Dual Classification Head Self-training Network for Cross-scene Hyperspectral Image Classification

2025-02-25

AI Summary

To address significant spectral shifts, large intra-class feature distribution discrepancies, and severe label scarcity in cross-scene hyperspectral image classification, this paper proposes the Dual-Head Self-Training Network (DHSNet). DHSNet introduces, for the first time in this task, a dual-branch self-training framework that jointly optimizes a class-level feature alignment loss and a center-aware feature attention module, effectively mitigating error accumulation in pseudo-labels and enhancing scene-invariant feature representation. Additionally, an adversarial domain alignment strategy is incorporated to further reduce inter-domain distribution divergence. Evaluated on three standard cross-scene benchmarks, DHSNet achieves an average overall accuracy improvement of 3.2–5.8 percentage points over state-of-the-art methods. The source code is publicly available.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Distributed Search

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Due to the difficulty of obtaining labeled data for hyperspectral images (HSIs), cross-scene classification has emerged as a widely adopted approach in the remote sensing community. It involves training a model using labeled data from a source domain (SD) and unlabeled data from a target domain (TD), followed by inferencing on the TD. However, variations in the reflectance spectrum of the same object between the SD and the TD, as well as differences in the feature distribution of the same land cover class, pose significant challenges to the performance of cross-scene classification. To address this issue, we propose a dual classification head self-training network (DHSNet). This method aligns class-wise features across domains, ensuring that the trained classifier can accurately classify TD data of different classes. We introduce a dual classification head self-training strategy for the first time in the cross-scene HSI classification field. The proposed approach mitigates domain gap while preventing the accumulation of incorrect pseudo-labels in the model. Additionally, we incorporate a novel central feature attention mechanism to enhance the model's capacity to learn scene-invariant features across domains. Experimental results on three cross-scene HSI datasets demonstrate that the proposed DHSNET significantly outperforms other state-of-the-art approaches. The code for DHSNet will be available at https://github.com/liurongwhm.

Problem

• Cross-scene hyperspectral image classification
• Domain gap in feature distribution
• Accumulation of incorrect pseudo-labels

Innovation

• Dual classification head self-training
• Central feature attention mechanism
• Cross-scene HSI classification enhancement