PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 21, 2026Results in Engineering0 citationsOpen Access

Pareto Calibrated Adversarial Sample Generation for SAR ATR with Aggression-Concealment Balance

View Full Paper
XSXinyuan SuSQSinong QuanZCZhihao Cai

Key Points

  • This research aims to optimize adversarial sample generation for Synthetic Aperture Radar Automatic Target Recognition by balancing attack effectiveness and visual concealment.
  • Developed the SMPI-FGSM method for generating adversarial samples.
  • Defined an aggression-concealment metric using Attack Success Rate and Structural Similarity.
  • Designed the NSTGWOA to identify optimal parameters for sample generation.
  • Employed the YOLOv8 detection network for disturbance detection on generated samples.
  • Achieved an average Attack Success Rate (ASR) of over 90%, surpassing mainstream methods by 4-5%.
  • Maintained a Structural Similarity index (SSIM) of 0.403 in balanced adversarial samples.
  • Obtained an ASR of 74.40% with a disturbance factor of only 0.089.

Abstract

• Adversarial sample generation is formulated as a joint optimization problem. • The SMPI-FGSM method is proposed to craft potent adversarial samples. • An aggression-concealment metric combining ASR and SSIM is defined. • The NSTGWOA is developed to obtain Pareto-optimal parameters. Adversarial attack is a key security issue in deep learning models for Synthetic Aperture Radar Automatic Target Recognition (SAR ATR). To balance attack effectiveness with visual concealment, this paper frames adversarial sample generation as a joint optimization problem and proposes an optimal generation method based on an Aggression-Concealment balance strategy. First, by integrating transformation and stabilized momentum mechanisms, a stabilized momentum with pre-convergence and input transformation fast gradient sign method (SMPI-FGSM) is proposed to enhance the potency of generated adversarial samples. Second, an aggression-concealment evaluation model is established, using the Attack Success Rate (ASR) and Structural Similarity (SSIM) index as objective functions for optimization. Third, by introducing Kent chaotic map, golden section strategy and time-varying weight into non-dominated sorting whale optimization algorithm (NSWOA), a non-dominated sorting time-varying golden-section whale optimization algorithm (NSTGWOA) is developed to solve the model and identify candidate optimal adversarial samples. Finally, the YOLOv8 detection network is employed to perform disturbance detection on samples within the Pareto set, thereby determining the optimal balance parameters. Experimental results demonstrate that the average ASR of SMPI-FGSM method for different classification networks is over 90%, which is 4-5% higher than that of mainstream methods. The optimal balanced adversarial samples maintain the concealment of SSIM = 0.403 while achieving ASR = 74.40 %, and the disturbance factor is only 0.089, which provides a general reference for other adversarial sample generation methods.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Su et al. (2026) studied this question.

synapsesocial.com/papers/69994ba9873532290d01fda0https://doi.org/10.1016/j.rineng.2026.109650
Ask AI
Helpful
Bookmark
Share
View Full Paper