基于极化相似性特征的极化SAR图像的谱分类

针对极化SAR图像分类存在的问题,提出了基于SAR目标的极化特征的二维谱聚类方法。该方法可以充分考虑目标的极化相似性特征,利用二维的谱聚类方法实现极化SAR图像的分类。它以两目标散射的极化相似性参数图像作为输入特征,用二维图权函数代替一维图权函数求权值,使采样点分类和特征矢量分类相一致,从而实现极化SAR图像的分类。实验结果表明,该方法具有更好的分类结果,明显优于K均值分类。     

基于Yamaguchi分解模型的全极化SAR图像分类

针对利用Yamaguchi分解模型的四个散射分量直接进行类别归属判断精度不高并且所分类别有限的问题,结合模糊C均值的理论,提出了一种基于Yamaguchi分解模型的全极化SAR分类算法,把四个散射分量组成一组归一化的特征矢量,进行FCM聚类分析。并且用日本机载L波段PiSAR数据验证了该算法具有较高的分类精度和较好的视觉效果。     

SAR images analysis based on polarimetric signatures

In the presented paper a new method of identification of canonical coherent scatterers in the quad-polarimetric SAR data are presented. The proposed method is based on the analysis of polarimetric signatures. The observed signatures are compared with the polarimetric signatures of four canonical objects: trihedral, dihedral and helix – right and left which represent basic scattering mechanisms: single bounce, double bounce and helix scattering. The polarimetric matrices are treated as vectors in a unitary space with a scalar product that generates the norm. A recognized object is classified to one of the four coherent classes by a Kohonen network. It is not trained in an iteration process but its weights are adjusted according to the given patterns. The network classification is supported by rules. The obtained maps of pixels that represent canonical objects are compared with a map of coherent scatterers which was obtained by using the polarimetric entropy approach. The developed method of canonical coherent scatterers identification based on the polarimetric signatures analysis allows us not only to identify precisely the canonical coherent scatterers but also to determine the type of scattering mechanism characteristic for each of them. Since the proposed method works on a single-look (non-averaged) SAR data, it does not cause any spatial nor spectral decrease of amount of information because averaging is not conducted. Moreover, the proposed method will enable us the identification of a type of scattering mechanism in the canonical coherent pixels. This is an improvement in comparison to the existing methods. The obtained results should be more precise because the full polarimetric information about the scatterers is used in the identification procedure.     

Applications of ICA for the enhancement and classification of polarimetric SAR images

Independent components analysis (ICA) based methods for polarimetric synthetic aperture radar (SAR) image speckle reduction and ground object classification are studied. Several independent components can be extracted from polarimetric SAR images using ICA directly. The component with lowest speckle index is regarded as the scene after speckle reduction. The disadvantage of this method is that only one image is kept and most polarization information will be lost. In this paper, we use ICA‐sparse‐coding shrinkage (ICA‐SPS) based speckle reduction method, which is implemented on each individual image and can keep polarization information. It is carried out on the combined channels obtained by Pauli‐decomposition rather than original polarization channels in order to keep relative phase information among polarization channels and get better performance. After ICA‐SPS, the effect of speckle suppression on SAR image classification can be compared favourably with other methods by combining the channels into a false colour image. At last, a new ICA‐based classification method is presented. In this method, four independent components are separated by ICA from five polarization and combined channels. One of these independent components which includes little ground object information is regarded as speckle noise and therefore be discarded. The remaining three components can be treated as subordination coefficients of three kinds of targets. A classified image can be obtained based on the components. And by composing these three channels in RGB colour pattern, a false colour image can be constructed.     

Evolutionary RBF classifier for polarimetric SAR images

In this paper, a robust radial basis function (RBF) network based classifier is proposed for polarimetric synthetic aperture radar (SAR) images. The proposed feature extraction process utilizes the covariance matrix elements, the H/α/A decomposition based features combined with the backscattering power (span), and the gray level co-occurrence matrix (GLCM) based texture features, which are projected onto a lower dimensional feature space using principal components analysis. For the classifier training, both conventional backpropagation (BP) and multidimensional particle swarm optimization (MD-PSO) based dynamic clustering are explored. By combining complete polarimetric covariance matrix and eigenvalue decomposition based pixel values with textural information (contrast, correlation, energy, and homogeneity) in the feature set, and employing automated evolutionary RBF classifier for the pattern recognition unit, the overall classification performance is shown to be significantly improved. An experimental study is performed using the fully polarimetric San Francisco Bay and Flevoland data sets acquired by the NASA/Jet Propulsion Laboratory Airborne SAR (AIRSAR) at L-band to evaluate the performance of the proposed classifier. Classification results (in terms of confusion matrix, overall accuracy and classification map) compared with the major state of the art algorithms demonstrate the effectiveness of the proposed RBF network classifier.     

基于商空间粒度的极化SAR图像分类

当前极化合成孔径雷达(SAR)图像的分类研究中,极化信息的不完全利用是影响极化SAR图像分类效果的重要原因之一。故将商空间粒度合成理论引入到极化SAR图像分类中,通过建立不同的支持向量机(SVM)分类器构建不同的商空间,从多个粒度层面实现对极化信息的综合利用。首先通过不同的极化分解方法得到不同的极化特征,分别对其建立不同的支持向量机分类器进行分类;再根据粒度合成理论对这些商空间进行融合,得到更细粒度上的改进的分类结果。最后,利用AIRSAR图像进行实验比较,算法改进后的结果在地物误分上有明显的抑制,各类别分类正确率都有所提高。     

Adaptive unsupervised classification of polarimetric SAR images using the improved affinity propagation clustering algorithm

A robust automatic classification system is critical for polarimetric synthetic aperture radar (POLSAR) terrain processing. In most of the conventional classification methods, the number of classes could not be calculated before classification. In this article, we present a new unsupervised classification algorithm with an adaptive number of classes for POLSAR data which is capable of estimating the class numbers automatically. The approach is mainly composed of three operations. First, region-based feature map combining the polarimetric statistical and spatial information is constructed based on the turbopixel method. This is followed by a clustering step performed through an improved affinity propagation clustering with Wishart distance. Finally, the result of the improved affinity propagation clustering is classified using Wishart classifier. The proposed approach takes the spatial information into consideration and makes good use of the inherent statistical characteristics of POLSAR data. The performance of the proposed classification approach on three real datasets is presented and analysed, and the experimental results show that the approach provides more accurate estimation under the condition of various numbers of classes compared with existing methods.     

A novel measure of target scattering randomness with application to polarimetric SAR image classification

A novel measure of target scattering randomness, called the average degree of randomness (ADoR), is introduced in this article. The proposed parameter ADoR is based on the degrees of polarization of the scattered waves using orthogonally polarized incident waves. Combining the ADoR and the Freeman decomposition, which is applied to discriminate the dominant scattering mechanism of the target, a new scheme for unsupervised classification of polarimetric synthetic aperture radar (PolSAR) images is designed. Considering that the preset intervals of the randomness measure may not fit the data distribution, an iterative classification method is developed. The effectiveness of the randomness measure and the proposed methods is demonstrated using a National Aeronautics and Space Administration (NASA)/Jet Propulsion Laboratory (JPL) AIRborne Synthetic Aperture Radar (AIRSAR) PolSAR image.     

Study on segmentation and interpretation of multi-look polarimetric SAR images

The segmentation and interpretation of multi-look polarimetric synthetic aperture radar (SAR) images is studied. We first introduce a multi-look polarimetric whitening filter (MPWF) to reduce the speckle in multi-look polarimetric SAR images. Then, by utilizing the wavelet multiresolution approach to extract the texture information in different scales and the Markov random field (MRF) model to characterize the spatial constraints between pixels in each scale level, a multiresolution segmentation algorithm (MSA) to segment the speckle-reduced SAR images is presented. The MSA first segments the image at the lowest resolution level and then proceeds to progressively higher resolutions until individual pixels are well classified. An unsupervised step to estimate both the optimal number of texture classes and their model parameters is also included in the MSA so that the segmentation can be implemented without supervision. Finally, in order to interpret the results of the unsupervised segmentation and to understand the whole polarimetric SAR image, we develop an image interpretation approach which jointly utilizes the scattering mechanism identification and target decomposition approaches. Experimental results with the real-world multi-look polarimetric SAR image demonstrate the effectiveness of the segmentation and interpretation approaches.     

Investigation of the effect of the incidence angle on land cover classification using fully polarimetric SAR images

Incidence angle is one of the most important imaging parameters that affect polarimetric SAR (PolSAR) image classification. Several studies have examined the land cover classification capability of PolSAR images with different incidence angles. However, most of these studies provide limited physical insights into the mechanism how the variation of incidence angle affects PolSAR image classification. In the present study, land cover classification was conducted by using RADARSAT-2 Wide Fine Quad-Pol (FQ) images acquired at different incidence angles, namely, FQ8 (27.75°), FQ14 (34.20°), and FQ20 (39.95°). Land cover classification capability was examined for each single-incidence angle image and a multi-incidence angle image (i.e., the combination of single-incidence angle images). The multi-incidence angle image produced better classification results than any of the single-incidence angle images, and the different incidence angles exhibited different superiorities in land cover classification. The effect mechanisms of incidence angle variation on land cover classification were investigated by using the polarimetric decomposition theorem that decomposes radar backscatter into single-bounce scattering, double-bounce scattering and volume scattering. Impinging SAR easily penetrated crops to interact with the soil at a small incidence angle. Therefore, the difference in single-bounce scattering between trees and crops was evident in the FQ8 image, which was determined to be suitable for distinguishing between croplands and forests. The single-bounce scattering from bare lands increased with the decrease in incidence angles, whereas that from water changed slightly with the incidence angle variation. Consequently, the FQ8 image exhibited the largest difference in single-bounce scattering between bare lands and water and produced the fewest confusion between them among all the images. The single- and double-bounce scattering from urban areas and forests increased with the decrease in incidence angles. The increase in single- and double-bounce scattering from urban areas was more significant than that from forests because C-band SAR could not easily penetrate the crown layer of forests to interact with the trunks and ground. Therefore, the FQ8 image showed a slightly better performance than the other images in discriminating between urban areas and forests. Compared with other crops and trees, banana trees caused stronger single- and double-bounce scattering because of their large leaves. As a large incidence angle resulted in a long penetration path of radar waves in the crown layer of vegetation, the FQ20 image enhanced the single- and double-bounce scattering differences between banana trees and other vegetation. Thus, the FQ20 image outperformed the other images in identifying banana trees.     

Similarity-enhanced target detection algorithm using polarimetric SAR images

Target detection and analysis using polarimetric synthetic aperture radar (PolSAR) images are currently of great interest in synthetic aperture radar (SAR) applications. For a complex target, the scattering characteristics are determined by different independent sub-scatterers and their interaction; therefore, the scattering characteristics should be described by a statistical method due to randomness and depolarization. Furthermore, the inherent speckle in SAR data must be reduced by spatial averaging at the expense of loss of spatial resolution. The polarimetric similarity parameter (PSP) is an effective parameter to analyse target characteristics. In order to describe a complex distributed target, two new methods for calculating PSP are proposed, namely Stokes matrix-based PSP (S-PSP) and multiple PolSAR similarity parameter (MPSP). The characteristics of a target can be described and extracted on the basis of the polarimetric similarity, and then the similarity-enhanced target detection methods using S-PSP and MPSP are implemented and demonstrated with German Aerospace Centre (DLR) experimental SAR L-band multiple temporal PolSAR images of Oberpfaffenhofen test site (DE), Germany. The results confirmed that the proposed methods are effective for detection and analysis of buildings in urban areas.     

Classification of polarimetric SAR images using adaptive neighbourhood structures

Two detail-preserving classification algorithms for polarimetric SAR images are proposed and their performance are evaluated on polarimetric complex SAR images. Neighbourhood structures are adaptively selected for modelling the polarimetric amplitudes and the region labels, and for achieving detailpreservation. Experimental results obtained from multi-frequency polarimetric SAR images show that the novel schemes produce significant visual improvements for detail preservation, and exhibit equivalent or higher classification performance with respect to usual classification schemes.     

.基于纹理和边缘的SAR图像SVM分类*

为实现SAR图像地物目标的有效分类,深入研究了基于灰度共生矩阵GLCM的四种纹理特征以及两个边缘特征。分析每个单独纹理或边缘特征在对SAR图像进行支持向量机SVM分类中对不同地物的分辨能力,选取不同的特征组合进行组合特征的SVM分类实验。对各种特征进行主成分分析PCA,并考察使用和不使用PCA两种情况下分类结果之间的差异。实验结果证明能量、边缘长度、对比度和相关度的特征组合在PCA作用下能够改善各类地物的分类精度,将总分类精度提高到90%以上。     

基于非下采样Contourlet变换的方向Teager能量的极化SAR图像融合*

提出一种基于非下采样Contourlet变换和方向Teager能量的极化SAR图像融合算法。采用具有多尺度、多方向和平移不变性特点的非下采样Contourlet变换对多个单极化强度图像进行分解,然后高频子带图像分别按行和列进行Teager能量计算,选取Teager能量作为度量来提取区域边缘与纹理信息。对于低频系数采用平均融合算法,根据高频子图Teager能量分布差异,对于方向高频系数采用不同最优加权算法实现极化图像的融合处理。实验结果表明,提出的算法与PWF算法相比在保留原始图像边缘和纹理信息的同时,可以有效地抑制相干斑噪声的影响,取得较好的融合视觉效果。     

A novel ship classification approach for high resolution SAR images based on the BDA-KELM classification model

Ship classification based on synthetic aperture radar (SAR) images is a crucial component in maritime surveillance. In this article, the feature selection and the classifier design, as two key essential factors for traditional ship classification, are jointed together, and a novel ship classification model combining kernel extreme learning machine (KELM) and dragonfly algorithm in binary space (BDA), named BDA-KELM, is proposed which conducts the automatic feature selection and searches for optimal parameter sets (including the kernel parameter and the penalty factor) for classifier at the same time. Finally, a series of ship classification experiments are carried out based on high resolution TerraSAR-X SAR imagery. Other four widely used classification models, namely k-Nearest Neighbour (k-NN), Bayes, Back Propagation neural network (BP neural network), Support Vector Machine (SVM), are also tested on the same dataset. The experimental results shows that the proposed model can achieve a better classification performance than these four widely used models with an classification accuracy as high as 97% and encouraging results of other three multi-class classification evaluation metrics.     

高阶条件随机场引导的多分支极化SAR图像分类

目的 针对极化合成孔径雷达（polarimetric synthetic aperture radar,PolSAR）小样本分类问题,基于充分挖掘有限样本的极化、空间特征考虑,提出一种由高阶条件随机场（conditional random field,CRF）引导的多分支分类网络模型。方法 利用Yamaguchi非相干目标分解方法,构建每个像素的极化特征向量。设计了由高阶CRF能量函数引导的多卷积分支特征提取网络,将像素点极化特征向量作为输入,分别提取像素点的像素特征、邻域特征和位置特征信息。将以上特征进行加和融合,并输入到softmax分类器中得到预分类结果。利用超像素方法对预分类结果图进行进一步修正和调优,平滑相邻像素之间的特异性和相似性。结果 采用1%的采样率对两组真实的极化SAR数据进行测试。同时,为了更好地模拟实际应用中训练样本位置分布不均匀的情况,考虑了空间不相交采样方法作为对比实验。综合两种采样策略的实验结果表明,相较于只利用像素级特征或简单利用空间特征的方法,本文方法总分类精度平均提升7%～10%,不同地物类别的分类精准度均在90%以上,运行速度相比于支持向量机（support vector machine,SVM）提高了2.5倍以上。结论 通过构建高阶CRF引导的卷积神经网络,将像素特征信息、同质区域特征和地理位置信息进行融合,有效建立了像素级和对象级数据之间的尺度关联,进一步扩充了像素点之间的空间依赖性,提取到了更强大更准确的表征特征,显著提高了标记样本数量较少情况下的卷积网络模型的分类性能,进一步保证了地物目标散射机制表征的全面性和可靠性。     

Comparison of polarimetric SAR observables in terms of classification performance

The advent of fully polarimetric systems has led to an increased amount of information acquired by synthetic aperture radar (SAR) sensors but also to an increased complexity of the data to be analysed and interpreted. In particular, the choice of several representations of the data, in terms of different parameters with peculiar characteristics and physical meaning, has been offered. With this work, we intend to address their systematic investigation with a twofold goal: (1) to provide a brief review of the polarimetric representations under consideration; and (2) to characterize and compare them with respect to their usefulness for classification purposes.

The analysis procedure consists of the accuracy estimation of classification tests performed on different parameters derived from L‐band polarimetric SAR data. In order to ensure a common basis for their comparison, a neural network classifier, the Multi‐Layer Perceptron trained by the Back‐Propagation learning rule, was used which permits us to operate on the data without making any a priori assumption on their statistics. In this way, the considered polarimetric parameters, in general characterized by different statistical distributions, may undergo the same classification process and the results compared.

Our results indicate that the overall classification performance varies depending on the polarimetric parameters used. However, these variations are relatively limited and do not permit us, at this stage, to define an ‘absolute’ best representation to identify the classes under investigation in an optimal way.     

Modeling multilook polarimetric SAR images with heavy-tailed rayleigh distribution and novel estimation based on matrix log-cumulants

With the improvements in modern radar resolution,the Gaussian-fluctuation model based on the central limit theorem does not accurately describe the scattering echo from targets.In contrast,the heavytailed Rayleigh distribution,based on the generalized central limit theorem,performs well in modeling the synthetic aperture radar(SAR) images,whereas its application to multi-look image processing is difficult.We describe successful modeling of multilook polarimetric SAR images with the heavy-tailed Rayleigh distribution and present novel parameter estimators based on matrix log-cumulants for the heavy-tailed Rayleigh distribution including the equivalent number of looks(ENL).First,a compound variable of heavy-tailed Rayleigh distribution is divided into a product of a positive alpha-stable variable and a complex Gaussian variable.The parameter estimations of the characteristic exponent and scale parameter based on log-cumulants in a single polarization channel are then derived.Second,the matrix log-cumulants(MLCs) for full polarization in multilook images are obtained,which can be applied to estimate model parameters.Therefore,a novel ENL estimator based on MLC is presented that describes the model more precisely.Extended to all other multivariable product models,this estimator performs better than existing methods.Finally,calculations on both simulated and real data are performed that give good fits with theoretical results.Multilook processing in one image with a fixed pixel number can improve parameter estimations over single-look processing.Our heavy-tailed Rayleigh model with its parameter estimation provides a new method to analyze the multilook polarimetric SAR images for target detection and classification.     

A study of land cover classification using polarimetric SAR parameters

Classification of the Earth's surface constitutes an important application of polarimetric synthetic aperture radar (SAR) data; in turn, it may represent an efficient way for investigating their different representations. The polarimetric parameters most frequently taken into account for classification have been the incoherent ones. A similar use of coherent methods appears to have been scarcely considered and remained neglected. In this contribution, we wish to address this issue, testing and comparing a wide range of polarimetric SAR parameters, coherent and incoherent. Another original aspect of this work is the study of the dependence of the classification results on the varying size of averaging windows of pixels. Such an analysis will permit us to evaluate the importance of speckle reduction and to prove if the chosen polarimetric parameters describe only point‐like physical properties of the targets or if they also contain ‘extended’, local information. The goal is to provide an objective estimate of the quality of the classification of polarimetric parameters and afford their comparison, an exercise hitherto unavailable in the literature in common knowledge.     

深度置信网络在极化SAR图像分类中的应用

目的 深度置信网络能够从数据中自动学习、提取特征,在特征学习方面具有突出优势。极化SAR图像分类中存在海量特征利用率低、特征选取主观性强的问题。为了解决这一问题,提出一种基于深度置信网络的极化SAR图像分类方法。方法 首先进行海量分类特征提取,获得极化类、辐射类、空间类和子孔径类四类特征构成的特征集;然后在特征集基础上选取样本并构建特征矢量,用以输入到深度置信网络模型之中;最后利用深度置信网络的方法对海量分类特征进行逐层学习抽象,获得有效的分类特征进行分类。结果 采用AIRSAR数据进行实验,分类结果精度达到91.06%。通过与经典Wishart监督分类、逻辑回归分类方法对比,表现了深度置信网络方法在特征学习方面的突出优势,验证了方法的适用性。结论 针对极化SAR图像海量特征的选取与利用,提出了一种新的分类方法,为极化SAR图像分类提供了一种新思路,为深度置信网络获得更广泛地应用进行有益的探索和尝试。