Hierarchical polarimetric SAR image classification based on feature selection and Genetic algorithm

In this letter, we propose a theoretically and computationally simple feature for synthetic aperture radar (SAR) image classification, the multilevel local pattern histogram (MLPH). The MLPH describes the size distributions of bright, dark, and homogenous patterns appearing in a moving window at various contrasts; these patterns are the elementary properties of SAR image texture. The MLPH is a very powerful descriptor of SAR images because it captures both local and global structural information. Additionally, it is robust to speckle noise. Experiments on a TerraSAR-X data set demonstrate that MLPH significantly outperforms four other widely used features in SAR image classification.

Synthetic aperture radar (SAR) image classification is an important part in the understanding and interpretation of SAR images. Each patch in SAR images has a scene category, but usually contains multiple land-cover classes or latent properties, which can be represented by topics in the probabilistic topic model (PTM). The representation and selection of discriminative features in PTM have a large impact on the classification results. Most of the existing feature learning methods do not make full use of high-level structure feature and the feature correlation within similar images to mine discriminative features. Therefore, this article proposes a discriminative sketch topic model with structural constraint (C-SSTM) for SAR image classification. In the proposed model, each image patch is characterized by structural and texture features. In particular, the sketch structural feature is based on the sketch map to represent the image local structure pattern. Then, the local image manifold information is preserved in terms of structure and texture. In the structural constraint, the texture and structure of each image patch are combined to learn discriminative latent semantic topics between image patches. Finally, each image patch is quantified by discriminative latent semantic topics instead of low-level representation. The experimental results tested on synthetic and real SAR images demonstrate that the proposed C-SSTM is able to learn effective structural feature representation from SAR images. Compared with other related approaches, C-SSTM produces competitive classification accuracies with high time efficiency.

Recently, Convolutional Neural Network (CNN) has achieved some success in synthetic aperture radar (SAR) image classification. This outstanding performance mainly depends on a large number of training samples, but achieved satisfactory classification results with limited training samples remains a challenge. To address this problem, we propose a dual-scale Siamese densely connected network with Markov Random Fields (DS-SDCNet-MRF) for single-polarization SAR image classification. First, the Siamese densely connected network (SDCNet) is proposed to fully extract discriminative features under limited samples. Then, the proposed DS-SDCNet is constructed with two SDCNets of different scales to produce complementary classification results. Among them, the large-scale SDCNet has a better classification result in the homogeneous region, while the small-scale SDCNet tends to provide good detail preservation. Finally, an improved MRF model which combined the category probability information is proposed to further improve the classification performance. Experimental results on simulated and real single-polarization SAR data demonstrate that the proposed method achieves more encouraging classification performance than the current state-of-the-art classification methods with limited samples.

Deep learning models for Synthetic Aperture Radar (SAR) image classification are highly vulnerable to adversarial perturbations, which can significantly compromise their reliability in critical applications.In this study, we investigate the impact of adversarial attacks on SAR image classification by generating imperceptible perturbations using the Fast Gradient Sign Method (FGSM).These perturbations are designed to deceive a pre-trained convolutional neural network (CNN) into making incorrect classifications.To mitigate this vulnerability, we employ Entropy-Based Bayesian Model Averaging (EB-BMA), which is an adaptive ensemble learning approach based on model uncertainty.We fine-tune multiple CNN architectures using transfer learning from ImageNet and apply EB-BMA to dynamically adjust model contributions based on prediction confidence.This adaptive weighting mechanism ensures that models with higher certainty contribute more to the final decision, while uncertain models are downweighted.Our results demonstrate that EB-BMA enhances resilience against adversarial perturbations, as validated on the MSTAR dataset.

Thanks to the availability of large-scale data, deep Convolutional Neural Networks (CNNs) have witnessed success in various applications of computer vision. However, the performance of CNNs on Synthetic Aperture Radar (SAR) image classification is unsatisfactory due to the lack of well-labeled SAR data, as well as the differences in imaging mechanisms between SAR images and optical images. Therefore, this paper addresses the problem of SAR image classification by employing the Generative Adversarial Network (GAN) to produce more labeled SAR data. We propose special GANs for generating SAR images to be used in the training process. First, we incorporate the quadratic operation into the GAN, extending the convolution to make the discriminator better represent the SAR data; second, the statistical characteristics of SAR images are integrated into the GAN to make its value function more reasonable; finally, two types of parallel connected GANs are designed, one of which we call PWGAN, combining the Deep Convolutional GAN (DCGAN) and Wasserstein GAN with Gradient Penalty (WGAN-GP) together in the structure, and the other, which we call CNN-PGAN, applying a pre-trained CNN as a discriminator to the parallel GAN. Both PWGAN and CNN-PGAN consist of a number of discriminators and generators according to the number of target categories. Experimental results on the TerraSAR-X single polarization dataset demonstrate the effectiveness of the proposed method.

A method for synthetic aperture radar (SAR) image classification in urban areas based on modified Unit-linking pulse coupled neural networks (Unit-linking PCNN) and texture feature is presented. Unit-linking PCNN is modified to be two levels in order to make it classify more classes. The primary level corresponds to determining the initial threshold value of the secondary level, and in the secondary level, the similar neurons are captured using Unit-linking PCNN. Because of the imaging characteristic of SAR building areas, the texture feature of the neuron’s n ×n window image is used as the input pulse signal. Experimental results show that the proposed method is effective.KeywordsSynthetic aperture radar (SAR)Image classificationUnit-linking PCNNUrban areas

The classification of high-resolution (HR) synthetic aperture radar (SAR) images is of great importance for SAR scene interpretation and application. However, the presence of intricate spatial structural patterns and complex statistical nature makes SAR image classification a challenging task, especially in the case of limited labeled SAR data. This paper proposes a novel HR SAR image classification method, using a multi-scale deep feature fusion network and covariance pooling manifold network (MFFN-CPMN). MFFN-CPMN combines the advantages of local spatial features and global statistical properties and considers the multi-feature information fusion of SAR images in representation learning. First, we propose a Gabor-filtering-based multi-scale feature fusion network (MFFN) to capture the spatial pattern and get the discriminative features of SAR images. The MFFN belongs to a deep convolutional neural network (CNN). To make full use of a large amount of unlabeled data, the weights of each layer of MFFN are optimized by unsupervised denoising dual-sparse encoder. Moreover, the feature fusion strategy in MFFN can effectively exploit the complementary information between different levels and different scales. Second, we utilize a covariance pooling manifold network to extract further the global second-order statistics of SAR images over the fusional feature maps. Finally, the obtained covariance descriptor is more distinct for various land covers. Experimental results on four HR SAR images demonstrate the effectiveness of the proposed method and achieve promising results over other related algorithms.

In this review paper, several studies and researches were surveyed for assisting future researchers to identify available techniques in the field of classification of Synthetic Aperture Radar (SAR) images. SAR images are becoming increasingly important in a variety of remote sensing applications due to the ability of SAR sensors to operate in all types of weather conditions, including day and night remote sensing for long ranges and coverage areas. Its properties of vast planning, search, rescue, mine detection, and target identification make it very attractive for surveillance and observation missions of Earth resources. With the increasing popularity and availability of these images, the need for machines has emerged to enhance the ability to identify and interpret these images effectively. This is due to the fact that SAR image processing requires the formation of an image from the measured radar scatter returns, followed by a treatment to discover and define the image's composition. After reviewing several previous studies that succeeded in achieving a classification of SAR images for specific goals, it became obvious that they could be generalized to all types of SAR images. The most prominent use of Convolutional Neural Networks (CNN) was successful in extracting features from the images and training the neural network to analyze and classify them into classes according to these features. The dataset used in this model was obtained from the Moving and Stationary Target Acquisition and Recognition (MSTAR) database, which consists of a set of SAR images of military vehicles, for which the application of the CNN approach achieved a final accuracy of 97.91% on ten different classes.

Automatic recognition of skin symptom plays an importance role in the skin diagnosis and treatment. Feature selection is to increase the classification performance of skin symptom. In this paper, the effects of feature selection on the classification of 4-class skin symptoms (chloasma, blackhead, freckled and comedone) are analyzed. Support vector machine (SVM) is employed to construct classifier, the parameters setting is very crucial to the learning results and generalization ability of SVM, and many irrelevant and redundant features degrade the performance of classification, furthermore, the choice of SVM parameters is influenced by the feature subset taken into account and vice versa. So a joint optimization method for selecting features and SVM parameters is proposed. Genetic algorithms, which have a lot of improvements about fitness, crossover and mutation in comparison with simple genetic algorithms, is to achieve a balance between the classification accuracy and the size of the feature subsets selected and have a fast convergence rate. Furthermore, ‘one-against-one’ approach is used to solve multi-class classification problems and a comparison of the model performances is made among the results of the joint optimization and the single optimization of features or SVM parameters in the experiments. The simulation result shows that our approach can find subsets of features for 4-class skin symptoms classification with higher average cross validation accuracy and a higher rate of convergence compared with single optimization of features or SVM parameters.

Convolutional neural networks (CNNs) have been applied to learn spatial features for high-resolution (HR) synthetic aperture radar (SAR) image classification. However, there has been little work on integrating the unique statistical distributions of SAR images which can reveal physical properties of terrain objects, into CNNs in a supervised feature learning framework. To address this problem, a novel end-to-end supervised classification method is proposed for HR SAR images by considering both spatial context and statistical features. First, to extract more effective spatial features from SAR images, a new deep spatial context encoder network (DSCEN) is proposed, which is a lightweight structure and can be effectively trained with a small number of samples. Meanwhile, to enhance the diversity of statistics, the nonstationary joint statistical model (NS-JSM) is adopted to form the global statistical features. Specifically, SAR images are transformed into the Gabor wavelet domain and the produced multi-subbands magnitudes and phases are modeled by the log-normal and uniform distribution. The covariance matrix is further utilized to capture the inter-scale and intra-scale nonstationary correlation between the statistical subbands and make the joint statistical features more compact and distinguishable. Considering complementary advantages, a feature fusion network (Fusion-Net) base on group compression and smooth normalization is constructed to embed the statistical features into the spatial features and optimize the fusion feature representation. As a result, our model can learn the discriminative features and improve the final classification performance. Experiments on four HR SAR images validate the superiority of the proposed method over other related algorithms.

The results of an experimental investigation of the use of textural features computed from the gray-level co-occurrence matrix for synthetic aperture radar (SAR) image classification are reported and discussed. The investigation, carried out on SAR images acquired with the SIR-C/X-SAR sensor in an Italian agricultural area, makes it possible to derive interesting information about the computation modalities and the effectiveness of the above textural features.

Presents a symbolic approach to synthetic aperture radar (SAR) image classification. The symbolic classifier presented has as input data samples of predefined groups (the learning set) represented in a usual table of values. The learning step organises a complete and discriminant description of each group, using either symbolic objects or disjunctions of symbolic objects. These disjunctions are obtained using symbolic operators and a mutual neighbourhood graph. The classification rule is based on dissimilarity functions between a standard description of an individual and a symbolic object which describes a group. To show the usefulness of this approach in SAR image analysis, images in amplitude format and three and eight looks, showing areas with different degrees of roughness are simulated assuming the multiplicative model. Segmentation of these images, after using the sigma Lee filter to diminish speckle, was performed with a standard region growing algorithm and these segments were used as the input of the symbolic classifier. A Monte Carlo experience is devised, and the feasibility of the symbolic approach is quantitatively assessed.

Computer-aided diagnosis has the potential of increasing diagnostic accuracy by providing a second reading to radiologists. In many computerized schemes, numerous features can be extracted to describe suspect image regions. A subset of these features is then employed in a data classifier to determine whether the suspect region is abnormal or normal. Different subsets of features will, in general, result in different classification performances. A feature selection method is often used to determine an "optimal" subset of features to use with a particular classifier. A classifier performance measure (such as the area under the receiver operating characteristic curve) must be incorporated into this feature selection process. With limited datasets, however, there is a distribution in the classifier performance measure for a given classifier and subset of features. In this paper, we investigate the variation in the selected subset of "optimal" features as compared with the true optimal subset of features caused by this distribution of classifier performance. We consider examples in which the probability that the optimal subset of features is selected can be analytically computed. We show the dependence of this probability on the dataset sample size, the total number of features from which to select, the number of features selected, and the performance of the true optimal subset. Once a subset of features has been selected, the parameters of the data classifier must be determined. We show that, with limited datasets and/or a large number of features from which to choose, bias is introduced if the classifier parameters are determined using the same data that were employed to select the "optimal" subset of features.

Classification of the polarimetric synthetic aperture radar (PolSAR) data is an important facet of synthetic aperture radar (SAR) image interpretation. In general, many polarimetric features can be extracted from PolSAR data. However, not all features are benefit for classification and too many polarimetric features result in the low classification efficiency. We introduce a method to deal with many polarimetric features and obtain both high classification accuracy and high classification efficiency. We first preprocess the PolSAR data to obtain objects, i.e., homogeneous regions. Next, we apply polarimetric feature processing to compute various polarimetric features of the PolSAR data. Based on the classification of the sample information, we then select the prominent polarimetric features using the data mining attributes of decision tree algorithm (DTA). Finally, we obtain classification results through the selected polarimetric features by DTA, the sample data and support vector machine (SVM). We verified the proposed method using the AirSAR L-Band PolSAR data. Our experiments indicate that the classification accuracy of the proposed method is equivalent to that of SVM and the computational efficiency is comparable to that of DTA. Thus, the proposed method integrates the advantages of both DTA and SVM.

Synthetic aperture radar (SAR) image classification is a challenging problem due to the complex imaging mechanism as well as the random speckle noise, which affects radar image interpretation. Recently, deep neural networks (DNNs) have been shown to outperform previous state-of-the-art techniques in computer vision tasks owing to their ability to learn relevant features from the data. However, the fragility of these models has received far less academic attention in the remote sensing community, which limits our understanding of the security of remote sensing image classification models. To explore the basic characteristic of adversarial examples of SAR images, we compare several mainstream adversarial methods and evaluate the securities of used DNNs from the perspective of attention. We subsequently perform other attempts. The experimental results provide data support and an effective reference for the defense capabilities of various DNNs regarding attack in SAR image classification models.