Hierarchical Semi-Supervised Active Learning for Remote Sensing

2025-11-22

AI Summary

High annotation costs and low utilization of unlabeled remote sensing data severely limit deep learning model performance. To address this, we propose a hierarchical semi-supervised active learning framework that innovatively integrates semi-supervised learning with hierarchical active learning. Our method employs a weak-strong augmentation self-training mechanism to enhance model robustness and introduces a hierarchical sampling strategy grounded in feature representation and uncertainty estimation; further, progressive clustering is incorporated to jointly optimize sample diversity, representativeness, and query efficiency. Evaluated on the UCM, AID, and NWPU-RESISC45 benchmarks, our approach achieves over 95% of the accuracy of fully supervised models using only 2%–8% labeled data. This significantly improves label efficiency and accelerates model convergence, demonstrating substantial gains in both data economy and training effectiveness.

Technology Category

• Machine Learning: Active Learning
• Computer Vision: Representation Learning for Vision
• Search and Optimization: Learning to Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

The performance of deep learning models in remote sensing (RS) strongly depends on the availability of high-quality labeled data. However, collecting large-scale annotations is costly and time-consuming, while vast amounts of unlabeled imagery remain underutilized. To address this challenge, we propose a Hierarchical Semi-Supervised Active Learning (HSSAL) framework that integrates semi-supervised learning (SSL) and a novel hierarchical active learning (HAL) in a closed iterative loop. In each iteration, SSL refines the model using both labeled data through supervised learning and unlabeled data via weak-to-strong self-training, improving feature representation and uncertainty estimation. Guided by the refined representations and uncertainty cues of unlabeled samples, HAL then conducts sample querying through a progressive clustering strategy, selecting the most informative instances that jointly satisfy the criteria of scalability, diversity, and uncertainty. This hierarchical process ensures both efficiency and representativeness in sample selection. Extensive experiments on three benchmark RS scene classification datasets, including UCM, AID, and NWPU-RESISC45, demonstrate that HSSAL consistently outperforms SSL- or AL-only baselines. Remarkably, with only 8%, 4%, and 2% labeled training data on UCM, AID, and NWPU-RESISC45, respectively, HSSAL achieves over 95% of fully-supervised accuracy, highlighting its superior label efficiency through informativeness exploitation of unlabeled data. Our code will be released at https://github.com/zhu-xlab/RS-SSAL.

Problem

• Addressing costly labeled data collection in remote sensing deep learning models
• Developing hierarchical framework combining semi-supervised and active learning methods
• Selecting most informative unlabeled samples for efficient model training

Innovation

• Integrates semi-supervised and hierarchical active learning
• Uses weak-to-strong self-training to refine feature representations
• Selects samples via clustering for diversity and uncertainty