Change detection is one of the fundamental applications of synthetic aperture radar (SAR) images. However, speckle noise presented in SAR images has a negative effect on change detection, leading to frequent false alarms in the mapping products. In this research, a novel two-phase object-based deep learning approach is proposed for multi-temporal SAR image change detection. Compared with traditional methods, the proposed approach brings two main innovations. One is to classify all pixels into three categories rather than two categories: unchanged pixels, changed pixels caused by strong speckle (false changes), and changed pixels formed by real terrain variation (real changes). The other is to group neighbouring pixels into superpixel objects such as to exploit local spatial context. Two phases are designed in the methodology: (1) Generate objects based on the simple linear iterative clustering (SLIC) algorithm, and discriminate these objects into changed and unchanged classes using fuzzy c-means (FCM) clustering and a deep PCANet. The prediction of this Phase is the set of changed and unchanged superpixels. (2) Deep learning on the pixel sets over the changed superpixels only, obtained in the first phase, to discriminate real changes from false changes. SLIC is employed again to achieve new superpixels in the second phase. Low rank and sparse decomposition are applied to these new superpixels to suppress speckle noise significantly. A further clustering step is applied to these new superpixels via FCM. A new PCANet is then trained to classify two kinds of changed superpixels to achieve the final change maps. Numerical experiments demonstrate that, compared with benchmark methods, the proposed approach can distinguish real changes from false changes effectively with significantly reduced false alarm rates, and achieve up to 99.71% change detection accuracy using multi-temporal SAR imagery.

In this article, a new type of feature, named two-dimensional (2D)-slice Zernike moments, is proposed for synthetic aperture radar (SAR) automatic target recognition (ATR). Target features play an extremely important role in the ATR system. Pixels with different scattering intensities distribute in different positions in SAR images, which represent target inherent signatures determined by the target’s characteristics, including global structure and local details. To extract these various scattering signatures, we developed a feature extraction technique named 2D-slice Zernike moments (2DS-ZMS), which can capture target global and local scattering field distribution information. First, the multilayer 2D-slices of a SAR image are extracted by uniformly cutting the 3D display SAR image along the amplitude direction. Then Zernike moments of each 2D-slice are calculated. Finally, the 2DS-ZMS of the SAR image are formulated into a column vector, called the feature vector. The obtained feature vectors of the targets are fed into a newly developed classifier, i.e. the sparse representation-based classifier (SRC). By projecting the testing sample feature vector on the dictionary made up of training samples feature vectors, the sparse coefficients are solved. The minimum reconstruction residual is adopted as the judgement criterion for predicting the test sample’s class label. Experiments on the moving and stationary target acquisition and recognition (MSTAR) data set validate the effectiveness and superiority of the proposed method.

Automatic target recognition (ATR) is a significant application scenario for Synthetic aperture radar (SAR) image interpretation. In recent years, deep-learning based SAR ATR approaches have made great progress, with the require of large amounts of labeled data for network training. However, the labeled SAR images are scarce, and the annotation of SAR images is expensive and time-consuming. In this paper, a novel semi-supervised learning (SSL) framework is proposed for SAR ATR, which effectively alleviates the need of labeled samples for network training, and allows for excavating the intrinsic semantic relationship information between samples belonging to different categories, greatly improving recognition performance. This method develops an epoch- and uncertainty-aware pseudo-label selection (EUAPS) mechanism, which takes advantage of the underutilized consistency between training epochs, and introduces the uncertainty estimates. EUAPS can select pseudo-label samples with high-confidence, allowing the network to perform well even when labels are extremely scarce. Furthermore, we propose a novel loss function, group relationship consistency loss (GRCL) that explicitly enforces the consistency of relations between different samples under perturbations. GRCL facilitates the network to learn rich sample relationship information and more discriminative features in sample group level, effectively reducing the recognition error rate. Extensive experiments were carried out on three public benchmark SAR datasets. And the experimental results illustrate that the proposed approach achieves the accuracy of 98.58% when 10 labeled samples per class and 88.65% when only 5 labeled samples per class respectively, outperforming the state-of-the-art SSL methods and demonstrating the superiority.

In this letter, we propose a synthetic aperture radar (SAR) target configuration recognition algorithm based on the fusion of multifeature low-rank representations (LRRs). First, Gabor, principal component analysis, and wavelet features are extracted for the SAR training set and test set, respectively. Second, with the LRR model, each feature of the test samples is represented by those of the training set, leading to the corresponding coefficient matrix. Then, the preliminary prediction labels of all features of the test sample are obtained according to the LRR coefficients. Third, in order to further improve the confidence of recognition and reduce the instability of the algorithm, a two-stage decision fusion strategy is adopted to obtain the final prediction labels. The first stage utilizes a vote fusion for the recognition results of multiaspect neighborhood test samples for each feature pattern, which exploits the strong correlation of these neighborhood samples. Furthermore, the second stage fuses the three results obtained in the first stage through Bayesian inference. Bayesian inference is widely used in decision fusion, which can improve the confidence of results by about 3%. Experiments on the moving and stationary target acquisition and recognition data set demonstrate the effectiveness and superiority of the proposed algorithm.

Polarimetric synthetic aperture radar (PolSAR) image classification plays an important role in various PolSAR image application. And many pixel-wise, region-based classification methods have been proposed for PolSAR images. However, most of the pixel-wise methods can not model local spatial relationship of pixels due to negative effects of speckle noise, and most of the region-based methods fail to figure out the regions with the similar polarimetric features. Considering that color features can provide good visual expression and perform well for image interpretation, in this work, based on the PolSAR pseudo-color image over Pauli decomposition, we propose a supervised PolSAR image classification approach combining learned superpixels and quaternion convolutional neural network (QCNN). First, the PolSAR RGB pseudo-color image is formed under Pauli decomposition. Second, we train QCNN with quaternion PolSAR data converted by RGB channels to extract deep color features and obtain pixel-wise classification map. QCNN treats color channels as a quaternion matrix excavating the relationship among the color channels effectively and avoiding information loss. Third, pixel affinity network (PAN) is utilized to generate the learned superpixels of PolSAR pseudo-color image. The learned superpixels allow the local information exploitation available in the presence of speckle noise. Finally, we fuse the pixel-wise classification result and superpixels to acquire the ultimate pixel-wise PolSAR image classification map. Experiments on three real PolSAR data sets show that the proposed approach can obtain 96.56%, 95.59%, and 92.55% accuracy for Flevoland, San Francisco and Oberpfaffenhofen data set, respectively. And compared with state-of-the-art PolSAR image classification methods, the proposed algorithm can obtained competitive classification results.

In this paper, we propose a two-stage multi-task learning representation method for the classification of synthetic aperture radar (SAR) target images. The first stage of the proposed approach uses multi-features joint sparse representation learning, modeled as a ${\ell }_{2,1}$ -norm regularized multi-task sparse learning problem, to find an effective subset of training samples. Then, a new dictionary is constructed based on the training subset. The second stage of the method is to perform target images classification based on the new dictionary, utilizing multi-task collaborative representation. The proposed algorithm not only exploits the discrimination ability of multiple features but also greatly reduces the interference of atoms that are irrelevant to the test sample, thus effectively improving classification performance. Conducted with the Moving and Stationary Target Acquisition and Recognition (MSTAR) public SAR database, experimental results show that the proposed approach is effective and superior to many state-of-the-art methods.