or short-duration drowsy driving using a fusion of features, obtained through time series analysis. To efficiently train the proposed network, we propose an imbalanced data-handling method that adjusts the ratio of normal driving data and drowsy driving data in the dataset by partially removing normal driving data. A dataset comprising 198.3 h of in-vehicle sensor data was acquired through a driving simulation that includes a variety of road environments such as urban environments and highways. The performance of the proposed model was evaluated with a dataset. This study achieved the detection of drowsy driving with an accuracy of up to 94.2%. |
| topic |
intelligent vehicle safety system driver status monitoring drowsy driving detection ensemble CNN |
| url |
https://www.mdpi.com/1424-8220/21/7/2372 |
| work_keys_str_mv |
AT yongsujeon ensemblecnntodetectdrowsydrivingwithinvehiclesensordata AT beomjunkim ensemblecnntodetectdrowsydrivingwithinvehiclesensordata AT yunjubaek ensemblecnntodetectdrowsydrivingwithinvehiclesensordata |
| _version_ |
1724190160507109376 |