Towards Interpretable PolSAR Image Classification: Polarimetric Scattering Mechanism Informed Concept Bottleneck and Kolmogorov-Arnold Network

2025-07-04

AI Summary

To address the limited interpretability of deep learning models in Polarimetric Synthetic Aperture Radar (PolSAR) image classification, this paper proposes a novel solution integrating polarimetric physical mechanisms with an interpretable neural architecture. Methodologically: (1) we design a Parallel Concept Bottleneck Network (PaCBN) that explicitly maps deep features to physically grounded scattering concepts—specifically, Polarimetric Target Decomposition (PTD) components—used as supervisory labels; and (2) we replace conventional multilayer perceptrons (MLPs) with Kolmogorov–Arnold Networks (KANs), leveraging learnable spline functions to model transparent, nonlinear concept-to-class mappings. Experiments on multiple benchmark PolSAR datasets demonstrate that our approach achieves state-of-the-art accuracy while providing strong interpretability: it disentangles physically meaningful, traceable scattering concepts and yields explicit, mathematically analyzable decision functions—thereby significantly enhancing model transparency and trustworthiness.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Interpretability, Explainability, and Transparency
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

In recent years, Deep Learning (DL) based methods have received extensive and sufficient attention in the field of PolSAR image classification, which show excellent performance. However, due to the ``black-box" nature of DL methods, the interpretation of the high-dimensional features extracted and the backtracking of the decision-making process based on the features are still unresolved problems. In this study, we first highlight this issue and attempt to achieve the interpretability analysis of DL-based PolSAR image classification technology with the help of Polarimetric Target Decomposition (PTD), a feature extraction method related to the scattering mechanism unique to the PolSAR image processing field. In our work, by constructing the polarimetric conceptual labels and a novel structure named Parallel Concept Bottleneck Networks (PaCBM), the uninterpretable high-dimensional features are transformed into human-comprehensible concepts based on physically verifiable polarimetric scattering mechanisms. Then, the Kolmogorov-Arnold Network (KAN) is used to replace Multi-Layer Perceptron (MLP) for achieving a more concise and understandable mapping process between layers and further enhanced non-linear modeling ability. The experimental results on several PolSAR datasets show that the features could be conceptualization under the premise of achieving satisfactory accuracy through the proposed pipeline, and the analytical function for predicting category labels from conceptual labels can be obtained by combining spline functions, thus promoting the research on the interpretability of the DL-based PolSAR image classification model.

Problem

• Achieve interpretability in DL-based PolSAR image classification
• Transform high-dimensional features into human-comprehensible scattering concepts
• Enhance model transparency using KAN for interpretable layer mapping

Innovation

• Uses Polarimetric Target Decomposition for interpretability
• Implements Parallel Concept Bottleneck Networks
• Applies Kolmogorov-Arnold Network for clearer mapping