Perceive, Act and Correct: Confidence Is Not Enough for Hyperspectral Classification

TL;DR

This paper introduces CABIN, a semi-supervised learning framework for hyperspectral classification that improves model reliability by estimating uncertainty, selectively sampling data, and dynamically correcting noisy labels to enhance accuracy and efficiency.

Contribution

The paper presents a novel semi-supervised framework, CABIN, that incorporates uncertainty estimation, dual sampling, and dynamic label correction for hyperspectral image classification.

Findings

CABIN improves classification accuracy across multiple models.

Enhanced labeling efficiency demonstrated in experiments.

State-of-the-art methods benefit from CABIN integration.

Abstract

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.