Riemannian Complex Hermit Positive Definite Convolution Network for Polarimetric SAR Image Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Junfei, Li, Yuke, Nie, Mengmeng, Liu, Fang, Jin, Haiyan, Li, Junhuai, Lin, Weisi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915372864634880
author Shi, Junfei
Li, Yuke
Nie, Mengmeng
Liu, Fang
Jin, Haiyan
Li, Junhuai
Lin, Weisi
author_facet Shi, Junfei
Li, Yuke
Nie, Mengmeng
Liu, Fang
Jin, Haiyan
Li, Junhuai
Lin, Weisi
contents Deep learning has been extensively utilized for PolSAR image classification. However, most existing methods transform the polarimetric covariance matrix into a real- or complex-valued vector to comply with standard deep learning frameworks in Euclidean space. This approach overlooks the inherent structure of the covariance matrix, which is a complex Hermitian positive definite (HPD) matrix residing in the Riemannian manifold. Vectorization disrupts the matrix structure and misrepresents its geometric properties. To mitigate this drawback, we propose HPDNet, a novel framework that directly processes HPD matrices on the Riemannian manifold. The HPDnet fully considers the complex phase information by decomposing a complex HPD matrix into the real- and imaginarymatrices. The proposed HPDnet consists of several HPD mapping layers and rectifying layers, which can preserve the geometric structure of the data and transform them into a more separable manifold representation. Subsequently, a complex LogEig layer is developed to project the manifold data into a tangent space, ensuring that conventional Euclidean-based deep learning networks can be applied to further extract contextual features for classification. Furthermore, to optimize computational efficiency, we design a fast eigenvalue decomposition method for parallelized matrix processing. Experiments conducted on three real-world PolSAR datasets demonstrate that the proposed method outperforms state-of-the-art approaches, especially in heterogeneous regions.
format Preprint
id arxiv_https___arxiv_org_abs_2502_08137
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Riemannian Complex Hermit Positive Definite Convolution Network for Polarimetric SAR Image Classification
Shi, Junfei
Li, Yuke
Nie, Mengmeng
Liu, Fang
Jin, Haiyan
Li, Junhuai
Lin, Weisi
Computer Vision and Pattern Recognition
Deep learning has been extensively utilized for PolSAR image classification. However, most existing methods transform the polarimetric covariance matrix into a real- or complex-valued vector to comply with standard deep learning frameworks in Euclidean space. This approach overlooks the inherent structure of the covariance matrix, which is a complex Hermitian positive definite (HPD) matrix residing in the Riemannian manifold. Vectorization disrupts the matrix structure and misrepresents its geometric properties. To mitigate this drawback, we propose HPDNet, a novel framework that directly processes HPD matrices on the Riemannian manifold. The HPDnet fully considers the complex phase information by decomposing a complex HPD matrix into the real- and imaginarymatrices. The proposed HPDnet consists of several HPD mapping layers and rectifying layers, which can preserve the geometric structure of the data and transform them into a more separable manifold representation. Subsequently, a complex LogEig layer is developed to project the manifold data into a tangent space, ensuring that conventional Euclidean-based deep learning networks can be applied to further extract contextual features for classification. Furthermore, to optimize computational efficiency, we design a fast eigenvalue decomposition method for parallelized matrix processing. Experiments conducted on three real-world PolSAR datasets demonstrate that the proposed method outperforms state-of-the-art approaches, especially in heterogeneous regions.
title Riemannian Complex Hermit Positive Definite Convolution Network for Polarimetric SAR Image Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2502.08137