IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2023-03-02 17:30
A Study on Data Augmentation by Pixel Value Transformation for Product Image Retrieval DNN
Yu Okamoto, Masaki Kishibe, Toshikazu Wada (Wakayama Univ.) PRMU2022-94 IBISML2022-101
Abstract (in Japanese) (See Japanese page) 
(in English) We are developing a product image retrieval DNN for supporting shelving allocation tasks. To train the DNN, product images photographed on various shelfs are necessary for each product. Although sample product images have been taken under an ideal condition for each product, it is very expensive to take product images on various shelves. For solving this problem, we apply luminance transformation of sample product images. This transformation is done by luminance Look Up Table (LUT) estimated by comparing luminance histograms of sample image and images of real product on the shelf. However, there are various LUTs, and it is impractical to use all LUTs for data augmentation. In this report, we propose two sampling methods for reducing the number of LUTs. In the experiment, we compared these two sampling methods in an actual product image retrieval system.
Keyword (in Japanese) (See Japanese page) 
(in English) Product Image Retrieval DNN / Data Augmentation / Color Transfer / / / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 404, PRMU2022-94, pp. 181-186, March 2023.
Paper # PRMU2022-94 
Date of Issue 2023-02-23 (PRMU, IBISML) 
ISSN Online edition: ISSN 2432-6380
Copyright
and
reproduction
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF PRMU2022-94 IBISML2022-101

Conference Information
Committee PRMU IBISML IPSJ-CVIM  
Conference Date 2023-03-02 - 2023-03-03 
Place (in Japanese) (See Japanese page) 
Place (in English) Future University Hakodate 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To PRMU 
Conference Code 2023-03-PRMU-IBISML-CVIM 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A Study on Data Augmentation by Pixel Value Transformation for Product Image Retrieval DNN 
Sub Title (in English)  
Keyword(1) Product Image Retrieval DNN  
Keyword(2) Data Augmentation  
Keyword(3) Color Transfer  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Yu Okamoto  
1st Author's Affiliation Wakayama University (Wakayama Univ.)
2nd Author's Name Masaki Kishibe  
2nd Author's Affiliation Wakayama University (Wakayama Univ.)
3rd Author's Name Toshikazu Wada  
3rd Author's Affiliation Wakayama University (Wakayama Univ.)
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 2023-03-02 17:30:00 
Presentation Time 10 minutes 
Registration for PRMU 
Paper # PRMU2022-94, IBISML2022-101 
Volume (vol) vol.122 
Number (no) no.404(PRMU), no.405(IBISML) 
Page pp.181-186 
#Pages
Date of Issue 2023-02-23 (PRMU, IBISML) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan