基于混合专家模型的航天遥感图像目标检测-航天遥感图像智能处理与分析专栏

卢光宇; 陈博文; 陈科研; 史振威; 邹征夏

doi:10.7527/S1000-6893.2025.32599

航空学报 >

0 1 - 0

DOI: https://doi.org/10.7527/S1000-6893.2025.32599

基于混合专家模型的航天遥感图像目标检测-航天遥感图像智能处理与分析专栏

卢光宇 ,
陈博文 ,
陈科研 ,
史振威 ,
邹征夏

展开

北京航空航天大学宇航学院

收稿日期: 2025-07-21

修回日期: 2025-10-22

网络出版日期: 2025-10-24

基金资助

国家自然科学基金

收起

Spaceborne Remote Sensing Image Object Detection Based on Mixture of Experts Mode

LU Guang-Yu ,
CHEN Bo-Wen ,
CHEN Ke-Yan ,
SHI Zhen-Wei ,
ZOU Zheng-Xia

Expand

Received date: 2025-07-21

Revised date: 2025-10-22

Online published: 2025-10-24

Fold

摘要

目标检测作为航天遥感领域的重要应用方向，在灾害应急、资源管理、环境监测等领域具有重要意义，本文以船只目标为例，研究航天遥感图像目标检测。航天遥感观测范围覆盖全球范围不同地区的海域，而这些海域地貌特征存在巨大差异，现有船只目标检测方法多采用统一的结构处理不同场景，难以适应不同区域及不同分辨率下地物特征的变化。为此，本文将混合专家模型（Mixture of Experts，MoE）引入航天遥感图像目标检测任务，提出一种基于混合专家模型的航天遥感图像目标检测模型，针对不同地理区域的差异特征，自适应地选择相应的专家处理，实现不同海域下更精准的船只目标检测。具体而言，模型包含两个专家组：一组由空间分辨率信息引导，以捕获不同分辨率下的地物尺度差异；另一组则利用经纬度信息引导，以适应不同地理区域的地物分布特征。通过两组专家的协同工作，模型能够更加精准地感知和定位图像中的船只目标，显著提高检测性能。此外，本文还构建了一个包含多分辨率信息和经纬度信息的数据集——地理信息船只检测数据集（Geographic Information Ship Detection Dataset , GISD），为研究提供数据支撑。本文在GISD数据集和公开数据集LEVIR-ship上进行了实验验证，本文方法的AP50分别达到了80.6%、85.7%，相较于两个数据集上性能最优的模型提升了3.8%、2%。

关键词： 遥感图像; 船只检测; 光学遥感; 地理信息; 混合专家

本文引用格式

卢光宇 , 陈博文 , 陈科研 , 史振威 , 邹征夏 . 基于混合专家模型的航天遥感图像目标检测-航天遥感图像智能处理与分析专栏[J]. 航空学报, 0 : 1 -0 . DOI: 10.7527/S1000-6893.2025.32599

Abstract

As an important application direction in the field of spaceborne remote sensing, object detection is of great significance in disaster emergency response, resource management, environmental monitoring and other fields. This paper takes ship targets as an example to study object detection in spaceborne remote sensing images. Spaceborne remote sensing covers sea areas in different regions worldwide, and these sea areas have significant differences in geomorphic features. Existing ship target detection methods mostly use a unified structure to handle different scenarios, making it difficult to adapt to changes in ground object features in different regions and with different resolutions. To address this, this paper introduces the Mixture of Experts (MoE) model into the task of spaceborne remote sensing image object detection, and proposes a spaceborne remote sensing image object detection model based on the Mixture of Experts model. It adaptively selects corresponding expert processing for the differential features of different geographical regions to achieve more accurate ship target detection in different sea areas. Specifically, the model includes two groups of experts: one group is guided by spatial resolution information to capture differences in ground object scales under different resolutions; the other group uses latitude and longitude information for guidance to adapt to the distribution characteristics of ground objects in different geographical regions. Through the collaborative work of the two groups of experts, the model can more accurately perceive and locate ship targets in images, significantly improving detection performance. In addition, this paper constructs a dataset containing multi - resolution information and latitude and longitude information - the Geographic Information Ship Detection Dataset (GISD) - to provide data support for the research. This paper conducts experimental verification on the GISD dataset and the public dataset LEVIR - ship. The AP50 of the proposed method reaches 80.6% and 85.7% respectively, which is 3.8% and 2% higher than the best - performing models on the two datasets.

Key words： Remote - sensing image; Ship detection; Optical remote sensing; Geographical information; Mixture of experts

参考文献

[1]LIN Y, LI J X, WEI S Q, et al. Optishipnet: Efficient ship detection in complex marine environments using optical remote sensing images, [J]. Journal of Marine Scence and Engineering, vol. 12, no. 10, p. 1786, 2024.
[2]LIU Z, ZHANG W J, YU H, et al. Improved yolov5s for small ship detection with optical remote sensing images, [J]. IEEE Geoscience and Remote Sensing Letters, vol. 20, pp. 1–5, 2023.
[3]肖欣林，施伟超，郑向涛，等．基于多模型协同的舰船目标检测［J］．航空学报， 2024， 45（14）： 630241．
XIAO X L，SHI W C， ZHENG X T， et al. Multiple models collaboration for ship detection［J］. Acta Aeronautica et Astronautica Sinica， 2024， 45（14）： 630241 （in Chinese）. doi：10. 7527/S1000-6893. 2024. 30241.
[4]赵其昌，吴一全，苑玉彬．光学遥感图像舰船目标检测与识别方法研究进展［J］．航空学报， 2024， 45（8）： 029025．
ZHAO Q C,WU Y Q， YUAN Y B. Progress of ship detection and recognition methods in optical remote sensing images［J］. Acta Aeronautica et Astronautica Sinica， 2024， 45（8）： 029025 （in Chinese）. doi：10.7527/S10006893.2023.29025
[5]ZOU Z X, CHEN K Y, SHI Z W, et al. Object detection in 20 years: A survey. [J]. Proceedings of the IEEE, 111(3), pp.257-276.
[6]CHEN B W, LIU L Q, ZOU Z X, et al. Target detection in hyperspectral remote sensing image: Current status and challenges. [J]. Remote Sensing 15, no. 13 (2023): 3223.
[7]SHI Z W, YU X R, JIANG Z G, et al. Ship Detection in High-Resolution Optical Imagery Based on Anomaly Detector and Local Shape Feature, in IEEE Transactions on Geoscience and Remote Sensing, vol. 52, no. 8, pp. 4511-4523, Aug. 2014, doi: 10.1109/TGRS.2013.2282355.
[8]黎经元, 厉小润, 赵辽英. 基于边缘线分析与聚合通道特征的港口舰船检测[J]. Acta Optica Sinica, 2019, 39(8): 0815004.
Li J Y, Li X R, ZHAO L Y. Docked Ship Detectio Based on Edge Line Analysis and Aggregation Channel Features [J]. Acta Optica Sinica, 2019, 39(8): 0815004.
[9]WANG, LI H, ZHU M, et al. Ship detection in optical remote sensing image based on visual saliency and AdaBoost classifier. [J]. Optoelectronics letters 13.2 (2017): 151-155. Acta Optica Sinica, 2019, 39(8): 0815004.
[10]GIRSHICK R, DOBAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2014: 580-587.
[11]REDMON J, DIVVALA S, GRISHICK R, et al. You only look once: Unified, real-time object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 779-788.
[12]CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]//European conference on computer vision. Cham: Springer International Publishing, 2020: 213-229.
[13]LIN H, SHI Z W and ZOU Z X, Fully Convolutional Network With Task Partitioning for Inshore Ship Detection in Optical Remote Sensing Images, in IEEE Geoscience and Remote Sensing Letters, vol. 14, no. 10, pp. 1665-1669, Oct. 2017, doi: 10.1109/LGRS.2017.2727515.
[14]CHEN J Q, CHEN K Y, CHEN H, et al. A degraded reconstruction enhancement-based method for tiny ship detection in remote sensing images with a new large-scale dataset[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-14.
[15]王子玲，熊振宇，杨璐铖，等. AIS和光学遥感图像引导的星载SAR舰船目标识别网络［J］. 航空学报， 2024， 45（2）： 328672.
WANG Z L， XIONG Z Y， YANG L C， et al. Spaceborne SAR ship target recognition network guided by AIS and optical remote sensing images［J］. Acta Aeronautica et Astronautica Sinica， 2024， 45（2）： 328672 （in Chinese）. doi：10.7527/S1000-6893.2023.28672
[16]XING Z J, REN J, FAN X Z, et al. S-DETR: A transformer model for real-time detection of marine ships[J]. Journal of Marine Science and Engineering, 2023, 11(4): 696.
[17]WANG P L, LIU B X, LI Y, et al. Iceregionship: Optical remote sensing dataset for ship detection in ice-infested waters, [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 17,pp. 1007–1020, 2023. 7
[18]YAN Y M, HE Y M, SU N, et al. Pnbt-cr: A cloud removal method for ship detection, [J]. IEEE Geoscience and Remote Sensing Letters, vol. 21, pp. 1–5, 2024.
[19]刘子力, 杨家俊, 王文静. 遥感图像云检测方法综述[J]. 中国空间科学技术, 2023, 43(1): 1-17.
LIU Z L, YANG J J, WANG W J, et al. Cloud detection methods for remote sensing images: a survey[J]. Chinese Space Science and Technology, 2023, 43(1): 1.
[20]LI Y X, HOU Q B, ZHENG Z H, et al. Large selective kernel network for remote sensing object detection[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2023: 16794-16805.
[21]MA D Q, LI S J, DANG B, et al. Fostc3net: A lightweight yolov5 based on the network structure optimization[C]//Journal of Physics: Conference Series. IOP Publishing, 2024, 2824(1): 012004.
[22]QIN C, WANG X Q, LI G, et al. An improved attention-guided network for arbitrary-oriented ship detection in optical remote sensing images[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 1-5.
[23]ZOU Z X, SHI Z W. Ship detection in spaceborne optical image with SVD networks[J]. IEEE transactions on geoscience and remote sensing, 2016, 54(10): 5832-5845.
[24]BI F K, Hou J Y, CHEN L, et al. Ship detection for optical remote sensing images based on visual attention enhanced network[J]. Sensors, 2019, 19(10): 2271.
[25]JACOBS R A, JORDAN M I, NOWLAN S J, et al. Adaptive mixtures of local experts[J]. Neural computation, 1991, 3(1): 79-87.
[26]FEDUS W, ZOPH B, SHAZEER N. Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity[J]. Journal of Machine Learning Research, 2022, 23(120): 1-39.
[27]CHEN B W, CHEN K Y, YANG M, et al. Heterogeneous mixture of experts for remote sensing image super-resolution[J]. IEEE Geoscience and Remote Sensing Letters, 2025.
[28]XIA G S, Bai X, DING J, et al. DOTA: A large-scale dataset for object detection in aerial images[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 3974-3983.
[29]LIU Z, LIN Y T, CAO Y, et al. Swin transformer: Hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2021: 10012-10022.
[30]SONG J N, ZHOU M L, LUO J, et al. Boundary-Aware Feature Fusion with Dual-Stream Attention for Remote Sensing Small Object Detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024.
[31]TIAN Z, SHEN C H, CHEN H, et al. Fcos: Fully convolutional one-stage object detection[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2019: 9627-9636.
[32]He K, Gkioxari G, Dollár P, et al. Mask r-cnn[C]//Proceedings of the IEEE international conference on computer vision. 2017: 2961-2969.
[33]ZHANG H, LI F, LIU S L, et al. Dino: Detr with improved denois-ing anchor boxes for end-to-end object detection, [DB/OL]. arXiv preprint :2203.03605, 2022.
[34]ZONG Z F, SONG G L, LIU Y. Detrs with collaborative hybrid assignments training[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2023: 6748-6758.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献