A Semantics-Aware Hierarchical Self-Supervised Approach to Classification of Remote Sensing Images

2025-10-06

AI Summary

Existing remote sensing image classification methods often fail to effectively model the inherent hierarchical structure of class labels. Method: This paper proposes a semantic-aware hierarchical self-supervised learning framework that constructs multi-granularity classification heads and a learnable hierarchy matrix; it discovers inter-class hierarchical relationships via a hierarchical consensus mechanism in a fully self-supervised manner—requiring no additional manual annotations—and enforces cross-level probabilistic consistency through a hierarchical loss function and a weighted ensemble strategy. The framework is backbone-agnostic and supports modeling complex semantic hierarchies. Contribution/Results: Evaluated on three remote sensing benchmark datasets with varying hierarchical complexities, the proposed method achieves significant improvements in classification accuracy and robustness, demonstrating strong generalization capability and model adaptability.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• 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: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Deep learning has become increasingly important in remote sensing image classification due to its ability to extract semantic information from complex data. Classification tasks often include predefined label hierarchies that represent the semantic relationships among classes. However, these hierarchies are frequently overlooked, and most approaches focus only on fine-grained classification schemes. In this paper, we present a novel Semantics-Aware Hierarchical Consensus (SAHC) method for learning hierarchical features and relationships by integrating hierarchy-specific classification heads within a deep network architecture, each specialized in different degrees of class granularity. The proposed approach employs trainable hierarchy matrices, which guide the network through the learning of the hierarchical structure in a self-supervised manner. Furthermore, we introduce a hierarchical consensus mechanism to ensure consistent probability distributions across different hierarchical levels. This mechanism acts as a weighted ensemble being able to effectively leverage the inherent structure of the hierarchical classification task. The proposed SAHC method is evaluated on three benchmark datasets with different degrees of hierarchical complexity on different tasks, using distinct backbone architectures to effectively emphasize its adaptability. Experimental results show both the effectiveness of the proposed approach in guiding network learning and the robustness of the hierarchical consensus for remote sensing image classification tasks.

Problem

• Integrating hierarchical class relationships into remote sensing image classification
• Addressing overlooked semantic hierarchies in deep learning classification methods
• Ensuring consistent probability distributions across hierarchical classification levels

Innovation

• Hierarchical classification heads for multi-granularity learning
• Trainable matrices guide self-supervised hierarchical structure learning
• Consensus mechanism ensures probability consistency across levels