| Title |
A Driving Attention Detection Method Based On Head Pose |
| ID_Doc |
1565 |
| Authors |
Li Y.; Li J.; Jiang X.; Gao C.; Zhang T. |
| Year |
2019 |
| Published |
Proceedings - 2019 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Internet of People and Smart City Innovation, SmartWorld/UIC/ATC/SCALCOM/IOP/SCI 2019 |
| DOI |
http://dx.doi.org/10.1109/SmartWorld-UIC-ATC-SCALCOM-IOP-SCI.2019.00124 |
| Abstract |
Head pose is an important indicator of driving attention detection. During driving, head pose, including head position and head movement, can infer the driver's attention. This paper presents a novel method for collecting driver's head pose information using a built-in accelerometer and gyroscope head-mounted inertial sensor. In our experimental study, we designed 10 scenes that are easy to distract from driving. And five subjects were asked to wear a head-mounted inertial sensor to drive the driving simulator. This driving simulator is equipped with real driving conditions such as brakes, steering wheels, accelerators and so on. While driving, subjects need to complete the designed driving scene in order. Subsequently, We perform pre-processing such as Savitzky-Golay filtering and windowing on data collected by inertial sensors with built-in accelerometers and gyroscopes. The time domain and frequency domain features of the data are then extracted in the corresponding window. Finally, we designed a random forest model to detect driving attention. Our simulation experiments show that our proposed method of collecting data using the built-in accelerometer and gyroscope's head-mounted sensor can achieve higher precision, recall and F1-Score. © 2019 IEEE. |
| Author Keywords |
Driving Attention Detection; Head Pose; Machine Learning; Wearable Sensor |
