Journal
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
Volume 19, Issue -, Pages -Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LGRS.2022.3163575
Keywords
Feature extraction; Training; Neurons; Task analysis; Neural networks; Forestry; Noise measurement; Change detection; deep learning (DL); deforestation detection; domain adaptation (DA); remote sensing (RS)
Categories
Funding
- Coordenacao de Aperfeicoamento de Pessoal de Nivel Superior (CAPES)
- Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq)
- Fundacao Carlos Chagas Filho de Amparo a Pesquisa do Estado do Rio de Janeiro (FAPERJ)
- NVIDIA Corporation
Ask authors/readers for more resources
Many deep learning-based domain adaptation methods for remote sensing applications rely on adversarial training strategies to align features extracted from images of different domains. This study proposes a DL-based representation matching approach for domain adaptation in change detection tasks, which effectively mitigates the issue of imbalanced classes and improves the accuracy of cross-domain deforestation detection.
Many deep-learning (DL)-based, domain adaptation (DA) methods for remote sensing (RS) applications rely on adversarial training strategies to align features extracted from images of different domains in a shared latent space. However, the performance of such representation matching techniques is negatively impacted when class occurrences in the target domain, for which no labeled data are available during training, are highly imbalanced. In this work, we propose a DL-based representation matching approach for DA in the context of change detection tasks. We further evaluate the approach in a deforestation mapping application, characterized by a high-class imbalance between the deforestation and no-deforestation classes. The domains represent different sites in the Amazon and Brazilian Cerrado biomes. To mitigate the class imbalance problem, we devised an unsupervised pseudolabeling scheme based on change vector analysis (CVA) that prevents the feature alignment to be biased toward the overrepresented class. The experimental results indicate that the proposed approach can improve the accuracy of cross-domain deforestation detection.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available