Advertisement
Published in:
2020 | OriginalPaper | Chapter
Using CNN and Channel Attention Mechanism to Identify Driver’s Distracted Behavior
Authors : Lu Ye, Cheng Chen, Mingwei Wu, Samuel Nwobodo, Annor Arnold Antwi, Chido Natasha Muponda, Koi David Ernest, Rugamba Sadam Vedaste
Published in: Transactions on Edutainment XVI
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
The driver’s distracted attention will cause a huge safety hazard to the traffic. In different types of distraction, it is illegal to make phone calls and smoke while driving, which will be fined in China. In order to solve this problem, a method of driver’s distracted behavior detection based on channel attention convolution neural network is proposed. SE module is added to the Xception network, which can distinguish the importance of different feature channels and enhance the expression ability of the network. SE module mainly assigns different weights to features to enhance more important features and suppress less influential features. The experiment uses Xception and SE-Xception for comparison. The experimental results show that the accuracy of SE-Xception is 92.60%, which has a good performance for the distracted driving behavior detection of drivers.