A Lightweight Riemannian Covariance Matrix Convolutional Network for PolSAR Image Classification
- 15 July 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-17
Abstract
Recently, deep learning methods have achieved superior performance for polarimetric synthetic aperture radar (PolSAR) image classification. Existing deep learning methods learn PolSAR data by converting the covariance matrix into a feature vector or complex-valued vector as the input, learning features in Euclidean space. However, it is well-known that covariance matrices are manifold data endowing in Riemannian space instead of Euclidean space. Existing methods cannot learn the geometric characteristics of covariance matrices directly and destroy the channel correlation. To learn features from covariance matrices directly, we propose a lightweight Riemannian covariance matrix convolutional network (LRCM_CNN) for PolSAR classification for the first time, which directly utilizes the covariance matrix as the network input and defines the Riemannian operations to learn complex matrix’s features in Riemannian space. The proposed LRCM_CNN network initially designs a lightweight Riemannian covariance matrix network (LRCMnet) to learn covariance matrix features by exploiting a series of Riemannian convolution, rectified linear unit (ReLu), and LogEig operations in Riemannian space, which breaks through the Euclidean constraint of conventional networks. Then, features learned from covariance matrices are converted from Riemannian to Euclidean space, and a CNN module is appended to enhance contextual covariance matrix features. Besides, a fast kernel learning method is developed for the proposed method to learn class-specific features and reduce the computation time effectively, which implements the lightweight RCMnet. Experiments are conducted on four sets of real PolSAR data with different bands and sensors. Experiments results demonstrate the proposed method can obtain superior performance than the state-of-the-art methods.Keywords
Funding Information
- National Natural Science Foundation of China (62006186, 62272383, 62372369)
- Youth Innovation Team Research Program Project of the Education Department in Shaanxi Province (23JP111)
This publication has 62 references indexed in Scilit:
- Unsupervised classification of polarimetric SAR images via SVMM with extended variational inferenceISPRS Journal of Photogrammetry and Remote Sensing, 2023
- Fruits yield estimation using Faster R-CNN with MIoUMultimedia Tools and Applications, 2021
- Covariance descriptors on a Gaussian manifold and their application to image set classificationPattern Recognition, 2020
- PolSAR image classification via a novel semi-supervised recurrent complex-valued convolution neural networkNeurocomputing, 2020
- Dense connection and depthwise separable convolution based CNN for polarimetric SAR image classificationKnowledge-Based Systems, 2020
- Complex Contourlet-CNN for polarimetric SAR image classificationPattern Recognition, 2020
- A Novel Multi-Feature Joint Learning Method for Fast Polarimetric SAR Terrain ClassificationIEEE Access, 2020
- Positive definite matrices and the S-divergenceProceedings of the American Mathematical Society, 2015
- : Two-directional two-dimensional PCA for efficient face representation and recognitionNeurocomputing, 2005
- A three-component scattering model for polarimetric SAR dataIEEE Transactions on Geoscience and Remote Sensing, 1998