Jiaying Li¹, Xiujuan Wang^1,*, Shuhan Han², Liya Xu¹, Changxing Wang¹

CMC-Computers, Materials & Continua, Vol.89, No.1, 2026, DOI:10.32604/cmc.2026.082631 - 13 August 2026

Abstract Deep neural networks are widely applied in computer vision tasks but remain highly vulnerable to adversarial attacks. Tiny and imperceptible perturbations can cause severe model misclassification. Most existing defense methods improve robustness but significantly reduce model accuracy on clean examples. To address this issue, we propose a defense framework combining pixel value transformation and spatial transformation. The proposed method divides the input image into two complementary regions. Feature compression is applied to one region to reduce model sensitivity to subtle perturbations. Intense reversible pixel transformation is applied to the other region to disrupt the spatial… More >

Li Peng^1,2, Xiangbing Li^1,2, Kun Zou¹, Yong Liu^1,2,*, Haibo Huang¹

CMC-Computers, Materials & Continua, Vol.88, No.3, 2026, DOI:10.32604/cmc.2026.082629 - 23 July 2026

Abstract In recent years, the transferability of adversarial examples has attracted significant attention. To improve the effectiveness of black-box attacks, a frequency-domain decay constraint is introduced, inspired by weight decay and regularization techniques commonly employed during model training. By treating adversarial perturbations as inputs in an optimization process, this constraint aims to mitigate the excessive reliance on low-frequency components during adversarial example generation, thereby enhancing transferability. Fourier heatmaps are utilized to analyze the sensitivity of input samples, enabling a decomposition of the frequency spectrum into low-frequency and high-frequency components. Based on this analysis, low-frequency attenuation is More >

Sudhir Kumar Pandey¹, Jian-Xun Mi^1,*, Zahid Ullah², Mona Jamjoom³

CMC-Computers, Materials & Continua, Vol.87, No.3, 2026, DOI:10.32604/cmc.2026.074811 - 09 April 2026

Abstract Adversarial examples in object detection frequently fail to transfer between different models because attacks overfit to the source model’s architecture and feature space. We propose AugTrans, a framework that addresses this limitation through input-space regularization. Our key innovation is a multi-stage augmentation pipeline that incorporates object-level semantic awareness into transformation design. The pipeline comprises three novel components: dynamic object-centric rotation with adaptive scheduling, multi-box aware resizing based on ground-truth annotations, and composite noise injection. These transformations are integrated within the Expectation over Transformation (EOT) framework. By optimizing perturbations to remain effective across semantically meaningful transformations, our… More >

Haolang Feng^1,2, Yuling Chen^1,2,*, Yang Huang^1,2, Xuewei Wang³, Haiwei Sang⁴

CMC-Computers, Materials & Continua, Vol.84, No.3, pp. 4469-4490, 2025, DOI:10.32604/cmc.2025.064896 - 30 July 2025

Abstract Deep neural networks remain susceptible to adversarial examples, where the goal of an adversarial attack is to introduce small perturbations to the original examples in order to confuse the model without being easily detected. Although many adversarial attack methods produce adversarial examples that have achieved great results in the white-box setting, they exhibit low transferability in the black-box setting. In order to improve the transferability along the baseline of the gradient-based attack technique, we present a novel Stochastic Gradient Accumulation Momentum Iterative Attack (SAMI-FGSM) in this study. In particular, during each iteration, the gradient information More >

Xiaoyin Yi^1,2, Long Chen^1,3,4,*, Jiacheng Huang¹, Ning Yu¹, Qian Huang⁵

CMC-Computers, Materials & Continua, Vol.83, No.1, pp. 157-175, 2025, DOI:10.32604/cmc.2025.059863 - 26 March 2025

Abstract Transfer-based Adversarial Attacks (TAAs) can deceive a victim model even without prior knowledge. This is achieved by leveraging the property of adversarial examples. That is, when generated from a surrogate model, they retain their features if applied to other models due to their good transferability. However, adversarial examples often exhibit overfitting, as they are tailored to exploit the particular architecture and feature representation of source models. Consequently, when attempting black-box transfer attacks on different target models, their effectiveness is decreased. To solve this problem, this study proposes an approach based on a Regularized Constrained Feature More >

Yaoxuan Zhu, Hua Yang, Bin Zhu^*

CMC-Computers, Materials & Continua, Vol.82, No.2, pp. 1947-1968, 2025, DOI:10.32604/cmc.2024.057866 - 17 February 2025

Abstract The emergence of adversarial examples has revealed the inadequacies in the robustness of image classification models based on Convolutional Neural Networks (CNNs). Particularly in recent years, the discovery of natural adversarial examples has posed significant challenges, as traditional defense methods against adversarial attacks have proven to be largely ineffective against these natural adversarial examples. This paper explores defenses against these natural adversarial examples from three perspectives: adversarial examples, model architecture, and dataset. First, it employs Class Activation Mapping (CAM) to visualize how models classify natural adversarial examples, identifying several typical attack patterns. Next, various common… More >

Younghoon Ban, Myeonghyun Kim, Haehyun Cho^*

CMES-Computer Modeling in Engineering & Sciences, Vol.139, No.3, pp. 3535-3563, 2024, DOI:10.32604/cmes.2023.046658 - 11 March 2024

Abstract Antivirus vendors and the research community employ Machine Learning (ML) or Deep Learning (DL)-based static analysis techniques for efficient identification of new threats, given the continual emergence of novel malware variants. On the other hand, numerous researchers have reported that Adversarial Examples (AEs), generated by manipulating previously detected malware, can successfully evade ML/DL-based classifiers. Commercial antivirus systems, in particular, have been identified as vulnerable to such AEs. This paper firstly focuses on conducting black-box attacks to circumvent ML/DL-based malware classifiers. Our attack method utilizes seven different perturbations, including Overlay Append, Section Append, and Break Checksum,… More >

Chengsheng Yuan^1,2, Baojie Cui^1,2, Zhili Zhou³, Xinting Li^4,*, Qingming Jonathan Wu⁵

CMC-Computers, Materials & Continua, Vol.78, No.1, pp. 899-914, 2024, DOI:10.32604/cmc.2023.045854 - 30 January 2024

Abstract In recent years, deep learning has been the mainstream technology for fingerprint liveness detection (FLD) tasks because of its remarkable performance. However, recent studies have shown that these deep fake fingerprint detection (DFFD) models are not resistant to attacks by adversarial examples, which are generated by the introduction of subtle perturbations in the fingerprint image, allowing the model to make fake judgments. Most of the existing adversarial example generation methods are based on gradient optimization, which is easy to fall into local optimal, resulting in poor transferability of adversarial attacks. In addition, the perturbation added… More >

Weizheng Wang^1,3, Xiangqi Wang^2,*, Xianmin Pan¹, Xingxing Gong³, Jian Liang³, Pradip Kumar Sharma⁴, Osama Alfarraj⁵, Wael Said⁶

CMC-Computers, Materials & Continua, Vol.76, No.3, pp. 3859-3876, 2023, DOI:10.32604/cmc.2023.041346 - 08 October 2023

Abstract Image-denoising techniques are widely used to defend against Adversarial Examples (AEs). However, denoising alone cannot completely eliminate adversarial perturbations. The remaining perturbations tend to amplify as they propagate through deeper layers of the network, leading to misclassifications. Moreover, image denoising compromises the classification accuracy of original examples. To address these challenges in AE defense through image denoising, this paper proposes a novel AE detection technique. The proposed technique combines multiple traditional image-denoising algorithms and Convolutional Neural Network (CNN) network structures. The used detector model integrates the classification results of different models as the input to… More >

Mingyu Dong, Diqun Yan^*, Rangding Wang

Computer Systems Science and Engineering, Vol.46, No.1, pp. 1-12, 2023, DOI:10.32604/csse.2023.034568 - 20 January 2023

Abstract Speech is easily leaked imperceptibly. When people use their phones, the personal voice assistant is constantly listening and waiting to be activated. Private content in speech may be maliciously extracted through automatic speech recognition (ASR) technology by some applications on phone devices. To guarantee that the recognized speech content is accurate, speech enhancement technology is used to denoise the input speech. Speech enhancement technology has developed rapidly along with deep neural networks (DNNs), but adversarial examples can cause DNNs to fail. Considering that the vulnerability of DNN can be used to protect the privacy in… More >