ECP-Mamba: An Efficient Multi-scale Self-supervised Contrastive Learning Method with State Space Model for PolSAR Image Classification

TL;DR

ECP-Mamba introduces a resource-efficient, self-supervised contrastive learning framework with a state space model backbone for accurate PolSAR image classification, reducing reliance on labeled data and improving computational efficiency.

Contribution

It proposes a novel multi-scale self-supervised contrastive learning method combined with a tailored state space model for efficient PolSAR classification.

Findings

Achieves 99.70% accuracy on Flevoland 1989 dataset

Balances high accuracy with resource efficiency

Outperforms existing methods on benchmark datasets

Abstract

Recently, polarimetric synthetic aperture radar (PolSAR) image classification has been greatly promoted by deep neural networks. However,current deep learning-based PolSAR classification methods encounter difficulties due to its dependence on extensive labeled data and the computational inefficiency of architectures like Transformers. This paper presents ECP-Mamba, an efficient framework integrating multi-scale self-supervised contrastive learning with a state space model (SSM) backbone. Specifically, ECP-Mamba addresses annotation scarcity through a multi-scale predictive pretext task based on local-to-global feature correspondences, which uses a simplified self-distillation paradigm without negative sample pairs. To enhance computational efficiency,the Mamba architecture (a selective SSM) is first tailored for pixel-wise PolSAR classification task by designing a spiral scan strategy. This strategy prioritizes causally relevant features near the central pixel, leveraging the localized nature of pixel-wise classification tasks. Additionally, the lightweight Cross Mamba module is proposed to facilitates complementary multi-scale feature interaction with minimal overhead. Extensive experiments across four benchmark datasets demonstrate ECP-Mamba's effectiveness in balancing high accuracy with resource efficiency. On the Flevoland 1989 dataset, ECP-Mamba achieves state-of-the-art performance with an overall accuracy of 99.70%, average accuracy of 99.64% and Kappa coefficient of 99.62e-2. Our code will be available at https://github.com/HaixiaBi1982/ECP_Mamba.