Research on Underwater Sonar Image Object Detection Method based on Improved YOLOv11n
Scientific Journal of Intelligent Systems Research · 14.05.2026 · 8 (4) · 21-30
DOI: 10.54691/8cydbk77 ↗Kurz gefasst
Fachartikel in der Zeitschrift Scientific Journal of Intelligent Systems Research, online erschienen am 14.05.2026. Verfasst von Weina Zhao und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Aiming at the problems of strong noise, low resolution, blurred target edges, severe geometric distortion of underwater sonar images, insufficient accuracy of traditional target detection algorithms, and poor adaptability of the original YOLOv11n model, an underwater sonar image target detection method based on improved YOLOv11n is proposed. The method adopts novel median filtering and single-scale Retinex for image denoising and enhancement, optimizes anchor boxes with K-means++, uses Mixup data augmentation, Focal-EIOU Loss and Soft-NMS, embeds ECA attention mechanism in the backbone network, and replaces the neck network with BiFPN structure. Experimental results show that the proposed method achieves 97.5% precision, 97.8% recall and 98.2% mAP@.5 on the self-built sonar dataset, which are 6.3, 6.2 and 6 percentage points higher than those of the original YOLOv11n respectively. It can effectively adapt to complex underwater environments and meet the high-precision detection requirements of marine exploration, maritime security and other engineering applications.
Zitieren
APA
Zhao, W., Qian, Y., Enrique, W., & Wang, M. (2026). Research on Underwater Sonar Image Object Detection Method based on Improved YOLOv11n. Scientific Journal of Intelligent Systems Research, 8(4), 21-30. https://doi.org/10.54691/8cydbk77
Vancouver
Zhao W, Qian Y, Enrique W, Wang M. Research on Underwater Sonar Image Object Detection Method based on Improved YOLOv11n. Scientific Journal of Intelligent Systems Research. 2026;8(4):21-30. doi:10.54691/8cydbk77
BibTeX
@article{Zhao2026,
author = {Zhao, Weina and Qian, Yiwei and Enrique, Willam and Wang, Mingyu},
title = {Research on Underwater Sonar Image Object Detection Method based on Improved YOLOv11n},
journal = {Scientific Journal of Intelligent Systems Research},
year = {2026},
volume = {8},
number = {4},
pages = {21--30},
doi = {10.54691/8cydbk77}
}
EUVIMED