Results 91 to 100 of about 619 (184)

Adversarial Patch Attack for Ship Detection via Localized Augmentation

CoRR
Current ship detection techniques based on remote sensing imagery primarily rely on the object detection capabilities of deep neural networks (DNNs). However, DNNs are vulnerable to adversarial patch attacks, which can lead to misclassification by the detection model or complete evasion of the targets.
Chun Liu 0008, Panpan Ding, Zheng Zheng 0003, Hailong Wang, Bingqian Zhu, Tao Xu, Zhigang Han, Jiayao Wang   +7 more
openaire   +2 more sources

Optimal Adversarial Sample Generation Method in SAR ATR Based on Joint Misleading and Fidelity Optimization

Leida xuebao
Adversarial sample generation is a key research direction for uncovering the vulnerabilities of deep neural networks and improving the robustness of Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) systems.
Xinyuan SU, Sinong QUAN, Zhihao CAI, Shiqi XING, Junpeng WANG   +4 more
doaj   +1 more source

Building centaur responders: is emergency management ready for artificial intelligence?

Disasters, Volume 50, Issue 3, July 2026.
Abstract This article examines the preparedness of emergency management (EM) for addressing questions pertaining to artificial intelligence (AI), encompassing its benefits to EM missions, the potential biases, the societal impacts, and more. We pinpoint two key shortcomings in early EM research on AI: (i) insufficient discussion of both AI's history ...
Christopher Whyte, Brittany ‘Brie’ Haupt   +1 more
wiley   +1 more source

Block-level masking and feature importance-based adversarial example generation

Journal of Big Data
This paper proposes a method to enhance the transferability of adversarial examples by combining a Learnable Patch-Wise Mask (LPM) generated through differential evolution algorithm with a Feature Importance Aware (FIA) attack.
Wenbo Qiu, Yafei Song
doaj   +1 more source

OD-SHIELD: Convolutional Autoencoder-Based Defense Against Adversarial Patch Attacks in Object Detection

IEEE Access
In the evolving landscape of deep neural network security, adversarial patch attacks present a serious challenge for object detection systems. We introduce OD-Shield, a novel defense approach that employs a convolutional autoencoder framework to detect ...
Byeongchan Kim, Heemin Kim, Minjung Kang, Hyunjee Nam, Sunghwan Park, Jaewoo Lee, Il-Youp Kwak   +6 more
doaj   +1 more source

Patch is enough: naturalistic adversarial patch against vision-language pre-training models

Visual Intelligence
Visual language pre-training (VLP) models have demonstrated significant success in various domains, but they remain vulnerable to adversarial attacks. Addressing these adversarial vulnerabilities is crucial for enhancing security in multi-modal learning.
Dehong Kong, Siyuan Liang, Xiaopeng Zhu, Yuansheng Zhong, Wenqi Ren   +4 more
doaj   +1 more source

Scalable One-Pixel Attacks on Deep Neural Networks for High-Resolution Images

Mathematics
Recent studies have shown that deep neural networks can be misled by adversarial examples that involve only imperceptible perturbations. Among these, one-pixel attacks (OPA) represent an extreme yet powerful threat, as they alter only a single pixel of ...
Wonhong Nam, Hyunwoo Moon, Kunha Kim, Hyunyoung Kil   +3 more
doaj   +1 more source

Perceptual Carlini-Wagner Attack: A Robust and Imperceptible Adversarial Attack Using LPIPS

IEEE Access
Adversarial attacks on deep neural networks (DNNs) present significant challenges by exploiting model vulnerabilities using perturbations that are often imperceptible to human observers.
Liming Fan, Anis Salwa Mohd Khairuddin, Haichuan Liu, Khairunnisa Binti Hasikin   +3 more
doaj   +1 more source

Rethinking adversarial attacks on neuromorphic models

Neuromorphic Computing and Engineering
Spiking neural networks (SNN) are biologically inspired artificial neural networks that emulate the behaviour of biological neurons in spiking-based computational units.
Soukaina Aji, Pierre Boulet, Mohamed Bouassida, Ihsen Alouani   +3 more
doaj   +1 more source

A comparative study of phantom sponge for monocular 3D object detection on edge devices

ICT Express
In this paper, we expand the Phantom Sponge attack to monocular 3D object detection to increase false positives and detection times, thereby impairing the performance of edge devices.
Shaheer Siddiqui, Jinpyung Kim, Jangho Lee   +2 more
doaj   +1 more source