SGFAR-YOLO: Semantic-Guided Multiscale Fusion and Anisotropic Refinement for Satellite Instance Segmentation and UAV-Based Damage Localization of Waterway Regulation Structures
Remote Sensing · 20.11.2026 · 18 (22) · 3995
DOI: 10.3390/rs18223995 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 20.11.2026. Verfasst von Aiguo Sun und 13 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Changjiang River Administration of Navigational Affairs, Hubei province.
Abstract (Originalsprache)
Regional inventories and local condition assessment of waterway regulation structures require complementary observations from satellite and unmanned aerial vehicle (UAV) imagery. Automated interpretation must preserve elongated structures with weak boundaries while separating targets from complex river backgrounds. We propose semantic-guided multiscale fusion and anisotropic refinement YOLO (SGFAR-YOLO), a novel pre-head feature design for independently trained satellite instance segmentation and UAV-oriented detection models. A semantic-guided feature pyramid network (SGFPN) coordinates coarse semantics and fine detail, while the residual strip enhancement module (RSEM) provides gated local and axial context at high-resolution pyramid levels. On the satellite validation split, SGFAR-YOLO achieves 57.31% horizontal-bounding-box AP and 43.14% mask AP, exceeding the YOLO11 baseline by 3.11 and 2.80 percentage points. UAV-oriented box AP reaches 34.75%, a gain of 5.71 percentage points. Ablation studies and qualitative comparisons examine the design across both tasks and illustrate boundary continuity, small-target recovery, and localization of surface damage conditions. The proposed method provides a practical framework for regional mapping of waterway regulation structures and local screening of their surface condition within the Jingjiang reach.
Förderung
- Changjiang River Administration of Navigational Affairs
- Hubei province
Zitieren
APA
Sun, A., Yang, S., Xiong, R., Sheng, C., Zhu, J., Xu, T., Jiang, Y., Hu, D., Han, Z., Zhang, X., Sun, A., Huang, X., Chen, Z., & Chen, N. (2026). SGFAR-YOLO: Semantic-Guided Multiscale Fusion and Anisotropic Refinement for Satellite Instance Segmentation and UAV-Based Damage Localization of Waterway Regulation Structures. Remote Sensing, 18(22), 3995. https://doi.org/10.3390/rs18223995
Vancouver
Sun A, Yang S, Xiong R, Sheng C, Zhu J, Xu T, et al. SGFAR-YOLO: Semantic-Guided Multiscale Fusion and Anisotropic Refinement for Satellite Instance Segmentation and UAV-Based Damage Localization of Waterway Regulation Structures. Remote Sensing. 2026;18(22):3995. doi:10.3390/rs18223995
BibTeX
@article{Sun2026,
author = {Sun, Aiguo and Yang, Shuang and Xiong, Rongjun and Sheng, Chenfei and Zhu, Jiayu and Xu, Tao and Jiang, Yuyao and Hu, Dongdong and Han, Zhiwei and Zhang, Xiaoning and Sun, Aifeng and Huang, Xin and Chen, Zeqiang and Chen, Nengcheng},
title = {SGFAR-YOLO: Semantic-Guided Multiscale Fusion and Anisotropic Refinement for Satellite Instance Segmentation and UAV-Based Damage Localization of Waterway Regulation Structures},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {22},
pages = {3995},
doi = {10.3390/rs18223995}
}
EUVIMED