OraL: An Observational Learning Paradigm for Unsupervised Hyperspectral Change Detection
- 8 January 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 35 (6) , 5380-5393
Abstract
Unsupervised hyperspectral change detection (UHCD), detecting subtle changes between bi-temporal images without manual annotations, is an essential but challenging task in the earth observation community. The current modus operandi often performs it in a feature comparison manner, which is limited by variations in imaging conditions. We observe that fully supervised paradigms using limited annotations are capable of overcoming this challenge. Based on this, we introduce a novel Observational Learning Paradigm (OraL) for UHCD by mimicking fully supervised paradigms. OraL comprises two sequential stages: Observation, which designs a spatial-temporal observation strategy (STO) that records the learning consistency of pixels under different training steps and views, to obtain reliable pseudo-labels. Reproduction, which retrains the model with these pseudo-labels and introduces a distribution-aware spectral learning strategy (DSL) to adaptively increase their learning difficulty according to spectral distributions, enhancing the robustness and generalization of the model. Extensive experiments on several public hyperspectral image datasets demonstrate its state-of-the-art performance and pluggability for previous unsupervised methods. Code will be made available.Keywords
Funding Information
- National Natural Science Foundation of China (62276197)
- National Natural Science Foundation of China (62171332)
- Postdoctoral Fellowship Program of CPSF (GZC20241321)
- China Postdoctoral Science Foundation (2024M762549)
This publication has 62 references indexed in Scilit:
- GPSFormer: A Global Perception and Local Structure Fitting-Based Transformer for Point Cloud Understanding2024
- Enhancement, integration, expansion: Activating representation of detailed features for occluded person re-identificationNeural Networks, 2023
- MOL: Towards accurate weakly supervised remote sensing object detection via Multi-view nOisy LearningISPRS Journal of Photogrammetry and Remote Sensing, 2023
- Combating Noisy Labels by Agreement: A Joint Training Method with Co-Regularization2020
- Learn2Perturb: An End-to-End Feature Perturbation Learning to Improve Adversarial Robustness2020
- Probabilistic End-To-End Noise Correction for Learning With Noisy Labels2019
- CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noise2018
- GPU Framework for Change Detection in Multitemporal Hyperspectral ImagesInternational Journal of Parallel Programming, 2017
- A post-classification change detection method based on iterative slow feature analysis and Bayesian soft fusionRemote Sensing of Environment, 2017
- Extending post-classification change detection using semantic similarity metrics to overcome class heterogeneity: A study of 1992 and 2001 U.S. National Land Cover Database changesRemote Sensing of Environment, 2008