PolSAR Image Research Articles

High-resolution fully-polarimetric synthetic aperture radar dataset

Abstract Objectives Fully-polarimetric synthetic aperture radar (PolSAR) data contain a rich body of elementary scattering physics information that is critically valuable for a broad range of applications and scientific purposes. However, there is a lack of available high-resolution (< 0.3048-m) data available for PolSAR phenomenology research. This article introduces a high-resolution PolSAR data set collected and provided by Sandia National Laboratories (SNL). The data sets were collected to support studying high-resolution scattering physics from different types of clutter and applications such as polarimetric-based terrain classification. Data description The data set presented consists of 39 200 m × 200 m complex-valued PolSAR image sets with 0.12-m (~ 4.7-inch) resolution and is the result of two different collection campaigns, one campaign in 2013 and the other in 2015. The data sets were collected with SNL’s airborne fully-polarimetric X-band (9.6 GHz center frequency) radar system and contains a variety of natural and cultural clutter environments with scene content ranging from agricultural fields to urbanized areas collected throughout the greater Albuquerque, New Mexico, U.S.A. region.

Channel adaptive CVFCN using a new transfer method for PolSAR terrain classification

ABSTRACT This paper mainly addresses the lack of labelled data and insufficient data utilization in PolSAR image classification. We propose a channel adaptive Complex-Valued Fully Convolutional Networks based on Transfer learning (TF-CVFCN). To solve the problem of less labelled data, we borrowed from the transfer learning method, which can use a large amount of easily obtained relevant or even irrelevant data for pre-training and then use a small amount of source data for network training fine-tuning. Since PolSAR data can obtain different data information by using different decomposition methods, the current research trend is to process PolSAR data in the complex-valued containing more information. In our work, to cater to the proposed network model and make the network design more liberal, a new transfer learning method is proposed, which uses a new method to realize partial layer parameter transfer, and achieves the purpose of optimizing the traditional Transfer learning method. On this basis, we optimize the network structure to improve the classification accuracy of PolSAR data, and evaluate its performance on three datasets. Furthermore, we lighten the TF-CVFCN and optimize the relevant parameters using Competitive Multi-Objective Particle Swarm Optimizer, emphasizing the importance of this parameter optimization in large networks.

A temporal difference matrix for historical cumulative change detection in time series PolSAR data

To detect historical cumulative changes in land cover, this study introduces a novel time series PolSAR change detection method. The objective is to address the inefficiency of traditional detection methods that use multiple pairwise comparisons, as well as to alleviate false positives and false negatives caused by the insufficient utilization of polarization and spatial context information in previous methods. The proposed method initially constructs a temporal difference matrix for time series PolSAR images and computes the difference image (DI) using the maximum eigenvalue of this matrix to depict the cumulative changes over time. Subsequently, the DI is segmented using an MRF-based image segmentation method with a limited amount of supervised information, yielding the cumulative change binary map. The efficacy of the proposed method is validated using three time series datasets covering different scenes acquired by different sensors, namely GaoFen3, Radarsat2 and Sentinel-1. The experimental results demonstrate that the method effectively characterizes cumulative changes in the time series while simultaneously considering disparities in radar backscatter intensity and polarization correlation. As a result, it significantly reduces erroneous and missed change detections and surpasses the existing methods by achieving the highest Recall, F1-score, IOU, Kappa, and overall accuracy. Furthermore, the DI calculation method serves as a versatile change descriptor applicable not only for capturing cumulative changes in time series but also for describing changes between dual temporal images.

Performance analysis of SAR filtering techniques using SVM and Wishart Classifier

The main objective of this work is to analyze the performance of different filtering techniques used to remove speckle noise from PolSAR images. The proposed work shall also provide an insight on the use of various filters on PolSAR images with regards to preservation of edges, point target detection and image classification. Quad-Pol, fully polarimetric SAR Data acquired form the ALOS PalSAR 1 Satellite of the San Francisco Region, USA is used in the study. The SAR data which is in the form of a scattering Matrix (S-Matrix) is first converted to a Coherence Matrix (T-Matrix). The T-Matrix has been used since it is the most suitable matrix representation of scattering and includes consideration of the phase of various polarization combinations. Different filters are applied on the T-Matrix. The filters used are Box Car Filter, Gaussian Filter, IDAN Filter, Lee Refined Filter, Lee Sigma Filter, Lopez Filter, Non-Local Means Filter, and Scattering Model Based Filter. The performance of the different filtering techniques is evaluated using the most used Support Vector Machines (SVM) with Radial Basis function as the kernel with cross validations and Wishart classifiers. The SVM and Wishart classifiers are considered in this study since they give highest accuracies and require the least time for training. A short comparative study is also presented by comparing SVM classifier with seven others to justify the use of SVM Classifier. Three different classes, namely Water, Forest and settlements are considered. Scores like individual class accuracies, Overall Accuracy (OA), Kappa Score, Precision, & Recall are used to determine the classification performance after each filtering technique. Box car filter has the highest overall accuracy of 99.5% after applying SVM Classifier. IDAN filter has the highest overall accuracy of 97.68% after applying Wishart Classifier on it. However, classification carried out on the Boxcar filtered image not only gives high overall accuracy but also maintains edges and point targets.

PolSAR Image Classification with Active Complex-Valued Convolutional-Wavelet Neural Network and Markov Random Fields

PolSAR image classification has attracted extensive significant research in recent decades. Aiming at improving PolSAR classification performance with speckle noise, this paper proposes an active complex-valued convolutional-wavelet neural network by incorporating dual-tree complex wavelet transform (DT-CWT) and Markov random field (MRF). In this approach, DT-CWT is introduced into the complex-valued convolutional neural network to suppress the speckle noise of PolSAR images and maintain the structures of learned feature maps. In addition, by applying active learning (AL), we iteratively select the most informative unlabeled training samples of PolSAR datasets. Moreover, MRF is utilized to obtain spatial local correlation information, which has been proven to be effective in improving classification performance. The experimental results on three benchmark PolSAR datasets demonstrate that the proposed method can achieve a significant classification performance gain in terms of its effectiveness and robustness beyond some state-of-the-art deep learning methods.

Multiobjective Evolutionary Superpixel Segmentation for PolSAR Image Classification

Superpixel segmentation has been widely used in the field of computer vision. The generations of PolSAR superpixels have also been widely studied for their feasibility and high efficiency. The initial numbers of PolSAR superpixels are usually designed manually by experience, which has a significant impact on the final performance of superpixel segmentation and the subsequent interpretation tasks. Additionally, the effective information of PolSAR superpixels is not fully analyzed and utilized in the generation process. Regarding these issues, a multiobjective evolutionary superpixel segmentation for PolSAR image classification is proposed in this study. It contains two layers, an automatic optimization layer and a fine segmentation layer. Fully considering the similarity information within the superpixels and the difference information among the superpixels simultaneously, the automatic optimization layer can determine the suitable number of superpixels automatically by the multiobjective optimization for PolSAR superpixel segmentation. Considering the difficulty of the search for accurate boundaries of complex ground objects in PolSAR images, the fine segmentation layer can further improve the qualities of superpixels by fully using the boundary information of good-quality superpixels in the evolution process for generating PolSAR superpixels. The experiments on different PolSAR image datasets validate that the proposed approach can automatically generate high-quality superpixels without any prior information.

Multichannel semi-supervised active learning for PolSAR image classification

Deep neural networks have recently been extensively utilized for Polarimetric synthetic aperture radar (PolSAR) image classification. However, this heavily relies on extensive labeled data which is both costly and labor-intensive. To lower the collection of labeling data and enhance the classification performance, a novel multichannel semi-supervised active learning (MSSAL) method is proposed for PolSAR image classification. First, a multichannel strategy-based committee model with cooperative representation classification is presented to explore more effective information in the limited training data. Second, a loss prediction (LP) module is designed to identify the most informative pixels, and an ensemble learning (EL) strategy is designed to select the pixels with the highest confidence. Then, the deep neural network is fine-tuned with the obtaining target pixels through LP and EL in each iteration. Finally, the trained deep model predicts labels for all unlabeled data, outputting the final classification results. The proposed method is evaluated on three real-world PolSAR datasets, demonstrating superior performance to other PolSAR image classification methods with limited labeled samples.

Multi-scale contrastive learning method for PolSAR image classification

Multi-scale contrastive learning method for PolSAR image classification

Dual Attention-Based Global-Local Feature Extraction Network for Unsupervised Change Detection in PolSAR Images

Dual Attention-Based Global-Local Feature Extraction Network for Unsupervised Change Detection in PolSAR Images

A Novel Causal Inference-Guided Feature Enhancement Framework for PolSAR Image Classification

A Novel Causal Inference-Guided Feature Enhancement Framework for PolSAR Image Classification

Supervised Manifold Learning-based Polarimetric-spatial Feature Extraction for PolSAR Image Classification

Supervised Manifold Learning-based Polarimetric-spatial Feature Extraction for PolSAR Image Classification

A Lightweight Riemannian Covariance Matrix Convolutional Network for PolSAR Image Classification

A Lightweight Riemannian Covariance Matrix Convolutional Network for PolSAR Image Classification

PolSAR Image Classification Framework With POA Align and Cyclic Channel Attention

PolSAR Image Classification Framework With POA Align and Cyclic Channel Attention

MLR-SimSiam: A Contrastive Pretraining Model Based on Polarimetric Jittering and Mutual Learning Regularizer for PolSAR Image Classification

MLR-SimSiam: A Contrastive Pretraining Model Based on Polarimetric Jittering and Mutual Learning Regularizer for PolSAR Image Classification

SemiPSCN: Polarization Semantic Constraint Network for Semi-Supervised Segmentation in Large-Scale and Complex-Valued PolSAR Images

SemiPSCN: Polarization Semantic Constraint Network for Semi-Supervised Segmentation in Large-Scale and Complex-Valued PolSAR Images

Polarimetry-Inspired Contrastive Learning for Class-Imbalanced PolSAR Image Classification

Polarimetry-Inspired Contrastive Learning for Class-Imbalanced PolSAR Image Classification

Adversarial Network With Higher Order Potential Conditional Random Field for PolSAR Image Classification

Adversarial Network With Higher Order Potential Conditional Random Field for PolSAR Image Classification

Meta-Graph Representation Learning for PolSAR Image Classification

Meta-Graph Representation Learning for PolSAR Image Classification

A 3-D Convolutional Vision Transformer for PolSAR Image Classification and Change Detection

A 3-D Convolutional Vision Transformer for PolSAR Image Classification and Change Detection

Adaptive Superpixel Generation for Time-series PolSAR Images Considering Time-varying Characteristics

Adaptive Superpixel Generation for Time-series PolSAR Images Considering Time-varying Characteristics