Classification of the PolSAR Data Using Dual Classifiers

The improvements in ship detection performance that can be obtained by using polarimetric synthetic aperture radar (SAR) data as compared to single-channel SAR data are investigated. Statistical decision theory is used to define decision variables that quantify the trade-off between the probability of missed detection and the probability of false alarm; performance is characterized by calculating receiver operating characteristics from single-channel and polarimetric SAR data by using likelihood ratio tests with the Neyman-Pearson criterion. It is shown that improvements in detection performance can be obtained with polarimetric SAR data as compared to single channel SAR data. We evaluate the results of these algorithms applied to single channel, dual channel amplitude only, dual channel with amplitude and phase, and fully polarimetric SAR data of known ships. In this way, the relative improvement in ship detection performance that is realized by introducing polarimetric information is quantified

This paper presents an application of the recent advances in the field of spherically invariant random vector (SIRV) modeling for coherency matrix estimation in heterogeneous clutter. The complete description of the polarimetric synthetic aperture radar (POLSAR) data set is achieved by estimating the span and the normalized coherency independently. The normalized coherency describes the polarimetric diversity, while the span indicates the total received power. The main advantages of the proposed fixed-point (FP) estimator are that it does not require any <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a priori</i> information about the probability density function of the texture (or span) and that it can directly be applied on adaptive neighborhoods. Interesting results are obtained when coupling this FP estimator with an adaptive spatial support based on the scalar span information. Based on the SIRV model, a new maximum-likelihood distance measure is introduced for unsupervised POLSAR classification. The proposed method is tested with both simulated POLSAR data and airborne POLSAR images provided by the Radar Ae¿roporte¿ Multi-Spectral d'Etude des Signatures system. Results of entropy/alpha/anisotropy decomposition, followed by unsupervised classification, allow discussing the use of the normalized coherency and the span as two separate descriptors of POLSAR data sets.

This letter presents a new estimation scheme for optimally deriving clutter parameters with high-resolution polarimetric synthetic aperture radar (POLSAR) data. The heterogeneous clutter in POLSAR data is described by the spherically invariant random vector model. Three parameters are introduced for the high-resolution POLSAR data clutter: the span, the normalized texture, and the speckle normalized covariance matrix. The asymptotic distribution of the novel span estimator is investigated. A novel heterogeneity test for the POLSAR clutter is also discussed. The proposed method is tested with airborne POLSAR images provided by the Office National d'Études et de Recherches Aerospatiales Radar Aéroporté Multi-spectral d'Etude des Signatures system.

Multi-Layer Perceptrons (MLPS) are powerful function approximators. In the last decades they were successfully applied to many different regression and classification problems. Their characteristics and convergence properties are well studied and relatively well understood, but they were originally designed to work with real-valued data. The main focus of this paper is the classification of polarimetric synthetic aperture radar (POLSAR) data which are a complex-valued signal. Instead of using an arbitrarily projection of this complex-valued data to the real domain, the paper proposes the usage of complex-valued MLPS (CV-MLPS), which are an extension of MLPS to the complex domain. The paper provides a detailed yet general derivation of the complex backpropagation algorithm and mentions related problems as well as possible solutions. Furthermore, it evaluates the performance of CV-MLPS in a land-cover classification task in POLSAR images under several learning conditions, and compares the proposed classifier with standard methods. The experimental results show that CV-MLPS are successfully applicable to classification tasks in POLSAR data. They show good convergence properties and a better performance if compared to real-valued MLPS.

An unsupervised segmentation method of polarimetric synthetic aperture radar (POLSAR) data is proposed in this paper. Firstly, this method performs initial classification of the POLSAR data based on polarimetric target decomposition. Then a model for the conditional distribution of the polarimetric coherency matrix is combined with a Markov random field (MRF) representation for the distribution of the region label to obtain the posterior distribution. Finally, iterated conditional modes (ICM) method is used to obtain the final segmentation results. A new MRF model with variable parameter is used in this method. Experiment results indicate that the proposed method achieves improvement over standard MRF method. (4 pages)

Hyperspectral images and polarimetric synthetic aperture radar (PolSAR) data are two important data sources, yet they barely appear under the same scope, even though multi-modal data fusion is attracting more and more attention. To our best knowledge, this paper investigates for the first time semi-supervised manifold alignment (SSMA) for the fusion of the hyperspectral image and PolSAR data. The SSMA searches a latent space where different data sources are aligned, which is accomplished by using the label information and the topological structure of the data. This paper is the first attempt to apply topological data analysis (TDA), a recent mathematic sub-field of data analysis, in remote sensing. It aims to reveal relevant information from the shape of a data in its feature space, and has been proven powerful in medicine. The paper also proposes a novel algorithm, MAPPER-regularized manifold alignment, which embeds the TDA into a semi-supervised manifold alignment for the fusion of the hyper-spectral image and PolSAR data. The proposed algorithm exhibits superior performance in fusing a simulated EnMAP data set and a Sentinel-1 data set for an image of Berlin.

Building orientation with respect to the radar look direction has a critical influence on the interpretation of multilook polarimetric synthetic aperture radar (PolSAR) data in urban areas. In this paper, its impacts on polarimetric orientation angle (POA) estimation and model-based decomposition are discussed. The discussion begins with the analysis of the general double-bounce scattering model, of which the characteristics are dependent on the electromagnetic and geometric parameters of the related dihedral structure. Then, for multilook PolSAR data, the polarimetric scattering mechanism in urban areas is modeled by two double-bounce scatterings from two orthogonal dihedral structures. From the model, the impacts of the building orientation on POA estimation can be revealed. With the increase of the building orientation, the POA difference between the two dihedral structures increases gradually, and the feasibility to estimate the building orientation via the estimated POA is reduced dramatically. Upon further analysis, we illustrate the impacts on the model-based decomposition. With the increase of the building orientation, the dominant scattering mechanism labeling technique based on the model-based decompositions will gradually become invalid. Moreover, the processing of POA compensation, which is helpful in reducing the impacts of the building orientation, also becomes invalid when the building orientation increases to a certain value. At last, three L-band data sets of San Francisco acquired by AIRSAR are used to verify the inferences. The experimental results show that, for L-band PolSAR data in urban areas, when the radar look angle is around 45, the threshold of building orientation for the validity of dominant scattering mechanism labeling is about ±3, and for the POA compensation, the threshold is about ±12.

This article proposes a novel autoencoder (AE) network based on the distribution of polarimetric synthetic aperture radar (POL-SAR) data matrix, called a mixture autoencoder (MAE). Through a detailed analysis of the data distribution POL-SAR data matrix, a normalization method is also presented in succession. The proposed MAE defines the data error term in the loss function according to the data distribution. It can be regarded as a process of unsupervised feature extraction designed specifically for POL-SAR data matrix. Then, a softmax classifier is trained with the help of data features and the corresponding label information. Next, a stacked MAE (SMAE) network is reasonably constructed by considering the data distribution among different layers. Finally, this article also presents a classification network through discarding the decoder process of the proposed SMAE and connecting with a softmax classifier. The SMAE is trained layer by layer using the unlabeled data. The softmax classifier is also trained with a small number of labeled pixels. With parameters obtained from the above-mentioned procedures as the initial parameters, the whole classification network is trained by the labeled pixels to get a well-trained model, which is used for predicting the corresponding label of the pixel in the data set. Three real POL-SAR data sets, including the AIR-SAR L-band data of Flevoland, The Netherlands, are used in the experiments. Compared with one classical algorithm and two related models with the similar structure, both the proposed methods show improvements in overall accuracy and efficiency as well as possess better adaptability of the parameter and preferable consistency with the classification performance.

The classification of urban areas in polarimetric synthetic aperture radar (PolSAR) data is a challenging task. Moreover, urban structures oriented away from the radar line of sight pose an additional complexity in the classification process. The characterization of such areas is important for disaster relief and urban sprawl monitoring applications. In this paper, a novel technique based on deep learning is proposed, which leverages a synthetic target database for data augmentation. The PolSAR dataset is rotated by uniform steps and collated to form a reference database. A stacked autoencoder network is used to transform the information in the augmented dataset into a compact representation. This significantly improves the generalization capabilities of the network. Finally, the classification is performed by a multilayer perceptron network. The modular architecture allows for easy optimization of the hyperparameters. The synthetic target database is created and the classification performance is evaluated on an Lband airborne UAVSAR dataset and L-band space-borne ALOS-2 dataset acquired over San Francisco, USA. The proposed technique shows an overall accuracy of 91.3%. An improvement over state-of-the-art techniques is achieved, especially in urban areas rotated away from the radar line of sight.

Abstract. In recent years, the use of Polarimetric Synthetic Aperture Radar (PolSAR) data in different applications dramatically has been increased. In SAR imagery an interference phenomenon with random behavior exists which is called speckle noise. The interpretation of data encounters some troubles due to the presence of speckle which can be considered as a multiplicative noise affecting all coherent imaging systems. Indeed, speckle degrade radiometric resolution of PolSAR images, therefore it is needful to perform speckle filtering on the SAR data type. Markov Random Field (MRF) has proven to be a powerful method for drawing out eliciting contextual information from remotely sensed images. In the present paper, a probability density function (PDF), which is fitted well with the PolSAR data based on the goodness-of-fit test, is first obtained for the pixel-wise analysis. Then the contextual smoothing is achieved with the MRF method. This novel speckle reduction method combines an advanced statistical distribution with spatial contextual information for PolSAR data. These two parts of information are combined based on weighted summation of pixel-wise and contextual models. This approach not only preserves edge information in the images, but also improves signal-to-noise ratio of the results. The method maintains the mean value of original signal in the homogenous areas and preserves the edges of features in the heterogeneous regions. Experiments on real medium resolution ALOS data from Tehran, and also high resolution full polarimetric SAR data over the Oberpfaffenhofen test-site in Germany, demonstrate the effectiveness of the algorithm compared with well-known despeckling methods.

Polarimetric synthetic aperture radar (PolSAR) image interpretation is widely used in various fields. Recently, deep learning has made significant progress in PolSAR image classification. Supervised learning (SL) requires a large amount of labeled PolSAR data with high quality to achieve better performance, however, manually labeled data are insufficient. This causes the SL to fail into overfitting and degrades its generalization performance. Furthermore, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">scattering confusion</i> problem is also a significant challenge that attracts more attention. To solve these problems, this article proposes a heterogeneous contrastive learning method (HCLNet). It aims to learn high-level representation from unlabeled PolSAR data for few-shot classification according to multifeatures and superpixels. Beyond the conventional CL, HCLNet introduces the heterogeneous architecture for the first time to utilize heterogeneous PolSAR features better. And it develops two easy-to-use plugins to narrow the domain gap between optics and PolSAR, including feature filter and superpixel-based instance discrimination, which the former is used to enhance the complementarity of multifeatures, and the latter is used to increase the diversity of negative samples. Experiments demonstrate the superiority of HCLNet on three widely used PolSAR benchmark datasets compared with state-of-the-art methods. Ablation studies also verify the importance of each component. Besides, this work has implications for how to efficiently utilize the multifeatures of PolSAR data to learn better high-level representation in CL and how to construct networks suitable for PolSAR data better.

When working with multilook fully polarimetric synthetic aperture radar (SAR) data, an appropriate way of representing the backscattered signal consists of the so-called covariance matrix. For each pixel, this is a 3/spl times/3 Hermitian positive definite matrix that follows a complex Wishart distribution. Based on this distribution, a test statistic for equality of two such matrices and an associated asymptotic probability for obtaining a smaller value of the test statistic are derived and applied successfully to change detection in polarimetric SAR data. In a case study, EMISAR L-band data from April 17, 1998 and May 20, 1998 covering agricultural fields near Foulum, Denmark are used. Multilook full covariance matrix data, azimuthal symmetric data, covariance matrix diagonal-only data, and horizontal-horizontal (HH), vertical-vertical (VV), or horizontal-vertical (HV) data alone can be used. If applied to HH, VV, or HV data alone, the derived test statistic reduces to the well-known gamma likelihood-ratio test statistic. The derived test statistic and the associated significance value can be applied as a line or edge detector in fully polarimetric SAR data also.

When working with multi-look fully polarimetric synthetic aperture radar (SAR) data an appropriate way of representing the backscattered signal consists of the so-called covariance matrix. For each pixel this is a 3/spl times/3 Hermitian, positive definite matrix which follows a complex Wishart distribution. Based on this distribution a test statistic for equality of two such matrices and an associated asymptotic probability for obtaining a smaller value of the test statistic are given and applied to change detection in polarimetric SAR data. In a case study EMISAR L-band data from 17 April 1998 and 20 May 1998 covering agricultural fields near Foulum, Denmark are used. The derived test statistic can be applied as a line or edge detector in fully polarimetric SAR data also.

Crop classification is one of the most significant applications for Polarimetric Synthetic Aperture Radar (PolSAR) data. Owing to the limited information obtained by the single-temporal PolSAR data, the multi-temporal data are used in this paper to further provide ample information within various crop growing stages. However, the polarization scattering decomposition for multi-temporal PolSAR data easily causes &#x201C;dimension disaster&#x201D;. Hence, a neural network of sparse auto-encoder with non-negativity constraints (NC-SAE) was employed to compress the data, yielding efficient features for accurate classification. Then, a novel classifier of multi-scale feature classification network (MSFCN) was constructed to improve the classification performance, which is proved to be superior to the popular classifiers of convolutional neural networks (CNN) and supper vector machine (SVM). The performances of the proposed method were evaluated and compared with the traditional methods by using simulated Sentinel-1 data provided by European Space Agency (ESA). The classification results indicate that the proposed methodology is promising for practical use in agricultural applications.

In this paper, we propose an object-based fusion approach for the joint use of polarimetric synthetic aperture radar (PolSAR) and hyperspectral data. The proposed approach extracts information from both datasets based on an object-level, which is used here for land use classification. The achieved classification result infers that the proposed methodology improves the classification performance of both hyperspectral and PolSAR data and can properly gather complementary information of the two kinds of dataset. The fusion approach also considers that only limited training samples are available, which is often the case in remote sensing.