Driver Physiological Drowsiness Behavior Detection and Analysis using Vision-based Multimodal Features for Driving Safety Clemson University

Format:

Book

Conference/Event

Author/Creator:

Li, Rui, author.

Contributor:

Brand, Howard

Gopinath, Aditya

Kamarajugadda, Srivatsav

Li, Bing

Wang, Weitian

Yang, Liang

Conference Name:

WCX SAE World Congress Experience (2020-04-21 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2020

Summary:

Driving safety has always been a fundamental concern in transportation systems. Driving inattention caused by drowsiness has been a significant reason for vehicle crash accidents according to United States Traffic Safety Culture Index report, and there is an essential need to improve assistance driving safety by understanding the driver behaviors. Towards real-time drowsy driving monitoring, we propose an in-vehicle driver assistant system to monitor driver states for drowsiness behavior recognition and analysis. First, an infrared camera is deployed inside the vehicle to capture the driver's facial and head information, in which scenarios, the driver is allowed to wear glasses or sunglasses during driving. Second, vision-based multimodal features, facial landmarks and head pose are extracted efficiently by the ensemble of regression trees based facial landmarks estimation method and a convolutional neural network (CNN) recognition model. Finally, an extreme learning machine (ELM) model is proposed to fuse the facial landmark, recognition model and pose orientation for drowsiness detection. The system gives promptly warning to the driver once a drowsiness event is detected. The proposed machine learning recognition model are trained in drowsy driving dataset and validated in both public dataset and field tests. Comparing to the end-to-end CNN recognition model, the proposed multimodal fusion with ELM detection model allows faster and more accurate detection with minimal intervention. The experimental result demonstrates that the proposed system is able to provide prompt and effective drowsy driving alerts under various light conditions to augment driving safety. The eld test demo of this research is available at https://youtu.be/Gag53nMqHdc

Notes:

Vendor supplied data

Publisher Number:

2020-01-1211

Access Restriction:

Restricted for use by site license