Weakly Supervised Oriented Object Detection in Remote Sensing via Geometry-Aware Enhancement Network
Top Cited Papers
Open Access
- 21 April 2026
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 18 (8) , 1253
Abstract
In remote sensing image oriented object detection tasks, weakly supervised learning methods based on horizontal bounding boxes have attracted much attention due to their lower annotation costs compared to fully supervised methods. However, remote sensing images, characterized by complex backgrounds, exhibit a wide range of target scales and diverse geometric characteristics across target categories. Existing methods exhibit inadequate exploitation of background and angular information under weak supervision, resulting in compromised perception of dense and high-aspect-ratio targets. Neglecting the imbalance in angle estimation samples further leads to excessively low detection accuracy for few-shot categories. To address the aforementioned issues, this paper proposes a Geometry-Aware Enhancement Network (WSOOD-GAEN) for weakly supervised oriented object detection tasks. First, in the backbone network stage, a channel-space deformable attention module (DAE-ResNet) was constructed. Through deformable sampling and screening of key regions, feature extraction has both morphological adaptability to complex shapes and semantic discriminability of key features in complex backgrounds. Secondly, in the feature pyramid stage, an Angle-Guided Feature Pyramid Network (AG-FPN) is proposed. This module dynamically applies rotation transformation to the sampling offsets of deformable convolutions, thereby enhancing the feature representation of objects with different orientations and scales. Furthermore, an adaptive geometric perception loss (AGL) was designed. Based on the geometric characteristics of different categories, it automatically learns differentiated rotation and flip consistency weights, thereby improving the prediction accuracy of small sample categories. Experiments on the DOTA-v1.0, HRSC, and RSAR datasets validate our approach. Specifically, under the AP75 evaluation metric, the proposed method outperforms existing weakly supervised methods by 1.51%, 9.86%, and 3.28%, respectively.Keywords
Funding Information
- Foundation Strengthening Fund Project (2023-JCJQ-JJ-0343)
- Fundamental Research Funds for the Central Universities (2042022dx0001)
This publication has 56 references indexed in Scilit:
- More Accurate Constraints for Self-Supervised Learning in Remote Sensing Images-Based Object DetectionIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025
- Point-Based Weakly Semisupervised Oriented Vehicle Detection in Optical Remote Sensing ImagesIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024
- Ship-Go: AR S ip mages In ainting via instance-to-image enerative Diffusion M delsISPRS Journal of Photogrammetry and Remote Sensing, 2023
- Feature Enhancement and Alignment for Oriented Object DetectionIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023
- A Comprehensive Review for Typical Applications Based Upon Unmanned Aerial Vehicle PlatformIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022
- Lightweight Oriented Object Detection Using Multiscale Context and Enhanced Channel Attention in Remote Sensing ImagesIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021
- Research Progress on Few-Shot Learning for Remote Sensing Image InterpretationIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021
- Object detection in optical remote sensing images: A survey and a new benchmarkISPRS Journal of Photogrammetry and Remote Sensing, 2020
- A survey on semi-supervised learningMachine Learning, 2019
- Remote Sensing Image Scene Classification: Benchmark and State of the ArtProceedings of the IEEE, 2017