Perceive, Act and Correct: Confidence Is Not Enough for Hyperspectral Classification
This addresses classification accuracy issues in hyperspectral imaging, particularly for applications with limited labeled data, but is incremental as it builds on existing semi-supervised and uncertainty-aware techniques.
The paper tackles the problem of misleading confidence in hyperspectral image classification, which causes confirmation bias under sparse annotations or class imbalance, and proposes CABIN, a semi-supervised framework that improves labeling efficiency and performance for state-of-the-art methods.
Confidence alone is often misleading in hyperspectral image classification, as models tend to mistake high predictive scores for correctness while lacking awareness of uncertainty. This leads to confirmation bias, especially under sparse annotations or class imbalance, where models overfit confident errors and fail to generalize. We propose CABIN (Cognitive-Aware Behavior-Informed learNing), a semi-supervised framework that addresses this limitation through a closed-loop learning process of perception, action, and correction. CABIN first develops perceptual awareness by estimating epistemic uncertainty, identifying ambiguous regions where errors are likely to occur. It then acts by adopting an Uncertainty-Guided Dual Sampling Strategy, selecting uncertain samples for exploration while anchoring confident ones as stable pseudo-labels to reduce bias. To correct noisy supervision, CABIN introduces a Fine-Grained Dynamic Assignment Strategy that categorizes pseudo-labeled data into reliable, ambiguous, and noisy subsets, applying tailored losses to enhance generalization. Experimental results show that a wide range of state-of-the-art methods benefit from the integration of CABIN, with improved labeling efficiency and performance.