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 2025-03-21 10:45
Defense Method Against Adversarial Example Attacks using Thermal Noise of a CMOS Image Sensor
Yuki Rogi, Kota Yoshida, Tatsuya Oyama, Takeshi Fujino, Shunsuke Okura (Ritsumeikan Univ.)
Abstract (in Japanese) (See Japanese page) 
(in English) With the rise of IoT technology, edge AI has become increasingly important, but it is vulnerable to adversarial example (AE) attacks. These attacks involve small, nearly invisible perturbations that can mislead convolutional neural networks (CNNs) into misclassifying images. One defense strategy is to introduce disturbance noise to input images to counter these attacks. However, this approach is less effective when the noise is predictable, such as when generated by arithmetic algorithms.
This paper proposes a novel defense method using unpredictable noise generated by thermal fluctuations from an image sensor. The thermal noise is integrated into the input images, and a post-processing method is employed to further enhance the defense. The method utilizes 5-bit disturbance noise generated from the image sensor to defend against AE attacks. To evaluate the randomness of the generated noise, the NIST 800-22 random number test was applied, and all items passed with high probability, confirming the high randomness of the noise. The proposed defense approach also demonstrated a 24% improvement in performance, indicating that unpredictable thermal noise and post-processing significantly enhance the effectiveness of defending against AE attacks on edge AI systems.
Keyword (in Japanese) (See Japanese page) 
(in English) CMOS image sensors / IoT / edge AI / hardware security / Image recognition / true random numbers / /  
Reference Info. ITE Tech. Rep., vol. 49, no. 13, IST2025-10, pp. 5-6, March 2025.
Paper # IST2025-10 
Date of Issue 2025-03-14 (IST) 
ISSN Online edition: ISSN 2424-1970
Download PDF

Conference Information
Committee IST  
Conference Date 2025-03-21 - 2025-03-21 
Place (in Japanese) (See Japanese page) 
Place (in English) Kikai-Shinko-Kaikan Bldg. 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To IST 
Conference Code 2025-03-IST 
Language English 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Defense Method Against Adversarial Example Attacks using Thermal Noise of a CMOS Image Sensor 
Sub Title (in English)  
Keyword(1) CMOS image sensors  
Keyword(2) IoT  
Keyword(3) edge AI  
Keyword(4) hardware security  
Keyword(5) Image recognition  
Keyword(6) true random numbers  
Keyword(7)  
Keyword(8)  
1st Author's Name Yuki Rogi  
1st Author's Affiliation Ritsumeikan University (Ritsumeikan Univ.)
2nd Author's Name Kota Yoshida  
2nd Author's Affiliation Ritsumeikan University (Ritsumeikan Univ.)
3rd Author's Name Tatsuya Oyama  
3rd Author's Affiliation Ritsumeikan University (Ritsumeikan Univ.)
4th Author's Name Takeshi Fujino  
4th Author's Affiliation Ritsumeikan University (Ritsumeikan Univ.)
5th Author's Name Shunsuke Okura  
5th Author's Affiliation Ritsumeikan University (Ritsumeikan Univ.)
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 ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2025-03-21 10:45:00 
Presentation Time 20 minutes 
Registration for IST 
Paper # IST2025-10 
Volume (vol) vol.49 
Number (no) no.13 
Page pp.5-6 
#Pages
Date of Issue 2025-03-14 (IST) 


[Return to Top Page]

[Return to ITE Web Page]


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