MPFT-UNet: A Boundary-Refined and Multi-Scale Dynamic Fusion Network for UAV-Based Port Ship Segmentation

Shi, Mengna; Qiu, Xiulin; Li, Ang; Yang, Yuwang; Ke, Yaqi; Chen, Yilan

doi:10.3390/jmse14100945

Open AccessArticle

MPFT-UNet: A Boundary-Refined and Multi-Scale Dynamic Fusion Network for UAV-Based Port Ship Segmentation

by

Mengna Shi

¹

,

Xiulin Qiu

^1,*

,

Ang Li

¹

,

Yuwang Yang

²

,

Yaqi Ke

²

and

Yilan Chen

¹

School of Automation, Jiangsu University of Science and Technology, Zhenjiang 212100, China

²

School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China

^*

Author to whom correspondence should be addressed.

J. Mar. Sci. Eng. 2026, 14(10), 945; https://doi.org/10.3390/jmse14100945

Submission received: 28 April 2026 / Revised: 14 May 2026 / Accepted: 17 May 2026 / Published: 19 May 2026

(This article belongs to the Topic Advanced Technologies and Applications for Unmanned Systems)

Download

Browse Figures

Versions Notes

Abstract

Ship semantic segmentation based on unmanned aerial vehicle (UAV) imagery has important application value in maritime scenarios such as marine surveillance, port management, and maritime safety. However, UAV images often contain large scale variations of ships, a high proportion of small targets, and complex background interference, including sea surface reflections, waves, and clouds. These factors make accurate segmentation and boundary localization difficult. To address these issues, this paper proposes a UAV-based ship semantic segmentation network, termed MPFT-UNet. The network introduces a Multi-scale Dynamic Sparse Cross-gating (MDSC) module to improve the representation of small targets. A Boundary Supervision Refinement (BSR) module is used to enhance boundary delineation. In addition, a Transformer-based Feature Fusion (FFT) module is applied at the bottleneck layer to strengthen global semantic representation. Experimental results show that MPFT-UNet achieves better performance than existing methods across multiple evaluation metrics. The model obtains an IoU of 0.8365, Dice coefficient of 0.9028, Recall of 0.8881, and AP of 0.95731. These results indicate stable segmentation performance under complex maritime conditions. Compared with the baseline U-Net model, the IoU is improved by approximately 5.1%.

Keywords: UAV remote sensing; ship semantic segmentation; small object segmentation; boundary-aware learning; multi-scale feature fusion; Transformer

Share and Cite

MDPI and ACS Style

Shi, M.; Qiu, X.; Li, A.; Yang, Y.; Ke, Y.; Chen, Y. MPFT-UNet: A Boundary-Refined and Multi-Scale Dynamic Fusion Network for UAV-Based Port Ship Segmentation. J. Mar. Sci. Eng. 2026, 14, 945. https://doi.org/10.3390/jmse14100945

AMA Style

Shi M, Qiu X, Li A, Yang Y, Ke Y, Chen Y. MPFT-UNet: A Boundary-Refined and Multi-Scale Dynamic Fusion Network for UAV-Based Port Ship Segmentation. Journal of Marine Science and Engineering. 2026; 14(10):945. https://doi.org/10.3390/jmse14100945

Chicago/Turabian Style

Shi, Mengna, Xiulin Qiu, Ang Li, Yuwang Yang, Yaqi Ke, and Yilan Chen. 2026. "MPFT-UNet: A Boundary-Refined and Multi-Scale Dynamic Fusion Network for UAV-Based Port Ship Segmentation" Journal of Marine Science and Engineering 14, no. 10: 945. https://doi.org/10.3390/jmse14100945

APA Style

Shi, M., Qiu, X., Li, A., Yang, Y., Ke, Y., & Chen, Y. (2026). MPFT-UNet: A Boundary-Refined and Multi-Scale Dynamic Fusion Network for UAV-Based Port Ship Segmentation. Journal of Marine Science and Engineering, 14(10), 945. https://doi.org/10.3390/jmse14100945

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

MPFT-UNet: A Boundary-Refined and Multi-Scale Dynamic Fusion Network for UAV-Based Port Ship Segmentation

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI