Enhancing Knowledge Transfer in Hyperspectral Image Classification via Cross-scene Knowledge Integration

2025-12-07

AI Summary

Cross-scene hyperspectral image classification faces two key challenges: spectral shifts induced by sensor discrepancies and semantic inconsistencies across heterogeneous scenes; moreover, existing methods struggle with extreme domain adaptation where source and target domains share no class labels. To address these issues, we propose Cross-scene Knowledge Integration (CKI), the first framework explicitly designed for such cross-scene transfer. CKI employs a domain-invariant spectral projection to align spectral features, introduces a Source Similarity Mechanism (SSM) to guide preference-aware cross-scene knowledge sharing, and incorporates multi-granularity alignment with adaptive complementary fusion to explicitly model and integrate target-specific knowledge. Evaluated on multiple cross-scene benchmarks, CKI significantly enhances transfer stability and achieves state-of-the-art performance—improving average accuracy by 9.3% under the zero-class-overlap setting.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Scene Analysis & Understanding
• Intelligent Robots: Multimodal Perception & Sensor Fusion

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: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Knowledge transfer has strong potential to improve hyperspectral image (HSI) classification, yet two inherent challenges fundamentally restrict effective cross-domain transfer: spectral variations caused by different sensors and semantic inconsistencies across heterogeneous scenes. Existing methods are limited by transfer settings that assume homogeneous domains or heterogeneous scenarios with only co-occurring categories. When label spaces do not overlap, they further rely on complete source-domain coverage and therefore overlook critical target-private information. To overcome these limitations and enable knowledge transfer in fully heterogeneous settings, we propose Cross-scene Knowledge Integration (CKI), a framework that explicitly incorporates target-private knowledge during transfer. CKI includes: (1) Alignment of Spectral Characteristics (ASC) to reduce spectral discrepancies through domain-agnostic projection; (2) Cross-scene Knowledge Sharing Preference (CKSP), which resolves semantic mismatch via a Source Similarity Mechanism (SSM); and (3) Complementary Information Integration (CII) to maximize the use of target-specific complementary cues. Extensive experiments verify that CKI achieves state-of-the-art performance with strong stability across diverse cross-scene HSI scenarios.

Problem

• Address spectral variations from different sensors in HSI classification
• Resolve semantic inconsistencies across heterogeneous hyperspectral scenes
• Incorporate target-private knowledge when label spaces do not overlap

Innovation

• Domain-agnostic projection reduces spectral discrepancies
• Source similarity mechanism resolves semantic mismatches
• Integrates target-specific complementary cues for transfer