東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Compressed-Domain Deep Learning With Application to Image Recognition and Universal Adversarial Attack.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Compressed-Domain Deep Learning With Application to Image Recognition and Universal Adversarial Attack./
作者:	Deng, Yingpeng.
面頁冊數:	1 online resource (129 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-02, Section: B.
Contained By:	Dissertations Abstracts International85-02B.
標題:	Electrical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30572098click for full text (PQDT)
ISBN:	9798380110105

Compressed-Domain Deep Learning With Application to Image Recognition and Universal Adversarial Attack.
Deng, Yingpeng.

Compressed-Domain Deep Learning With Application to Image Recognition and Universal Adversarial Attack. - 1 online resource (129 pages)

Source: Dissertations Abstracts International, Volume: 85-02, Section: B.

Thesis (Ph.D.)--Arizona State University, 2023.

Includes bibliographical references

Researchers have shown that the predictions of a deep neural network (DNN) for an image set can be severely distorted by one single image-agnostic perturbation, or universal perturbation, usually with an empirically fixed threshold in the spatial domain to restrict its perceivability. However, current universal perturbations have limited attack ability, and more importantly, limiting the perturbation's norm in the spatial domain may not be a suitable way to restrict the perceptibility of universal adversarial perturbations. Besides, the effects of such attacks on DNN-based texture recognition have yet to be explored.Learning-based image compression was shown to achieve a competitive performance with state-of-the-art transform-based codecs. This motivated the development of learning-based image compression systems targeting both humans and machines. Also, the learning-based compressed-domain representations can be utilized to perform computer vision tasks directly in the compressed domain.In the context of universal attacks, a novel method is proposed to compute more effective universal perturbations via enhanced projected gradient descent on targeted classifiers. The perturbation is optimized by accumulating small updates on perturbed images consecutively. Performance results show that the proposed adversarial attack method can achieve much higher fooling rates as compared to state-of-the-art universal attack methods.In order to reduce the perceptibility of universal attacks without compromising their effectiveness, a frequency-tuned universal attack framework is proposed to adopt JND thresholds to guide the perceptibility of universal adversarial perturbations. The proposed frequency-tuned attack method can achieve cutting-edge quantitative results, realize a good balance between perceptibility and effectiveness in terms of fooling rate on both natural and texture image datasets.In the context of compressed-domain image recognition, a novel feature adaptation module integrating a lightweight attention model is proposed to adaptively emphasize and enhance the key features within the extracted channel-wise information. Also, an adaptation training strategy is designed to utilize the pretrained pixel-domain weights. The obtained performance results show that the proposed compressed-domain classification model can distinctly outperform the existing compressed-domain classifiers, and that it can also yield similar accuracy results with a much higher computational efficiency as compared to the decoded image trained pixel-domain models.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798380110105Subjects--Topical Terms:

649834
Electrical engineering.
Subjects--Index Terms:

Deep learningIndex Terms--Genre/Form:

542853
Electronic books.

Compressed-Domain Deep Learning With Application to Image Recognition and Universal Adversarial Attack.
LDR:04008nmm a2200409K 4500 001 2363234
005 20231116093832.5
006 m o d
007 cr mn ---uuuuu
008 241011s2023 xx obm 000 0 eng d
020 $a 9798380110105
035 $a (MiAaPQ)AAI30572098
035 $a AAI30572098
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Deng, Yingpeng. $3 3703988
245 1 0 $a Compressed-Domain Deep Learning With Application to Image Recognition and Universal Adversarial Attack.
264 0 $c 2023
300 $a 1 online resource (129 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-02, Section: B.
500 $a Advisor: Karam, Lina J.
502 $a Thesis (Ph.D.)--Arizona State University, 2023.
504 $a Includes bibliographical references
520 $a Researchers have shown that the predictions of a deep neural network (DNN) for an image set can be severely distorted by one single image-agnostic perturbation, or universal perturbation, usually with an empirically fixed threshold in the spatial domain to restrict its perceivability. However, current universal perturbations have limited attack ability, and more importantly, limiting the perturbation's norm in the spatial domain may not be a suitable way to restrict the perceptibility of universal adversarial perturbations. Besides, the effects of such attacks on DNN-based texture recognition have yet to be explored.Learning-based image compression was shown to achieve a competitive performance with state-of-the-art transform-based codecs. This motivated the development of learning-based image compression systems targeting both humans and machines. Also, the learning-based compressed-domain representations can be utilized to perform computer vision tasks directly in the compressed domain.In the context of universal attacks, a novel method is proposed to compute more effective universal perturbations via enhanced projected gradient descent on targeted classifiers. The perturbation is optimized by accumulating small updates on perturbed images consecutively. Performance results show that the proposed adversarial attack method can achieve much higher fooling rates as compared to state-of-the-art universal attack methods.In order to reduce the perceptibility of universal attacks without compromising their effectiveness, a frequency-tuned universal attack framework is proposed to adopt JND thresholds to guide the perceptibility of universal adversarial perturbations. The proposed frequency-tuned attack method can achieve cutting-edge quantitative results, realize a good balance between perceptibility and effectiveness in terms of fooling rate on both natural and texture image datasets.In the context of compressed-domain image recognition, a novel feature adaptation module integrating a lightweight attention model is proposed to adaptively emphasize and enhance the key features within the extracted channel-wise information. Also, an adaptation training strategy is designed to utilize the pretrained pixel-domain weights. The obtained performance results show that the proposed compressed-domain classification model can distinctly outperform the existing compressed-domain classifiers, and that it can also yield similar accuracy results with a much higher computational efficiency as compared to the decoded image trained pixel-domain models.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Electrical engineering. $3 649834
650 4 $a Scandinavian studies. $3 3176828
650 4 $a Computer engineering. $3 621879
650 4 $a Information technology. $3 532993
653 $a Deep learning
653 $a Image recognition
653 $a Universal adversarial attack
653 $a Deep neural network
653 $a Perturbation
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0544
690 $a 0613
690 $a 0489
690 $a 0464
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a Arizona State University. $b Electrical Engineering. $3 1671741
773 0 $t Dissertations Abstracts International $g 85-02B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30572098 $z click for full text (PQDT)