ITE Technical Group Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2023-09-07 09:50
Improving Performance of Convolutional Neural Network-Based Driver Behavior Recognition
Shengbiao Wang, Koji Iwano (Tokyo City Univ.)
Abstract (in Japanese) (See Japanese page) 
(in English) This study investigates the automatic recognition of driver behaviors using images captured by in-vehicle cameras for the purpose of realizing distracted driving detection. While existing methods for behavior recognition utilizing standard convolutional neural networks (CNN) have been explored, their recognition performance has been less than satisfactory. To address this issue, we propose two approaches aimed at improving the performance: (1) the incorporation of an attention mechanism designed to emphasize distinctive regions, and (2) the integration of a neural network utilizing skeleton information extracted by a skeleton detection model as input for behavior recognition, in conjunction with the CNN. Experiments on classification of 10 distinct behaviors using a dataset of approximately 14,000 images from 44 drivers shows that the combined implementation of these two methods improves the accuracy by 3.0 to 6.5%; the maximum recognition accuracy of 86.8% is achieved.
Keyword (in Japanese) (See Japanese page) 
(in English) Driver behavior recognition / Image recognition / Convolutional neural network / Attention mechanism / Skeleton information / / /  
Reference Info. ITE Tech. Rep., vol. 47, no. 25, ME2023-90, pp. 7-12, Sept. 2023.
Paper # ME2023-90 
Date of Issue 2023-08-31 (ME) 
ISSN Online edition: ISSN 2424-1970
Download PDF

Conference Information
Committee ME IEICE-EMM IEICE-IE IEICE-LOIS IEE-CMN IPSJ-AVM  
Conference Date 2023-09-07 - 2023-09-07 
Place (in Japanese) (See Japanese page) 
Place (in English) Osaka Metropolitan Univ. 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To ME 
Conference Code 2023-09-ME-EMM-IE-LOIS-CMN-AVM 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Improving Performance of Convolutional Neural Network-Based Driver Behavior Recognition 
Sub Title (in English)  
Keyword(1) Driver behavior recognition  
Keyword(2) Image recognition  
Keyword(3) Convolutional neural network  
Keyword(4) Attention mechanism  
Keyword(5) Skeleton information  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Shengbiao Wang  
1st Author's Affiliation Tokyo City University (Tokyo City Univ.)
2nd Author's Name Koji Iwano  
2nd Author's Affiliation Tokyo City University (Tokyo City Univ.)
3rd Author's Name  
3rd Author's Affiliation ()
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2023-09-07 09:50:00 
Presentation Time 20 minutes 
Registration for ME 
Paper # ME2023-90 
Volume (vol) vol.47 
Number (no) no.25 
Page pp.7-12 
#Pages
Date of Issue 2023-08-31 (ME) 


[Return to Top Page]

[Return to ITE Web Page]


The Institute of Image Information and Television Engineers (ITE), Japan