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 2020-05-29 15:10
Distant pedestrian detection from nighttime NIR video by moving-region zooming
Atsuki Hiramatsu, Yusuke Kameda (TUS), Hiroshi Ikeoka (Fukuyama Univ.), Takayuki Hamamoto (TUS)
Abstract (in Japanese) (See Japanese page) 
(in English) Highly accurate pedestrian detection technology is required for functions such as automatic emergency braking, which are important in preventing traffic accidents and reducing damage. Visible light cameras are generally used for pedestrian detection. However, there is a problem that clear images cannot be captured in low-light environments such as at night. On the other hand, the combination of near-infrared (NIR) camera and near-infrared light can capture clear images even in low-light environments. Thus it is expected to develop the pedestrian detection technology using them. Previous studies on pedestrian detection from NIR images include methods using deep convolutional neural networks (CNN) with real-time computation and high detection accuracy. However, it is difficult to detect distant pedestrians in small areas of the image because convolution and pooling are combined in the previous study CNN. In this paper, we propose a method to equalize the apparent size of distant pedestrians as preprocessing of CNN input. Specifically, it extracts the time-varying pixels from the input consecutive frames and adaptively chooses the pixel blocks to be enlarged. We aim to improve the accuracy of pedestrian detection in the postprocessing by inputting the enlarged image to the CNN.
Keyword (in Japanese) (See Japanese page) 
(in English) vehicle-mounted camera / pedestrian detection / near-infrared image / CNN / / / /  
Reference Info. ITE Tech. Rep.
Paper #  
Date of Issue  
ISSN Print edition: ISSN 1342-6893  Online edition: ISSN 2424-1970
Download PDF

Conference Information
Committee IEICE-MI IEICE-IE IEICE-SIP IEICE-BioX IST ME  
Conference Date 2020-05-28 - 2020-05-29 
Place (in Japanese) (See Japanese page) 
Place (in English) Online 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Image and signal processing/analysis/AI technology, and their application 
Paper Information
Registration To IEICE-IE 
Conference Code 2020-05-MI-IE-SIP-BioX-IST-ME 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Distant pedestrian detection from nighttime NIR video by moving-region zooming 
Sub Title (in English)  
Keyword(1) vehicle-mounted camera  
Keyword(2) pedestrian detection  
Keyword(3) near-infrared image  
Keyword(4) CNN  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Atsuki Hiramatsu  
1st Author's Affiliation Tokyo University of Science (TUS)
2nd Author's Name Yusuke Kameda  
2nd Author's Affiliation Tokyo University of Science (TUS)
3rd Author's Name Hiroshi Ikeoka  
3rd Author's Affiliation Fukuyama University (Fukuyama Univ.)
4th Author's Name Takayuki Hamamoto  
4th Author's Affiliation Tokyo University of Science (TUS)
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
Date Time 2020-05-29 15:10:00 
Presentation Time 20 
Registration for IEICE-IE 
Paper #  
Volume (vol) ITE-44 
Number (no)  
Page  
#Pages ITE- 
Date of Issue  


[Return to Top Page]

[Return to ITE Web Page]


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