Generative Data Augmentation Method for Sonar Images Based on Diffusion Model
Applied Sciences · 29.07.2026 · 16 (15) · 7557
DOI: 10.3390/app16157557 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 29.07.2026. Verfasst von Siqi Cui und 4 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Hainan University, Department of Science and Technology of Hainan Province.
Abstract (Originalsprache)
Sonar object detection is constrained by scarce instance-level annotations, high acquisition costs, and long-tailed category distributions in real underwater environments. To address these limitations, this paper proposes a generative data augmentation framework based on the Stable Diffusion Model (SDM) for synthesizing sonar images together with target bounding boxes. The framework first fine-tunes SDM with an instance-level slice cropping strategy to strengthen the alignment between text prompts and local acoustic target structures. It then introduces a cross-modal sparse localization module (CMSL), which uses denoising features and text priors to infer 2D bounding boxes for generated samples. Synthetic long-tail samples are mixed with real URPC2022 data under a fixed-ratio saturation compensation strategy and evaluated through UTD-SCnet fine-tuning. The results show that instance-level cropping provides the best generation quality among the tested strategies (FID = 31.42, DR = 0.815, CCSCR = 0.075), and that a 60% synthetic-data injection ratio yields the best detection performance. These findings indicate that diffusion-based augmentation can provide a practical, semi-automated supplement for long-tail sonar detection, while excessive synthetic data may introduce domain-shift effects.
Förderung
- National Natural Science Foundation of China
- Hainan University
- Department of Science and Technology of Hainan Province
Zitieren
APA
Cui, S., Wu, K., Zhang, Z., Li, P., & Zhu, R. (2026). Generative Data Augmentation Method for Sonar Images Based on Diffusion Model. Applied Sciences, 16(15), 7557. https://doi.org/10.3390/app16157557
Vancouver
Cui S, Wu K, Zhang Z, Li P, Zhu R. Generative Data Augmentation Method for Sonar Images Based on Diffusion Model. Applied Sciences. 2026;16(15):7557. doi:10.3390/app16157557
BibTeX
@article{Cui2026,
author = {Cui, Siqi and Wu, Kaitao and Zhang, Zhipeng and Li, Pengcheng and Zhu, Rongxin},
title = {Generative Data Augmentation Method for Sonar Images Based on Diffusion Model},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {15},
pages = {7557},
doi = {10.3390/app16157557}
}
EUVIMED