Superpixel-based graph convolutional network (SGCN) has shown the advantages of less computational time and global modeling ability for polarimetric synthetic aperture radar (PolSAR) image classification. However, the effectiveness is heavily dependent on the superpixel segmentation result. Existing superpixel segmentation methods usually produce edge errors due to speckle and scattering confusion, which directly results in the mistakes of the final classification. To address this issue, a novel hybrid weighted fuzzy SGCN method(HF-SGCN) is proposed to correct the edge pixels by defining a fuzzy projection matrix (FPM). The FPM can transform features from superpixel to pixel, by which features of edge pixels can be calculated from all the neighboring superpixels with a certain probability, so as to correct edges to the most similar region. In addition, a hybrid weighted adjacent matrix is formulated by incorporating both the revised Wishart and multi-feature distances, which can enhance the discriminating features effectively. The proposed HF-SGCN method is capable of capturing the global contextual information and rectifying edges, while disregarding the local individual features for each pixel. To combine global and local features, we further propose the HF-SGCN-CNN method, which integrates the superpixel-wise HF-SGCN network and the pixel-wise 3D-CNN network into a unified framework. Thus, we can fuse the features extracted from two subnetworks, producing complementary global and local features that significantly improve classification accuracy. Experiments are conducted on four publicly real PolSAR datasets with different sensors and bands. Experimental results demonstrate the effectiveness of the proposed methods.

Polarimetric synthetic aperture radar (PolSAR) image classification is one of the most important study areas for PolSAR image processing. Many kinds of PolSAR features can be extracted for PolSAR image classification, such as the scattering, polarimetric or image features. However, it is difficult to improve the classification accuracy of PolSAR images by using all these low-level features directly, since they may conflict with each other for classification. Hence, how to joint learn these low-level features to obtain high-level discriminating features is a challenging task. To solve this problem, a novel fast multi-feature joint learning method(fMF-JLC) is proposed for PolSAR image classification. The proposed method extract three kinds of low-level features of PolSAR data at first. Then, a multi-feature joint sparse representation model(MF-JSR) is proposed by designing joint sparse constraints on the extracted features above. Moreover, the joint sparse features are further compressed to overcome the dimension curse and acquire semantic features by the topic model. By this way, the low-level features are fused and discriminating high-level features are acquired. However, the pixel-wise feature learning method is time consuming. To speed the proposed method, a superpixel-based fast learning method is designed by involving the contextual relationship. Experiments are taken on three sets of real PolSAR data with different sensors and bands, and several compared methods are used to verify the effectiveness of the proposed method. The experimental results illustrate that the proposed method can obtain better performance than the state-of-art methods, especially for the heterogenous areas.

Deep learning methods excel in Polarimetric SAR (PolSAR) image classification. However, existing methods typically sample an image block for each pixel with a fixed-size square window, which always contains inconsistent/incomplete content with the central pixel, resulting in many misclassifications especially in boundary and heterogeneous regions. So, a size-fixed square window is not enough for representing various terrain objects. To address this issue, we develop a content-adaptive multi-region deep network to obtain contextual consistent sampling windows for diverse terrain objects. Firstly, a complex scene of PolSAR image is partitioned into homogeneous, heterogeneous and boundary regions. Then, sampling windows with adaptive direction and scale are designed for three distinct regions. Besides, windows with central and global regions are proposed to provide additional local and global information. Finally, a fusion network is designed to adaptively combine different sampling windows to enhance classification performance. Experimental results on three real data sets demonstrate that the proposed method can achieve superior performance in both edge details and heterogeneous terrain objects compared with the state-of-the-art methods.

The edge detection plays an important role in post-processing of PolSAR images. It is still a great challenge for extracting all the edge features and suppress speckle noises, especially when weak/strong edges appear simultaneously outside and within heterogenous areas. In this paper, a novel hybrid edge detection framework is proposed to address this problem. The proposed method is designed by fusing two initial edge detectors, which can detect complementary edge information. One is an improved polarimetric constant false alarm rate (IP_CFAR) edge detector, which can detect weak edges well, but fail to detect the edges in the heterogeneous regions. The other is the proposed weighted gradient-based (WG) detector which can detect edges in heterogeneous areas well, but loses some weak edges and produces some false edges due to the speckle noises. Secondly, based on the two detectors above, a wavelet-based hybrid edge detection method is proposed by combining their merits and suppress their shortcomings. To fuse them effectively, wavelet transformation is utilized and semantic rules are defined to extract their advantages. Moreover, a despeckling scheme is designed to suppress the false edges in the wavelet domain. Experimental results demonstrate that the proposed method outperforms the state-of-art methods in extracting both weak edges and strong edges within heterogeneous regions.

Nearest-regularized subspace (NRS) algorithm is a kind of effective representation learning method, which can obtain both accuracy and speed for PolSAR image classification. However, existing NRS methods use the polarimetric feature vector instead of the PolSAR original coherency matrix (known as Hermitian positive definite (HPD) matrix) as the input. This will destroy the matrix structure, and miss the instinct correlation among channels. How to utilize the original coherency matrix to NRS method is a key problem. To address this limitation, a Riemannian NRS method is proposed, which considers the HPD matrices endowed in a Riemannian space. First, to utilize the PolSAR original data, a Riemannian NRS method (RNRS) is proposed by constructing HPD dictionary and HPD distance metric. Then, a new Tikhonov regularization term is designed to reduce the differences within the same class. Finally, the optimization method is developed to resolve the proposed model, and the first-order derivative is inferred. Besides, only coherency matrix T is used as the input in the proposed method, while multiple features are utilized for compared methods in the experiments. Experimental results demonstrate the proposed method can outperform the state-of-the-art algorithms even with fewer features.

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.