Geospatial Prior Guided Confused-Class Data Augmentation for Semantic Segmentation of Remote Sensing Images
- 9 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64 (01962892) , 1-19
Abstract
Semantic segmentation of high spatial resolution remote sensing images plays a crucial role in diverse applications. Deep learning-based methods have emerged as the dominant approach in this field. However, existing methods face the challenge of effectively using the information in annotated training data, resulting in limited model performance. Data augmentation is a commonly used technique to better reveal the information embedded in the data through appropriate transformations. However, these methods usually fail to account for the unique characteristics of remote sensing images, such as the high spectral similarity among different classes and intra-class variation among distinct objects. This oversight hinders the model’s ability to distinguish between spectrally similar classes, resulting in persistent misclassifications. To address these challenges, we propose a novel geospatial prior guided confused-class data augmentation (GCA) method for semantic segmentation of remote sensing images. By emphasizing the occurrence of confused classes within augmented training data, GCA compels the model to focus on learning discriminative features to improve class separability. GCA consists of three key modules: geospatial prior information extraction module (module G), confused-class similar units detection module (C), and confused-class data augmentation module (module A). Specifically, module G extracts geographic units and their spatial relationships. Module C identifies highly similar units from confused classes that are prone to misclassification. Module A generates challenging training data by emphasizing confused-class similar units under the guidance of geospatial priors. Experiments on the GID and LoveDA datasets demonstrate that the proposed GCA method shows a better performance than state-of-the-art data augmentation methods. The GCA code is publicly available at https://github.com/Jingying-Li/GCA.Keywords
Funding Information
- Natural Science Foundation of Henan Province (232300421071)
- National Natural Science Foundation of China (32160369)
- National Natural Science Foundation of China (41961053)
- Program for Innovative Research Team in University of Henan Province (26IRTSTHN003)
- State Key Laboratory of Natural Medicines
This publication has 55 references indexed in Scilit:
- B3-CDG: A pseudo-sample diffusion generator for bi-temporal building binary change detectionISPRS Journal of Photogrammetry and Remote Sensing, 2024
- Effective variance attention-enhanced diffusion model for crop field aerial image super resolutionISPRS Journal of Photogrammetry and Remote Sensing, 2024
- Coarse-to-fine semantic segmentation of satellite imagesISPRS Journal of Photogrammetry and Remote Sensing, 2024
- Toulouse Hyperspectral Data Set: A benchmark data set to assess semi-supervised spectral representation learning and pixel-wise classification techniquesISPRS Journal of Photogrammetry 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
- An Effective Attention-based CNN Model for Fire Detection in Adverse Weather ConditionsISPRS Journal of Photogrammetry and Remote Sensing, 2023
- Deep Semantic-Visual Alignment for zero-shot remote sensing image scene classificationISPRS Journal of Photogrammetry and Remote Sensing, 2023
- BiSeNet V2: Bilateral Network with Guided Aggregation for Real-Time Semantic SegmentationInternational Journal of Computer Vision, 2021
- Parsing very high resolution urban scene images by learning deep ConvNets with edge-aware lossISPRS Journal of Photogrammetry and Remote Sensing, 2020
- SSD: Single Shot MultiBox Detector2016