The undesirable distortions of synthetic aperture radar (SAR) images pose a challenge to intuitive SAR interpretation. SAR-to-optical (S2O) image translation provides a feasible solution for easier interpretation of SAR and supports multisensor analysis. Currently, diffusion-based S2O models are emerging and have achieved remarkable performance in terms of perceptual metrics and fidelity. However, the numerous iterative sampling steps and slow inference speed of these diffusion models (DMs) limit their potential for practical applications. In this letter, an efficient end-to-end diffusion model (E3Diff) is developed for real-time one-step S2O translation. E3Diff not only samples as fast as generative adversarial network (GAN) models, but also retains the powerful image synthesis performance of DMs to achieve high-quality S2O translation in an end-to-end manner. To be specific, SAR spatial priors are first incorporated to provide enriched conditional clues and achieve more precise control from the feature level to synthesize optical images. Then, E3Diff is accelerated by a hybrid refinement loss, which effectively integrates the advantages of both GAN and diffusion components to achieve efficient one-step sampling. Experiments show that E3Diff achieves real-time inference speed (0.17 s per image on an A6000 GPU) and demonstrates significant image-quality improvements (35% and 27% improvement in Frechet inception distance (FID) on the UNICORN and SEN12 dataset, respectively) compared to existing state-of-the-art (SOTA) diffusion S2O methods. This advancement of E3Diff highlights its potential to enhance SAR interpretation and cross-modal applications. The code is available athttps://github.com/DeepSARRS/E3Diff.

In recent years, contrastive learning (CL) methods have achieved remarkable success in few-shot classification of polarimetric synthetic aperture radar (PolSAR) images. However, existing CL models based on Euclidean metric typically vectorize PolSAR data into real-valued or complex-valued vectors with independent channels, disrupting the inherent correlations between polarimetric channels. To address this issue, we propose a novel CL framework, the Riemannian-Euclidean CL network (RECLNet). The proposed RECLNet allows direct input of polarimetric covariance matrices, overcoming the limitations of conventional Euclidean-based CL models that require vectorizing PolSAR data. First, RECLNet constructs hard positive samples by leveraging polarimetric information within the Riemannian space. Second, a series of carefully designed Riemannian manifold operation (RMO) layers are used to extract the Riemannian geometric features of PolSAR data while preserving its matrix structure. Finally, the vision transformer (ViT) is adopted as the backbone of the Euclidean metric encoding to capture contextual information. Experiments conducted on two widely used PolSAR datasets demonstrate that the proposed approach achieves superior performance compared with existing state-of-the-art methods.

Effectively fusing multifrequency polarimetric synthetic aperture radar (PolSAR) data has emerged as a key research focus, as it holds significant potential for optimizing the interpretation of diverse scattering mechanisms in complex environments. However, current fusion technologies primarily rely on data-driven machine learning or statistical models. While these methods effectively achieve multifrequency data fusion, they often lack clear physical interpretation, making it challenging to directly explain or quantify the contributions of different frequencies. To address these challenges, a general fusion method is proposed in this letter, which introduces five canonical scattering models to guide the fusion of multifrequency data. This method fully considers the scattering diversity within individual pixels while ensuring the physical significance of the fusion process. Specifically, we systematically investigate various combinations of multifrequency data and apply the sequential quadratic programming (SQP) algorithm to optimize the similarity between the fused data and the canonical scattering models. The combination that maximizes this similarity is identified as the optimal multifrequency data fusion configuration. The RADARSAT-2 C-band full-polarized data and the ALOS-2 L-band full-polarized data acquired over the San Francisco area are applied to validate the proposed method. The experimental results demonstrate that the fused image offers superior performance in interpreting ground features compared to the original images.

Scattering Power Decomposition of polarimetric synthetic aperture radar (PolSAR) coherency matrix is essential for PolSAR image interpretation. However, current research faces two challenges: (1) the contradiction between limited scattering models (i.e. covariance matrix models or coherency matrix models) and the complex, diverse scattering types in reality, leading to incomplete and ambiguous characterization; (2) too many manually configured branches in the solving process due to models having more unknown parameters than observations. To address the above issues, a scattering power decomposition method based on Polarimetric Interferometric Synthetic Aperture Radar (PolInSAR) image and sparse representation theory is proposed. Specifically, to eliminate scattering ambiguity in polarization data, PolInSAR coherence and self-organizing maps (SOM) are introduced to develop unambiguous decomposition schemes. Besides, to enhance the diversity of scattering mechanisms and models, the rotated double-bounce and coherent volume scattering mechanisms for different types of buildings are added; the scattering power decomposition is implemented through sparse representation with an overcomplete dictionary of diverse scattering models. Experiments on three pairs of PolInSAR data validate the effectiveness of the proposed method. This work reveals the essence of the scattering characterization system, offers effective approaches to address its complexities without extensive scattering modeling, and has valuable applications in target detection and fine land cover classification.

In recent years, continuous improvements in SAR resolution have significantly benefited applications such as urban monitoring and target detection. However, these improvements in resolution have also led to increased discrepancies in scattering characteristics, posing challenges to the generalization ability of target detection models. While domain adaptation technologies provide a potential solution, the inevitable discrepancies caused by resolution differences often result in blind feature adaptation and unreliable semantic propagation, ultimately degrading the domain adaptation performance. To address these challenges, this paper proposes a novel SAR target detection method, termedCR-Net, which incorporates structure priors and evidential learning theory into the detection model, enabling reliable domain adaptation for cross-resolution detection. To be specific,CR-Netintegrates Structure-induced Hierarchical Feature Adaptation (SHFA) and Reliable Structural Adjacency Alignment (RSAA). TheSHFAmodule is designed to establish structural correlations between targets and achieve structure-aware feature adaptation, thereby enhancing the interpretability of the adaptation process. Afterwards, theRSAAmodule is proposed to enhance reliable semantic alignment, by leveraging the secure adjacency set to transfer valuable discriminative knowledge from the source domain to the target domain. This further improves the discriminability of the detection model in the target domain. Based on experimental results from different-resolution datasets, the proposedCR-Netsignificantly enhances cross-resolution adaptation by preserving intra-domain structures and improving discriminability. It achieves state-of-the-art (SOTA) performance in cross-resolution SAR target detection.

The attribute scattering center (ASC) is a common target characteristic in synthetic aperture radar (SAR) image. In addition to the optical features, the physical parameters extracted by ASC provide the specific target structure information related to the SAR system. However, the existing methods for ASC extraction usually extract all parameters at the same time, which cause many errors in the extracted results. In this paper, a hierarchical ASC extraction method is proposed, in which the parameters of each scattering center are extracted hierarchically. In the experimental section, a SAR image simulation algorithm is used to simulate SAR images of some typical structures and the effect of ASC extraction is analyzed. Besides, a set of measured data is used for validation as well. Experimental results show that the hierarchical ASC extraction method can extract the scattering centers more accurately.

In this paper, a high-resolution spotlight SAR signal simulation method using electromagnetic field distribution of the target calculated by FDTD is proposed. Aiming at the difficulty of introduce Doppler history into the existing SAR echo simulation methods, the extrapolation boundary setting and near-to-far transformation method is improved, then the received electric field of antenna in all azimuth can be calculated by compensating Doppler history to simulation data in single azimuth. In addition, a sub-aperture simulation method is proposed to solve the problem that the error of waveform forming method increases in the simulation of spotlight SAR echo. Numerical experiments are made and results show the accuracy of FDTD method and the feasibility of proposed method in spotlight SAR signal simulation.

Aircraft detection using Synthetic Aperture Radar (SAR) images plays a crucial role in transportation and military applications. Nevertheless, the unique imaging characteristics of SAR often render aircraft targets as discrete points, and their complex geometric structures vary under different imaging conditions. Moreover, the complex background, especially with strong-scattering elements like buildings, significantly complicates detection. To overcome these challenges, this paper introduces a multi-aspect feature enhancement network called MAFEN for aircraft detection in high-resolution SAR images. MAFEN combines a Multi-scale Feature Enhancement Module (MSFEM) with a Key Structure Enhancement Module (KSEM), thereby enhancing detection accuracy and efficiency. Experimental results on public datasets demonstrate significant improvements in detecting aircraft targets in complex scenes.

Observing that the integration of polarimetric features with physical attributes in Polarimetric Synthetic Aperture Radar (PolSAR) images yields superior decoupling compared to individual statistical features. In response to the challenge, the Attention-based Feature Selection and Edge Preservation CNN (AFE-CNN) is proposed, which integrates a channel attention mechanism to dynamically assign weights to input features and employs a multi-scale spatial attention mechanism to prioritize edge information. Specifically addressing edge confusion in Polarimetric Synthetic Aperture Radar (PolSAR) image classification, this approach ensures the preservation of crucial edge details through judicious selection and utilization of input features. The effectiveness of AFE-CNN is validated through end-to-end classification of PolSAR images on two widely utilized datasets.

Large vertical cylindrical structure is an important type of structure in polarimetric synthetic aperture radar (PolSAR) images. Objects with this structure, such as oil tanks, granaries and rotor sails, attract much attention in multiple fields. Before detecting these targets with these structures, scattering mechanism analysis and feature extraction are necessary prerequisite steps. However, due to the lack of scattering mechanism analysis and characterization for such structures, traditional feature extraction methods, like polarimetric decomposition methods, usually fail to distinguish these targets. In this letter, according to high frequency approximation theory, the scattering mechanism of large vertical cylindrical structure with finite radius and height is analyzed. The coherency matrix of this structure with multiple pixels is derived further for scattering characterization. The model is called vertical cylindrical scattering model (VCDM), which is further introduced into a polarimetric decomposition method for feature extraction. Experiments prove that the model can describe the scattering mechanism of large vertical cylindrical structure accurately. Besides, the feature extraction method can effectively extract features for large vertical cylindrical structures in PolSAR images, which helps detect targets with this kind of structures.

Significant progress has been achieved by using the graph convolutional network (GCN) in image change detection. However, the limited quantity of labeled data and the inherent speckle noise adversely impact the generalization ability of the existing GCN-based methods in practical synthetic aperture radar (SAR) image applications. To address these challenges, we introduce the structure-optimized complex-valued graph contrastive learning network (SCGCLN) for semi-supervised SAR image change detection. Specifically, we explore how to learn effective feature representations from complex-valued SAR data with limited supervised information using the GCN architecture. We present a structure-optimized graph reconstruction strategy based on optimizing node features and edge structures. By combining efficient spectral clustering with graph reconnection, our method learns high-quality graph structures that enable the network to capture long-range dependencies, thereby mitigating the impact of speckle noise. Moreover, we construct a complex-valued graph contrastive learning (GCL) network to train a graph feature representation model from unlabeled SAR data. Subsequently, the pretrained model is fine-tuned for the downstream limited labeled SAR change detection task. The effectiveness of SCGCLN is validated through experimental results on three SAR image datasets.

Polarimetric synthetic aperture radar (PolSAR) image classification is a significant research area in the interpretation of PolSAR images. In recent years, PolSAR classification algorithms based on deep learning (DL) have been widely used, and achieved excellent performance in the case of sufficient training samples. However, due to the particularity of radar imaging, the scarcity of annotated data poses a challenge for PolSAR image classification based on DL. Contrastive learning (CL) is an effective approach to solve the problem of insufficient training samples. However, current CL methods have overlooked the utilization of the rich polarimetric information which is unique to PolSAR data. Thus, we propose a novel CL model to address the aforementioned issue. Firstly, inspired by the polarimetric scattering mechanism, the proposed method constructs hard positive samples combining polarimetric information. Then, a mutual learning regularizer (MLR) is used to design a loss function that bridges the semantic gap between different positive instances while introducing the hard positive samples. Finally, a transductive CL framework and gradient stop method are employed for pre-training, preventing model collapse and excessive computing requirements. The proposed method has been tested on two benchmark PolSAR datasets. On the same network backbone, the proposed method achieves up to 4.07% and 5.43% improvement in overall accuracy (OA) and kappa coefficient (KC), and up to 1.08% and 1.43% improvement compared to the the state-of-the-art methods. The experimental results demonstrate impressive performance on few-shot classification task of PolSAR image classification.

In recent years, self-supervised synthetic aperture radar (SAR) image change detection methods have achieved remarkable results, particularly in reducing dependence on expensive supervised signals. However, some critical issues remain open: how to design the optimal unsupervised feature representation model, and how to utilize noisy pseudo-labels for self-training to improve the change detection model? In this article, efforts are made to find a principled and fundamental solution to the above issues from the new perspective of causal reasoning, proposing a self-supervised SAR change detection framework named CausalCD. Specifically, we first construct a structural causal model (SCM) to formalize the self-supervised SAR image change detection process, and carry out a principled analysis to assess the influence of the unsupervised feature representation module and the pseudo-label-driven self-training module on the performance of change detection model. On this basis, we present a feature representation module that employs graph contrastive learning (GCL) and leverages the causal invariant mechanism to extract the optimal augmented representation, thus improving the model’s generalizability and soundness. In addition, we employ a causal intervention strategy to mitigate the negative impact of confusing bias from noisy pseudo-labels by blocking the backdoor path. CausalCD’s strengths lie in reducing the model’s dependence on labeled samples through causal GCL and obtaining optimal feature representation. Concurrently, it helps disentangle the noisy pseudo-label’s impact, thereby improving the performance of the self-supervised SAR change detection model. Finally, comprehensive experiments on three bitemporal SAR scenes demonstrate that CausalCD significantly outperforms several mainstream change detection models, confirming the effectiveness of CausalCD.

The presence of speckles and geometric distortions poses a serious challenge to the visual interpretation of synthetic aperture radar (SAR) images. SAR-to-optical (S2O) image translation technology provides a feasible solution and has attracted increasing attention. Restricted by substantial gaps between optical and SAR images, current S2O translation methods unavoidably result in geometric distortions, target missing, and generating low-fidelity images, thereby limiting subsequent cross-modal applications. In this article, we propose an augmented conditional denoising diffusion probabilistic model with spatial-frequency refinement (SFDiff) for high-fidelity S2O image translation. SFDiff progressively narrows the gap between synthesized and real images in both spatial and frequency perspectives, showcasing notable performance in terms of quality and consistency. Specifically, to incorporate rich spatial content priors provided by SAR images, we design an SAR context prior extractor (SCPE) with denoising enhancement to extract multiscale conditional representations, thereby aiding SFDiff in capturing more descriptive cues for S2O translation. In addition, a spatial-frequency complementary learning (SFCL) module is designed to learn spatial semantics and simultaneously enhances informative frequency components and global dependencies. Furthermore, SFDiff is optimized using the joint spatial-frequency refinement loss, facilitating iterative refinement in both spatial and frequency domains to enhance content consistency and fidelity in the synthesized images. Based on the experimental findings from the UNICORN dataset and the SEN12 dataset, SFDiff maintains a high level of content and structural consistency, resulting in visually appealing translation results that surpass the state-of-the-art (SOTA) methods. In particular, SFDiff exhibits excellent performance in preserving small targets and details, which is crucial in cross-modal detection applications.

Benefiting from the advantages of height sensitivity, the short time span, and the low data cost, interferometric synthetic aperture radar (InSAR) technology has the potential for 3-D reconstruction. However, the layover problem has always been the limitation for the InSAR-based height reconstruction of buildings in complex scenes because of the mixing of scatterings from building facade, roof, and other interferers. In this article, the above layover problem was addressed, and the height reconstruction of buildings in complex scenes was first achieved based on one single interferogram. Specifically, the layover mechanism in complex scenes in the InSAR system was fundamentally explored and was summarized as a general interferometry model. Based on the established model, a height reconstruction method for high-rise buildings was proposed. The main idea is to recognize the facade scattering-dominated area and then reconstruct the complete facade interferometric phase based on the interferometry model and derived facade phase gradient characteristic. Finally, the building height can be obtained based on the phase-height conversation. TerraSAR-X InSAR data in six different complex scenes were used for experiments. For results based on the proposed method, the mean absolute error and root-mean-square error are less than 1.2 m, and compared with the building height reconstruction based on the original layover phase, the accuracy is improved by several meters to tens of meters, indicating that the height reconstruction of buildings in complex scenes can be achieved with high accuracy based on the proposed method.

Polarimetric synthetic aperture radar (PolSAR) target decomposition is an effective way to obtain the scattering mechanism information. However, the problem of volume scattering overestimation in rotated built-up areas has not been completely addressed even with extensive efforts in scattering modeling. The root lies in the scattering ambiguity in polarization response (S matrix, T matrix, and C matrix) which is difficult to remove. To handle this problem, the repeat-pass polarimetric interferometric synthetic aperture radar (PolInSAR) coherence is introduced to PolSAR target decomposition in this article, to distinguish the volume scattering and the scattering generated by rotated dihedral structures and then achieve the volume scattering component correction. Specifically, a building descriptor$P_{bd}$is established based on PolInSAR coherence and then used to distinguish natural and built-up areas before target decomposition. Then, two different scattering model sets, where the volume scattering and the rotated double-bounce scattering do not appear at the same time, are applied in natural and built-up areas, respectively. Applying the above two points to classic decomposition methods, a series of improved methods are proposed, named repeat-pass PolInSAR coherence-assisted target decomposition methods. Experiments on three sets of PolInSAR data confirm the validity of the proposed decomposition methods. Besides, time series PolInSAR data are used to analyze the performance of the algorithm under different temporal baseline conditions. This work may enlighten how to efficiently correct the volume scattering component in rotated built-up areas by introducing repeat-pass PolInSAR coherence into PolSAR target decomposition, to reduce the large amount of energy spent on the complex scattering modeling.

Owing to various factors, including severe speckle noise and orbit direction differences, performing multitemporal synthetic aperture radar (SAR) image registration with high accuracy and robustness may become difficult. Herein, an efficient self-supervised deep learning registration network for multitemporal SAR image registration, SAR-superpoint and transformation aggregation network (SSTA-Net), is proposed. The SSTA-Net consists of three parts: 1) the SAR-Superpoint detection network (SS-Net); 2) the transformation aggregation feature matching network (TA-Net); and 3) the unstable point removal module. Specifically, a pseudolabel generation method is adopted without additional annotations. It transfers the characteristics of real SAR data to synthetic data through a feature transition module, which can generate feature point labels for real SAR images for self-training SS-Net. Furthermore, a position–channel aggregation attention is proposed and embedded into the SS-Net to efficiently capture position and channel information and to increase the stability and accuracy of feature point identification. Finally, a unique transformation aggregation strategy is designed to improve the robustness of feature matching, and an unstable point removal module is adopted to eliminate the mismatched point pairs caused by orbit differences. Six sets of multitemporal SAR images were used to evaluate the registration performance of the SSTA-Net, and our model was also compared with the traditional and deep learning algorithms. The experimental results demonstrate that the SSTA-Net outperforms various state-of-the-art approaches for SAR image registration.

Ship detection in SAR images is a challenging problem. CNN-based ship detection method in SAR images has achieved remarkable results. Due to the multi scale of the ships and interference from complex sea conditions or nearshore background in SAR images, many false alarms and missed detections can occur in ship detection. To solve these problems, a multi-scale ship detection network in SAR images based on attention and weighted fusion is proposed in this paper. First, a higher-resolution detect head is added based on the YOLOv5 framework for detecting tiny-scale ships in SAR images. Then, the coordinate attention block is introduced to refine the location features of ship targets and suppress the interference of complex background. Finally, in the feature fusion stage, adaptive weighted feature fusion is used to reduce feature redundancy. Experiments on the SSDD dataset show the effectiveness of the proposed method.

Noting the fact that the high-dimensional data composed of various polarimetric features has better decouplability than polarimetric synthetic aperture radar (PolSAR) image source data, in this letter, multiple polarimetric features are extracted and stacked to form a high-dimensional feature cube as the input of convolutional neural networks (CNNs) to improve the performance of PolSAR image classification. Directly utilizing the polarimetric features will produce a performance degradation and the recalibration of them is indispensable. However, classical feature selection methods are independent of the classifier, which means that the stimulated features may not be the classification-friendly ones. To avoid separated procedures and improve the performance, attention-based polarimetric feature selection convolutional network, called AFS-CNN, is proposed to implement end-to-end feature selection and classification. The relationship between input polarimetric features can be captured and embedded through attention-based architecture to ensure the validity of high-dimensional data classification. Experiments on two PolSAR benchmark data sets verify the performance of the proposed method. Furthermore, this work is quite flexible, which is reflected in that the proposal can be used as a plug-and-play component of any CNN-based PolSAR classifier.

Man-made target detection of PolSAR image is an important part of the application of remote sensing data, and generally, the target occupies very few pixels of the image. Convolutional sparse representation (CSR) is proved able to effectively combine polarimetric and spatial information of a PolSAR image to achieve target detection. However, the acquisition of prior knowledge and the modeling of background are difficult. Moreover, the solution process is highly dependent on the alternating direction method of multipliers (ADMMs) algorithm, which introduces new parameters and has high computational complexity. To solve these problems, this letter proposes a man-made target detection method for PolSAR images based on local CSR (Local-CSR). In this method, a novel target detection framework to combine the polarimetric and spatial information based on CSR is proposed, the target dictionary is constructed by the Local-CSR to avoid the aforementioned problems of ADMM solution procedure and finally realize the man-made target detection of the PolSAR image. Two sets of fully polarimetric SAR data sets are used to demonstrate the performance and the experimental results prove the capacity and validity of the proposed method.

The canopy vertical structure, especially the average height, has always been regarded as an important factor in monitoring forest changes. The coherent scattering model links forest canopy information with radar observations and the random volume over ground (RVoG) model has been extensively applied to the polarimetric SAR interferometry (PolInSAR) data since it was proposed. The complex coherence of the RVoG model was originally derived in a simplified way by neglecting some factors due to the complexity of the scattering process. Thence, this letter proposed a slope three-layer scattering (STLS) model for forest parameters’ estimation in sloping mountain forest region. This model separates the vertical structure of the forest into three layers: the ground layer, the tree-trunk layer, and the canopy layer which account for the simultaneous effects of three scattering components on complex coherence. Moreover, it also corrects the distortion caused by the local terrain slope. The STLS model provides a better understanding of the microwave scattering process in the terrain slope area compared with the traditional RVoG model, S-RVoG model, and general three-layer scattering model (GTLSM) model. Finally, the STLS model has been quantitatively tested with the simulated PolInSAR data with different terrain slopes from PolSARProSim software and qualitatively tested with the spaceborne SIR-C data in Tian-Shan Mount area. The results validate the potential of the proposed STLS model in forest parameter inversion.

Change detection is a key technology in the field of polarimetric synthetic aperture radar (PolSAR) image processing. The current research on the change detection mainly focuses on studying PolSAR images with the same angle or small angle difference, and the angle problem is not considered. However, when the angle difference occurs, especially a large angle difference, some pixels might be falsely detected because the angle difference can affect the polarimetric characteristics. In this letter, we propose a multilevel information fusion-based (MIFB) method, which is suitable for extracting change information from PolSAR images with angle difference. In particular, the proposed method first adopts data resolution correction, then applies an improved feature-based registration algorithm, and finally, incorporates weighted graph theory with the superpixel segmentation algorithm to extract and merge pixel-based and object-based change areas to eliminate false alarms. Experimental results for multitemporal and multiangle PolSAR images reveal that the MIFB method can effectively eliminate false detection caused by angle differences and improve the detection accuracy.

Ship detection using polarimetric synthetic aperture radar (PolSAR) images is a major area of interest within the field of target detection. It is because the polarimetric information in PolSAR images is beneficial to extract features corresponding to different ship structures. Most of the work carried out on target detection is by extracting target features combined with detectors such as constant false alarm rate detector (CFAR) and support vector machine (SVM). However, much of the research up to now has been undertaken by adjusting parameters manually which depends on experience to a large extent resulting in unstable consequence and unintelligent detection. In order to achieve the goal of automatic detection, a novel ship detection method based on simulated annealing by fuzzy matching (SAFM) methods is proposed in this letter, noted SAFM for convenience, which can achieve the automatic adjustment of parameters. In addition, this letter applies SAFM to adjust the number of features and parameters of the detector simultaneously through which self-adaptive feature screening and adjustment of parameters in the detector are realized and data characteristics are considered. The experimental results suggested that the method proposed in this letter can achieve automatic detection and get good results on ships, especially for docked ships.

Small-scale target detection (such as vehicles) in complex synthetic aperture radar (SAR) image scenes has always been a pain point for the advanced convolutional neural network (CNN)-based target detectors because of the downsampling operations and the local receptive field characteristics of CNNs. To tackle these limitations, a vehicle detector named SCEDet for the small-scale vehicles in SAR images is proposed to improve the detection performance in this letter. SCEDet mainly consists of two parts: subaperture semantic feature extraction and subaperture semantic-context enhancement (SCE) with SCE module. First, ResNet34 with subaperture decomposition is used to efficiently exploit the latent subaperture semantic features. Then, the SCE module is proposed to balance the multiscale semantic information as well as aggregate the global context information for vehicle detection with a small number of parameters and computation costs. The experimental results on the FARAD dataset (0.1 m$\times0.1$m, Ka-band) demonstrate that both the detection performance and the speed are much better than other detection methods under the same hardware conditions.

As one of the most popular topics in polarimetric synthetic aperture radar (PolSAR) community, PolSAR image classification has always been an important way for PolSAR applications. Constructing representations is the most critical part of PolSAR image classification. With the maturity of deep learning technique, many data-driven PolSAR representation methods have been proposed, most of which are based on convolutional neural networks (CNNs). Despite some achievements, the bottleneck of CNN-based methods may be related to the locality induced by their inductive biases. Considering this problem, the state-of-the-art method in natural language processing, i.e., transformer, is introduced into PolSAR image classification for the first time. Specifically, a vision transformer (ViT)-based representation learning framework is proposed in this article, which covers both supervised learning and unsupervised learning. For supervised learning, we use self-attention to replace convolution, which shifts the focus from the information in local neighborhoods to the long-range interactions between each pixel. Beyond supervised learning, we introduce an improved contrastive-based strategy to implement simple unsupervised representation learning. Compared with CNN and its variants, ViT constructs more global representations by explicitly modeling the relationship between each pixel, so as to improve the classification performance. Experimental results on four widely used PolSAR image datasets indicate that the representation obtained by the ViT-based methods is better for PolSAR image classification, whether supervised (up to about 5% accuracy improvement) or unsupervised (up to about 4%). In addition, we also prove the robustness of ViT to the initial input form. These discoveries may arouse rethinking of the dominance of CNNs in PolSAR image classification.

The generality and robustness of a man-made target detection method are of high practical value in SAR applications. However, current methods hardly adapt to different SAR images simultaneously because data property and target characteristics are not identical in images with various resolutions and scenes. The key to achieving high generality and robustness is extracting stable and invariant information in different SAR images. This paper analyzes the scattering of man-made targets and natural backgrounds in SAR images and assumes that the noise characteristics are relatively stable and invariant when resolution and observing background change. Then the Relative Spectral Stationarity (RSS) is proposed based on two-dimensional spectrum analysis to measure the distance between the observed data and a manually generated noise. RSS takes the manually generated noise with known and definite properties as a standard, so it is irrelevant to image parameters and scenes. A low RSS indicates that the characteristics of observing data are dominated by noise, whereas a high value means there may be targets in the observing data weakening the noise characteristics. An efficient segmentation algorithm Sparsity-based Format-free Segmentation within errore(SFFe) is proposed to process the RSS map and complete the detection method. SAR images with resolutions ranging from 0.1m to 8m are employed as testing data. Various testing scenes are constructed to simulate different practical conditions. Experimental results validate that the proposed RSS-based method works well in SAR images with different resolutions, bands, and observing scenes, obtaining reliable and robust detection results and outperforming canonical methods on various criteria.

Deep learning and convolutional neural networks (CNNs) have made progress in polarimetric synthetic aperture radar (PolSAR) image classification over the past few years. However, a crucial issue has not been addressed, i.e., the requirement of CNNs for abundant labeled samples versus the insufficient human annotations of PolSAR images. It is well known that following the supervised learning paradigm may lead to the overfitting of training data, and the lack of supervision information of PolSAR images undoubtedly aggravates this problem, which greatly affects the generalization performance of CNN-based classifiers in large-scale applications. To handle this problem, in this article, learning transferrable representations from unlabeled PolSAR data through convolutional architectures is explored for the first time. Specifically, a PolSAR-tailored contrastive learning network (PCLNet) is proposed for unsupervised deep PolSAR representation learning and few-shot classification. Different from the utilization of optical processing methods, a diversity stimulation mechanism is constructed to narrow the application gap between optics and PolSAR. Beyond the conventional supervised methods, PCLNet develops an unsupervised pretraining phase based on the proxy objective of instance discrimination to learn useful representations from unlabeled PolSAR data. The acquired representations are transferred to the downstream task, i.e., few-shot PolSAR classification. Experiments on two widely used PolSAR benchmark data sets confirm the validity of PCLNet. Besides, this work may enlighten how to efficiently utilize the massive unlabeled PolSAR data to alleviate the greedy demands of CNN-based methods for human annotations.

Image segmentation has been an important procedure in object-based image analysis (OBIA), which takes image object as a processing unit. The spatial scale in image segmentation has great importance in OBIA. Due to the high heterogeneity and large dynamic range of polarimetric synthetic aperture radar (PolSAR) images, it is often difficult to choose optimal spatial scales. This letter proposes a polarimetric semivariogram-based spatial scale selection method for PolSAR image segmentation. The optimal spatial bandwidth parameter in the mean-shift algorithm is pre-estimated based on the combined polarimetric and statistical analysis of PolSAR images. By implementing a quantitative evaluation of segmentation result, the effectiveness of the proposed method in optimal spatial bandwidth selection for PolSAR image segmentation is verified. Experiments on both the EMISAR and UAVSAR L-band PolSAR data sets testify the validity of the proposed adaptive optimal bandwidth selection strategy for PolSAR images.

Built-up area extraction is a primary and fundamental processing in applications such as city planning using remotely sensed high-resolution synthetic aperture radar (SAR) images. One of the critical challenges is that canopy-covered area is always falsely extracted due to similarity between canopy and building in scattering power and texture pattern, resulting in high false alarm and low overall accuracy. In this letter, physical scatterings on built-up areas and canopy-covered areas are analyzed, seeking the distinct differences between these two ground types under various observing scales. A spectral reconfiguration (SR) descriptor is proposed in frequency domain to describe differences that can be strengthened by frequency modulating strategy. Both theoretical and experimental analyses show that the SR descriptor can separate buildings and canopy greatly. Meanwhile, it enjoys both slight computational burden and low operative complexity. Based on this descriptor, an SR-intensity-based built-up extraction algorithm is proposed. Experimental results validate that the SR-intensity-based algorithm acquires low false alarm rate with high accuracy, showing great practical meaning and potential of improvement for the application of urban remote sensing.

Recently, an object-based Markov random field (OMRF) with auxiliary fields (OMRF-A) was developed for the processing of optical remote sensing images and it provided satisfactory results. However, it cannot be directly applied to polarimetric synthetic aperture radar (PolSAR) images with high heterogeneity. In addition, polarimetric information which plays a dominant role in PolSAR image interpretation cannot be effectively incorporated by the OMRF-A model. In order to solve this problem, an OMRF with polarimetric auxiliary fields (OMRF-PA) is developed in this letter for the classification of PolSAR images. A polarimetric index is developed to evaluate the information loss during iterations of label fields and auxiliary label fields. Then, an improved conditional probability distribution which incorporates the polarimetric index is proposed to account for the interactions between label fields and auxiliary label fields. Experiments on PolSAR data sets acquired by ESAR and EMISAR systems demonstrate that the proposed OMRF-PA model can generate a higher classification accuracy compared to the original OMRF-A method.

Target detection is of great significance for polarimetric SAR (PolSAR) image applications, and independent target detection in a large scene can be considered as a sparse problem. In high-resolution PolSAR, the independent targets generally present a cluster of similar pixels, and therefore, the detection principle should include not only the internal characteristics but also the spatial information. This letter proposes an unsupervised adaptive convolution sparse representation (ACSR) method for PolSAR image independent target detection. The proposed method updates the dictionary to realize adaptive spatial information injection into polarimetric features, and then the independent target can be detected through the iteration. Two sets of high-resolution, full PolSAR images of unmanned aerial vehicle synthetic aperture radar (UAVSAR) system are used to validate the performance of the proposed method. The results indicate the potential of the proposed method in target detection of PolSAR image.

Convolutional neural networks (CNNs) have shown good performance in polarimetric synthetic aperture radar (PolSAR) image classification. Excellent hand-crafted CNN architectures incorporated the wisdom of human experts, which is an important reason for CNNs success. However, the design of architectures is a difficult problem, which needs a lot of professional knowledge as well as computational resources. Moreover, the manually designed architecture might be suboptimal, because it is only one of the thousands of unobserved but objective existed paths. Considering that the success of deep learning is largely due to its automation of the feature engineering process, how to design automatic architecture search methods to replace the hand-crafted ones is an interesting topic. In this article, the application of neural architecture search (NAS) in the PolSAR area is explored for the first time. Different from the utilization of existing methods, a PolSAR-tailored Differentiable Architecture Search (DARTS) method, called PDAS, is proposed in order to adapt NAS to the PolSAR classification. A PolSAR-tailored search space and an improved one-shot search strategy are equipped with the proposed method. By PDAS, the architecture (corresponds to the hyperparameter but not the topology) parameters can be optimized with high efficiency by a stochastic gradient descent (SGD) method. The optimized architecture parameters should be transformed into corresponding architecture and retrained to achieve classification. In addition, a complex-valued PDAS (CVPDAS) is developed to fit the data form of PolSAR images so as to improve the performance. Experiments on three benchmark data sets show that the architectures obtained by searching have better classification performance than hand-crafted ones.

In recent years, using mid-level features to interpret images has attracted more and more attention. Mid-level features can solve the semantic gap between the low-level features and the high-level features. This paper proposes a PolSAR image classification method using Support Vector Machines (SVM) classifier based on an improved bag-of-visual-words (BOW) model combined with contextual information to generate mid-level semantic features. In order to solve the synonym word or polysemic word problem caused by the uncertainty of the number of visual words in the traditional BOW model, the improved visual-vocabulary construction method is proposed by defining visual words using the prior knowledge to reduce the uncertainty and feature dimension effectively. Since the traditional BOW representation ignores the correlation between words, the scale context information between words will be integrated by down-sampling to construct a multi-scale visual-vocabulary. Moreover, comparative experiments are implemented to prove the feasibility of the proposed method.

With the development of representation learning, deep learning based methods have become the state-of-the-art method in some related fields of pattern recognition. This phenomenon brings new challenges and opportunities to polarimetric SAR image interpretation. In this paper, we propose a novel classification method for polarimetric SAR image based on a fresh technique in deep learning: DenseNet. A 20-layers (with 3 dense block and 2 transition layers) DenseNet is built to implement polarimetric SAR image classification. The proposed method effectively prevents gradient vanish and overfitting by feature reuse while automatically extracting high-level features and performing pixel-wise multi-class classification. Last but not least, the proposed method achieves the state-of-the-art experimental result on PolSAR Flevoland 15-class benchmark dataset.

A large-scale SAR image covers various land covers and targets with different complexity and statistical characteristics. Single algorithm always fails to deal with the interpretation tasks of large-scale remote sensing images, due to the mismatch between local background conditions and the algorithm requirements. This paper proposes a novel processing method named Image-Key-Element Decomposition, aiming at decomposing a SAR image into several regions under a certain criterion. Every decomposed key element is the subset of the whole image and represents a pixel set with a particular characteristic. Considering that many algorithms have certain requirements about the complexity status of backgrounds, homogeneous is used as the criterion in this paper and the SAR images will be decomposed into homogeneous and heterogeneous elements. The experimental result shows that homogeneous and heterogeneous regions are separated by the proposed model and the following processing may take advantage of the result to improve the performance.

Aiming at the problem that only the boundaries near the radar appear as strong scattering points and the other boundaries are weakly scattered, this paper proposes an azimuth angle estimation method of vehicle based on target restoration. Through in-depth analysis of the vehicle target scattering structure and error sources, the method introduces the principle of symmetry to restore the target structure as much as possible, and uses the restored target to perform azimuth angle estimation, which solves the problem of inaccurate estimation value and unstable error of traditional methods. The experimental results of MSTAR measured data verify the accuracy and stability of the method.

At present, the Internet of Things (IoT) has attracted more and more researchers' attention. Electromagnetic scattering calculation usually has the characteristics of large-scale calculation, high space-time complexity, and high precision requirement. For the background and objectives of complex environment, it is difficult for a single computer to achieve large-scale electromagnetic scattering calculation and to obtain corresponding large data. Therefore, we use Finite-Difference Time-Domain (FDTD) combined with Internet of Things, cloud computing, and other technologies to solve the above problems. In this paper, we focus on the FDTD method and use it to simulate electromagnetic scattering of electrically large objects. FDTD method has natural parallelism. A computing network cluster based on MPI is constructed. POSIX (Portable Operating System Interface of UNIX) multithreading technology is conducive to enhancing the computing power of multicore CPU and to realize multiprocessor multithreading hybrid parallel FDTD. For two-dimension CPU and memory resources, the Dominant Resource Fairness (DRF) algorithm is used to achieve load balancing scheduling, which guarantees the computing performance. The experimental results show that the hybrid parallel FDTD algorithm combined with load balancing scheduling can solve the problem of low computational efficiency and improve the success rate of task execution.

CFAR is a widely used algorithm for target detection in SAR images. The simplicity of computation and stable performance make it a key role in practice. However, for man-made target detection, conventional CFAR has a limited performance because of the non-adaptive processing window and the varieties of categories, sizes and structures of targets. In order to detect man-made targets with different size and complex structures in high resolution SAR images, an improved CFAR algorithm with an adaptive processing window named Adaptive-Window CFAR is proposed. A global guard window obtained by pre-detection adaptively is used to take place of the guard window in conventional CFAR makes AW -CFAR an algorithm with both adaptive threshold and adaptive processing window. In this case, the size of detectable targets is not fixed anymore and targets of different sizes and complex structures are detectable in AW-CF AR. Images with different resolutions and environments, which contain different categories of man-made targets, are used in the experiments. The experimental results show that the AW -CFAR inherits the simplicity of computation and stable performance of the conventional CFAR and has a better performance of man-made target detection in high resolution SAR images.

Terrain classification is an important application of polarimetric SAR (PolSAR) data. Traditional classification methods need to extract the feature and then classify by classifiers. Besides, it should consider the influence of speckle noise. As a new method for image processing, convolutional neural network (CNN) has attracted more and more attention because of its good performance in image processing. It can deal with the original image directly with a higher classification accuracy without considering the impact of speckle noise. Moreover, three-dimensional convolutional neural network (3D CNN) has stronger feature extraction capability compared with traditional two-dimensional convolutional neural network (2D CNN). In this paper, the application of 3D CNN in terrain classification is studied, in which a new convolutional neural network architecture is designed and the elements of polarimetric coherency matrix are used as the input data of this network. The experiments of two real PolSAR data are conducted to verify the performance of the proposed network.

Polarimetric SAR interferometry (PolInSAR) is a promising remote sensing technique for extracting forest height in which the random volume over ground (RVoG) model has been extensively applied to polarimetric interferometry SAR data for the retrieval of forest geophysical parameters. The complex interferometric coherence of the RVoG model was originally derived in a simplified way by neglecting some factors, like the terrain slope and the double bounce interaction with the stems or trunks. But in many cases, none of them should be ignored. Therefore, a slope three-layer scattering (STLS) model is proposed to correct the terrain distortion for forest parameters estimation in a sloping forest area. At the same time, this model separates the vertical structure of forest into three layers: ground, tree-trunk and the canopy layer which account for the simultaneous effects of three scattering components on complex coherence. A significant model complexity reduction is achieved by aligning the reference frame along the local terrain slope and changing the corresponding radar geometrical configuration. The STLS model provides a better understanding of microwave scattering process in the terrain slope area compared to traditional RVoG model.

Morphology profiles (MPs) have been applied to the processing of different types of imagery, which have highly improved the segmentation and classification results. MPs can both preserve spatial structures of objects and construct the multiscale description of images. Since the speckle noise inherent in the images, the classification of PolSAR images cannot obtain satisfied results. In this paper, the framework of polarimetric object-based morphological profiles (POMPs) is proposed in which the MPs features and polarimetric features are combined and object-based image analysis (OBIA) is introduced to PolSAR image classification. Based on statistical measures of central tendency, the POMPs are obtained by integrating the feature vector of images with the segments of meanshift segmentation. To evaluate the effectiveness of the proposed method, POMPs are introduced to support vector machine (SVM) classifier. Experiments on the EMISAR image using object-based polarimetric features, object-based MPs and POMPs features respectively are implemented and compared. The results show that the proposed method can effectively improve classification accuracy of high resolution PolSAR images.

A coastline detection method for PolSAR image, which uses polarimetric characteristics and mathematical morphology, is proposed in this paper. In this method, both intensity and orientation angle are used to separate sea water and land. Then, in order to reduce the impact of cultivated land affection during aforementioned process, i.e., fuzzy boundary and small cavities, mathematical morphology is utilized for better coastline processing. Experiments are conducted over the UAVSAR L-band PolSAR data. The results show that the proposed method can effectively and accurately detect the coastline in PolSAR images.

Polarimetric target decomposition is the most commonly used method of extracting information from polarimetric synthetic aperture radar (SAR) images. Coherent target decomposition methods are usually suitable for high-resolution images. Recently, Paladini reviewed coherent target decomposition methods and proposed a new approach, lossless and sufficient target decomposition (LSTD), using the special unitary matrix SU(4). However, this method suffers from parameter dependence and commutation problems that could introduce errors in parameter estimation such as an erroneous odd-even bounce ratio. In order to overcome these problems, a new model to decompose the circular polarization scattering vector is proposed. In this paper, the model applies a mapping from SU(4) to SU(6) to simplify the target representation while meaningful parameters, which are independent, can be extracted. Fully polarimetric L-band UAVSAR data are used to validate the proposed method. The most important odd-even bounce ratio parameter is used to compare the estimation accuracy between the proposed method and LSTD. Results show that the proposed method can extract parameters more accurately.

Compared to traditional SAR, high-resolution PolSAR not only can provide texture and geometry information, but also can provide polarimetric information, which have been used extensively for various surface features recognition. Traditional methods only based on images characteristics which don't apply to high-resolution PolSAR images interpretation, causing the algorithm redundancy and low recognition rate. However, human's image cognition system is an efficient and intelligent image processing system, which nothing can be comparable to in targets recognition. Based on human image cognition mechanism, a new method for features recognition in PolSAR images is proposed in this paper to overcome above shortcomings. The proposed method utilizes hierarchical cognition model to identify different features: the first layer is visual cognition, the second layer is logical cognition and the third layer is psychology cognition. Image segmentation and visual sensitive features extraction and integration are applied in visual cognition to derive preliminary recognition results. Based on the results from first procedure, fuzzy logic theory and Neural Network Model are both utilized in logical cognition. Background characteristics are utilized to identify features precisely in psychology cognition. The whole cognitive procedure is under the guidance of a priori knowledge, which is represented in accordance with production rules. Experiments are conducted over the EMISAR L-band PolSAR data and the E-SAR L-band PolSAR data. The results show that the proposed method can effectively and precisely identify different features in PolSAR images.

PolSAR images have been used extensively for various surface features recognition and buildings recognition is an important research topic of PolSAR image interpretation. Traditional methods are only based on PolSAR image characteristics and lack subjective knowledge of human image cognition, making low target recognition rate and algorithm redundancy. To overcome this shortcoming, based on human image cognition mechanism, a new method for buildings recognition in PolSAR images is proposed in this paper. The proposed method utilizes hierarchical cognition model to identify buildings: the first layer is visual cognition and the second layer is logical cognition. In visual cognition, visual sensitive features are extracted and integrated under the guidance of a priori knowledge to derive preliminary recognition results. In logical cognition, based on the results from first process, fuzzy logic theory and Neural Network Model are both utilized to identify buildings precisely. The whole cognition procedure is guided by the knowledge, which is represented in accordance with production rules. Experiments are conducted over the EMISAR L-band PolSAR data, the E-SAR L-band PolSAR data and Convair-SAR C-band PolSAR data. The results show that the proposed method can identify buildings from PolSAR images effectively and precisely.

With the improvement of SAR image resolution, more accurate detection method fit for complicated scenes should be developed to satisfy most commercial and military customers' interests. In this paper, a novel case based reasoning (CBR) method is investigated and well exploited to rule out a mathematical model which performs high detection accuracy in very high resolution SAR images. The cordial innovation of the proposed method is a feature based case library which is established through empirical knowledge as well as existing interpretational results. Each target to be detected will be matched with the most likely case in the case library and its detection result can be determined according to the description of the matched case. The experimental results from several real SAR images with resolution of 0.1m×0.1m show that the satisfactory performance of interpretation can be obtained by the proposed method.

A building detection method for PolSAR image, which is based on the human's visual cognition mechanism, is proposed in this paper. Study on human's visual cognition mechanism is done and it is found that human's visual cognition system is efficient and intelligent in target detection due to efficient processing units and structure, as well as complete experience and knowledge. Considering both the human's visual cognition mechanism and PolSAR image characteristics, the proposed building detection method selects appropriate primary features from PolSAR image, integrates primary features to derive interest map under the guidance of priori knowledge, and then detects building areas based on the interest map. Scale filter and shape judgment are also utilized to remove false alarms. The EMISAR L-band PolSAR data is used to validate the proposed method and the results show that the proposed method can effectively detect building areas from PolSAR images.

In this paper, a new high-precision electromagnetic scattering computing model, which is based on the parallel finite-difference time-domain (FDTD) method, is proposed to compute the SAR echo. Since the FDTD method can only compute radar return at a fixed azimuth angle once, several FDTD simulations should be combined so that a flight path can be simulated. The FDTD method suffers from low computation speed and large computation cost, thus a parallel FDTD model based on MPI library is built to overcome these shortcomings in this paper. The purpose of the computing model is how to use the parallel FDTD to get high-resolution SAR echoes rather than imaging techniques or electromagnetic propagation and scattering effects. Then the SAR echoes of some typical objects including tank and car are computing. To validate the correctness of the proposed model, SAR images derived from the computed radar echoes are given in this paper. To evaluate the performance of the proposed method, experiments are done with different number of threads and PCs. The results show that big improvement has been made in speedup and computational scale by the proposed model.

Summary form only given. Grid-based model is used to describe 3D objects in many circumstances. It can represent fine structure of objects by using small grid. However, small grid causes problem that data of grid-based model occupies much space, which leads to difficulties of transmission and storage. This paper presents an effective compression method for 3D data. In this method, 8-byte float coordinates of each grid are transferred to 1-bit binary codes. Then the binary data is coded by octree further. The experiment result shows that the 3D data can be efficiently compressed. The method is lossless. The complete raw data can be obtained by decoding.

This paper investigates the segmentation problem for very high resolution airborne synthetic aperture radar (SAR) images. In addition to the instinct speckles, these images show two extra characteristics: scene complexity and intensity inhomogeneity, which make segmentation more difficult. An unsupervised solution is proposed based on level set method. First, a new level set evolution method is put forward, it can get global minimum without initial contour, thus can handle complex images automatically. And the new evolution function also introduces the localizing idea from region-scalable-fitting (RSF) model to deal with the intensity inhomogeneity. Then the two segmentation results for background and targets are fused. The experimental results on real images demonstrate the effectiveness of the proposed method.

Feature extraction and image classification using PolSAR images is currently of great interest in SAR applications. On the basis of the sparse characteristics of the features for PolSAR image classification, a supervised PolSAR image classification method based on sparse representation is proposed in this paper, in which the test data can be firstly projected onto a subset of training vectors from the dictionary, then the residual errors with respect to each atom are evaluated and considered as the criteria for classification, and the ultimate class results can be obtained according to the atom with the least residual error. In addition, a Simplified Matching Pursuit (SMP) algorithm is proposed to solve the optimization problem of sparse representation of PolSAR images. The experimental results of Danish EMISAR L-band fully polarimetric SAR data of Foulum Area (DK) confirm our method outputs an excellent result and moreover the classification process is simpler and less time consuming.

Polarimetric SAR (POLSAR) can offer more information than single-polarized SAR that has the capability to image in all weather and day-night conditions. The multi-polarization mode and wide swath requirements result in a huge amount of data, which may exceed the on-board storage and downlink bandwidth. Effective compression of POLSAR raw data is clearly one of the best solutions. This paper presents a compression method for POLSAR raw data with the relative phase between co-polarized and cross-polarized channel being reserved. In this method, the amplitude is quantized for Rayleigh distribution with the phase of HH and VV channel quantized for uniform distribution. In order to reserve relative phase information, the phase difference between co-polarized (HH and VV) and cross-polarized (HV and VH) channel is optimally quantized for triangular distribution. Results show that by quantizing the phase difference, the relative phase information between HH (VV) and HV (VH) channel can be well preserved.

To broaden the future application of CHANG'E-1 imagery, including hyperspectral imagery (low spatial resolution of 200m) and CCD imagery (relatively high spatial resolution of 120m), an ARSIS-based method for spatial-spectral fusion is proposed in this paper, which aims at combine high spatial and high spectral resolution. Firstly, ARSIS concept is employed, in which Àtrous wavelet is used to describe images at different resolutions for multiresolution analysis. Secondly, Pulse Coupled Neural Network (PCNN) is employed to search and model a relationship between the high frequencies of the images to be fused for missing information. The ARSIS method preserves the spectral content of the original image for its very definition, and Àtrous wavelet and PCNN prove to be effective means to implement it on CHANG'E-1 Imagery. The experimental results demonstrate that the visual improvement and spectral fidelity of the proposed method outperform many conventional methods of image fusion.

Polarimetric SAR (POLSAR) can offer more information than single-polarized SAR. At the same time dual-channel POLSAR results in much more data than the single-polarized SAR which is difficult to process on board. This paper presents a method for dual channel POLSAR raw data compression with the relative phase between HH and HV channel being reserved. In this method, the amplitude is quantized for Rayleigh distribution with the phase of HH and VV channel quantized for uniform distribution. The phase difference between HH and HV channel is optimally quantized for triangular distribution. Results show that by quantizing the phase difference, the relative phase information between HH and HV channel can be well preserved. Also the method can be used in full-polarimetric mode.

An asymmetric decomposition method is proposed in this paper, which describes surface, double-bounce, volume, and asymmetric scattering terms. The proposed method is based on the orientation-angle-compensated coherency matrix and is used to deal with non-reflection symmetric scattering cases. The proposed decomposition method applies the correlation coefficient between like- and cross-polarized channels to solve the unknown parameters and discriminate the urban areas and the natural distributed areas. It has been found that the asymmetric scattering is predominant in urban areas and disappears in natural distributed areas. The results of the experiment confirmed that the proposed decomposition method can be used effectively to distinguish different types of terrains.

Polarimetric hyperspectral images can provide spectral, spatial, and polarimetric information of a scene, which are unique and comprehensive for remote sensing applications such as growth monitoring of crops, analysis of water quality, and geology mapping, etc. The researches on polarimetric hyperspectral imaging mechanism and on image characteristics are of great importance for further information extraction and utilization of the images. The purposes of this paper are to analyze the mechanism of polarimetric hyperspectral imaging and to model such a process. The outcome of the paper will help designers and users of a polarimetric hyperspectral imaging system to further understand the system and take full advantages of it. In this paper, a polarimetric hyperspectral imaging model is proposed, in which the influence of skylight on polarization is considered, and subpixel model, polarized reflectance models, and the classical fast canopy reflectance model are combined to model the vegetation canopy. Then, a simulated scene that includes a woodland area with low shrubbery and a road is obtained by using the imaging model. Experiments analyze and discuss the simulation condition and parameters of the imaging models, the uniqueness, and usefulness of the integration of polarimetric and spectral information.

In this paper, the bandwidth of channels is finite in a practical video compression and transmission system, down-sampling techniques are usually adopted to reduce the rate of bit stream in the transmitter before video compression. Thus the reconstructed frames need to be inserted into the video sequences as the missing frames in the receiver after video decompression. Based on traditional temporal frame interpolation method using bidirectional motion vectors, an improved algorithm is proposed to classify the motional relationship to determine the weighting ratio between the adjacent frames, which is used to reconstruct the missing frame by interpolation.

Polarimetric SAR Interferometry (PolInSAR) provides a useful tool for forestry parameters retrieval at L-band. Three-stage inversion process is a traditional way to retrieve these parameters, but it has several shortcomings while ground phase estimation is processed. In order to overcome these shortcomings, an algorithm of ground topography estimation using PolInSAR image by means of coherence set is proposed in this paper. The coherence region derived from numerical range is combined with RVoG model to obtain accurate estimation of ground topography. Simulated image and SIR-C image of forestry area are used to validate the proposed algorithm.

PolInSAR is more advantageous than either polarimetric or interferometric SAR alone in extracting information of vertical object structure. Building parameters could be extracted from PolInSAR data based on scattering model by using nonlinear inversion algorithms. However, inversion results are subject to initialization parameters. To overcome this problem, ESPRIT could be introduced combined with scattering model. In this paper, a method based on polarimetric target decomposition and polarimetric correlation coefficient is used to detect building areas and then building parameters are estimated using a hybrid scattering model and ESPRIT algorithm. Experimental results indicate that building parameters could be effectively extracted by the proposed method.

Target analysis and detection using Polarimetric Synthetic Aperture Radar (PolSAR) image is currently of great interest in SAR applications. The scattering mechanism may be very complex because of speckle and the vector superposition of the scattering echo. In the existing polarimetric features, Polarimetric Similarity Parameter (PSP) is an effective parameter to analyze the scattering characteristics. Based on the similarity or coherence of the target in the multiple PolSAR images, the Multiple PolSAR Similarity Parameter (MPSP) is proposed and defined using the eigenvalues of two polarimetric coherence matrices. Therefore, the characteristic of a target can be described and extracted using MPSP, and then the similarity-enhanced target detection method based on MPSP is implemented and demonstrated with DLR E-SAR L-band multiple-temporal PolSAR images of Oberpfaffenhofen test site. The results confirmed that the proposed method is effective for detection and analysis of buildings in urban areas.

China's first lunar orbiter Chang'e-1 (CE-1) has the capability to concurrently capture hyperspectral imagery with resolution of 200 m along with CCD imagery with a higher resolution of 120 m. To better serve for future lunar terrain exploration, and preparation work for the landing of probe vehicle, resolution of hyperspectral images from CE-1 needs to be enhanced. In order to make use of information from a high-resolution CCD sensor, this paper introduces a resolution enhancing method combining MAP estimation with SMM, and develops an iteration optimization algorithm and a Possion optimization algorithm. Through a set of experiments, it is demonstrated that the proposed method is more effective for practical use compared with conventional methods in resolution enhancement of hyperspectral imageries of CE-1.

For SAR signals with different frequency, the scattering characteristic of objects is generally different. Hence, more information can be obtained with multi-frequency SAR. It is easier to detect different targets by using multi-frequency PolSAR data, and even for the man-made targets in forest area. To effectively detect natural and man-made targets, the scattering components of different scattering mechanisms obtained from Land C-band PolSAR data by using polarimetric target decomposition could be combined. The experiment result indicates that different targets can be detected by using multi-frequency PolSAR data.

Polarimetric synthetic aperture radar (SAR) interferometry (PolInSAR) combines SAR polarimetry and SAR interferometry and is much more sensitive to the distribution of orientated scatterers compared with polarimetric or interferometric data alone. The polarimetric similarity parameter is an efficient parameter to analyze target characteristics using the similarity between a target and the canonical target. In this letter, the polarimetric interferometric eigenvalue similarity parameter (PIESP) is proposed based on the similarity between two polarimetric SAR images obtained by two interferometric antennas. The PIESP is defined by the eigenvalues of two polarimetric coherence matrices in the PolInSAR system, and the eigenvalues of polarimetric coherence matrix are independent on the target orientation angle; therefore, the PIESP is rotation invariant. PolInSAR systems use two antennas to measure the same ground area with slightly different image geometry. Thus, the PIESP can be used to distinguish the target based on coherence and similarity. Then, the target detection method using the PIESP is implemented with the DLR experimental SAR L-band full polarized image of the Oberpfaffenhofen test site of Germany obtained on September 30, 2000. The results confirmed that the proposed model is accurate and effective for the detection and the analysis of buildings in urban areas.

Target detection is a hot topic and key technique of SAR image interpretation. There are many detection methods, such as CFAR detector and Extended Fractal (EF) feature detector. In order to overcome their shortcomings and combine their merits at the same time, the combination of some different detection methods need be implemented. Granularity computing is just an approach that solves the problem at different granularity space due to different principles. Therefore, SAR image target detection based on granularity synthetic algorithm of quotient space theory is proposed in this paper. Firstly, CFAR detector and EF feature detection method are performed to generate different detection results as coarse granularity spaces. Then combine the different quotient spaces and construct the fine granularity space by using granularity synthesis algorithm. Finally, obtain the final target detection result. The experimental result of RADARSAT-I C band SAR image proves that the proposed algorithm is effective.

This paper proposes a two-stage registration method for SAR and optical images based on multi-features and multi-constraints. In the first stage, closed regions are extracted automatically to achieve the coarse mapping parameters as geometrical restriction. In the second stage, Harris corner points and cross-road features are extracted, and then correlation analysis and mutual information are utilized to match the corresponding control points. After that, multi-constraints are used to delete the false matched points. The retained ones are served as ground control points for registration. The experimental results show that the method can reduce the possibility of false matching effectively and the registration error is within one pixel.

Hyperspectral image provides rich information of ground covers, which has been brought great attention. Combining hyperspectral spatial and spectral constraint, an image segmentation method based on region active contour is proposed in this paper. The energy function in Chan-Vese's method is improved and both spatial and spectral information are employed. Spatial term of the function is restricted by global spatial information and intensity consistency in homogeneous region, while spectral term is restricted by spectral similarity of the target. Finally the image is segmented by minimizing the energy function. In order to verify the effectiveness of the proposed method, experiments are conducted on AVIRIS data. Results show that the method can extract indistinct contours of interested target in hyperspectral image, and reduce the influence caused by mixed pixels and the shadows around the target, and also adapts to complex scene in some degree.

POLSAR image classification plays an important role in remote sensing. POLSAR data are a type of mass data and have more independent features which can represent different physical significances than optical image. Therefore, POLSAR image classification is actually a high dimensional nonlinear mapping problem. Because of the nonlinear mapping function of BP neural network, it can be used to classify POLSAR image. But BP neural network classifier is sensitive to initial weights and thresholds. Quantum Clonal Evolutionary Algorithm (QCEA) can converge to an optimal value quickly and can be used to optimize the initial weights and thresholds of BP neural network. Therefore, in this paper, BP classifier based on QCEA was used for POLSAR image classification. Firstly, optimize the initial weights and thresholds of BP neural network using QCEA. Secondly, train the optimized BP neural network classifier by gradient descent algorithm. Finally, classify the POLSAR image using the trained classifier. The validity test is demonstrated using Danish EMISAR L-band fully polarimetric data of Foulum Area (DK), Denmark. The preliminary result indicates that this method can classify most of the areas correctly.

Hyperspectral imaging (HSI) systems can acquire both spectral and spatial information of ground surface simultaneously and have been used in a variety of applications such as object detection, material identification, land cover classification, etc. Through simulation of a HSI process, it is in favor of finding key contributors to optimize system performance and sensor design. Although it is difficult to be expressed accurately, a hyperspectral image simulator could be built on the premise of some simplifications. In this paper, a complex scenario that includes a woodland area with low shrubbery and a woodland area surrounded by high building walls is simulated. The simulated scene model is built, in which a bidirectional reflectance distribution function (BRDF) is introduced and reflectance functions for ground materials are represented using a combination of canopy BRDF and a subpixel fraction model. To improve our previous work, we introduce in the model the secondary illumination based on geometric analysis methods to simulate the effects of reflected solar radiance from other buildings, which improves atmospheric modeling on the basis of nonhomogeneous and non-Lambertian surface. Experiments and analysis show that reasonable results could be achieved.

Hyperspectral images have been used in many applications including object detection, land cover classification and material identification. By building simulation models of hyperspectral system, the key contributors to system performance might be found. Scene model is the most complex subsystem in the hyperspectral simulation system. However, through various simplifying assumptions, simulation model of hyperspectral scene could be built. A scene model based on Monte Carlo simulation approach has been presented by Kerekes. In support of some special scene, such as a park with low shrubbery and high building walls around, Kerekes's hyperspectral simulation model is extended to an artificial woodland scene model in this paper. Bidirectional reflectance distribution functions (BRDFs) has been introduced into the new model and the reflectance functions for the ground materials are represented by combination of canopy BRDFs and subpixel fraction model. The new model also introduces secondary illumination based on geometric analyzing method that can be used to simulate the effects of reflected solar radiance from other building. To study the performance of the model, some validation experiments have been implemented and reasonable results have been achieved.

The motion of moving targets causes distortions in SAR image, which present a challenge on image interpretation. Comparing to stationary targets' signatures, signatures of moving target are distorted from azimuth displacement to smearing beyond detectability. In this paper, a moving target detection algorithm is proposed based on polarimetric whitening filter (PWF) as well as time-frequency (TF) analysis. An experiment is implemented using multi-look X-band PolSAR image acquired by PiSAR. Both moving target and stationary target are included in the image. Results shows PWF method and time-frequency analysis could be used detect moving targets as respected.

A multiple-component scattering model (MCSM) is proposed to decompose polarimetric synthetic aperture radar (PolSAR) images. The MCSM extends a three-component scattering model, which describes single-bounce, double-bounce, volume, helix, and wire scattering as elementary scattering mechanisms in the analysis of PolSAR images. It can be found that double-bounce, helix, and wire scattering are predominant in urban areas. These elementary scattering mechanisms correspond to the asymmetric reflection condition that the copolar and cross-polar correlations are not close to zero. The MCSM is demonstrated with a German Aerospace Center (DLR) Experimental Synthetic Aperture Radar (ESAR) L-band full-polarized image of the Oberpfaffenhofen Test Site Area (DE), Germany, which was obtained on September 30, 2000. The result of this decomposition confirmed that the proposed model is effective for analysis of buildings in urban areas.

Based on the basic principle of PolInSAR and the coherent scattering model of random volume over ground, the inversion of forest parameters of PolInSAR can be characterized by a six-dimensional non-linear parameter optimization problem. However, the global optimal can't be obtained using the traditional gradient-based optimization algorithms. Therefore, a global optimization inversion scheme of forest parameters of PolInSAR based on genetic algorithm is presented. We generate a validity test using SIR-C L-band repeat-pass PolInSAR data of the area of Tien Shan, China. The preliminary results accord with the range of the parameters of the fact. Performances of different GAs and effects of different parameters are compared. SGA is influenced by the mutation rate strongly, but GA with tournament of two generations is independent of the mutation rate.

With the emerging of multi-frequency InSAR system, extraction of 3-D and variation information of earth surface using multi-frequency InSAR data has drawn more attention. Methods based on Maximum-Likelihood Estimation (MLE), combining with some conventional phase unwrapping algorithm can be used to achieve this purpose. In this paper, MLE integrated with weighted multigrid phase-unwrapping method for high-sloped terrain profile reconstruction is presented and studied in detail. The performance of this method is also analyzed. Experiment result using simulated InSAR data shows that the method can be used to reconstruct high-sloped terrain with relatively high accuracy.

This paper introduces a method for building location information extraction using airborne C-band polarimetric SAR data. The method is based on the analysis of building feature. The method decomposes the scattering covariance matrix into three simple mechanism, i.e., odd-bounce; even-bounce; cross-bounce. Taking into account that the typical strong T-shaped echoes from quite large buildings are visible, a method is introduced to extract the location of the buildings.

SAR interferometry allows height reconstruction of the earth surface. A method based on the use of multi-frequency interferograms and Maximum Likelihood Estimation (MLE) has recently been proposed. However without a priori knowledge of the terrain, the result of the reconstruction is unsatisfied in practical cases. In this paper, we present a novel method to reconstruct highly sloped and discontinuous terrain height profiles using multi-frequency interferograms. It is based on MLE using multi-frequency interferograms joint statistic property, combining with some conventional signal frequency phase unwrapping algorithm. The method can not only improve efficiency of the MLE, but also ensure reliability of the estimation.

Data fusion is widely used in many fields in the last two decades. With the development of the hyperspectral sensor technique, the concept of data fusion is introduced into the classification investigation of hyperspectral data recently. Consensus theory is one of the data fusion methods, in which how to properly choose and assign the weights is very important for the improvement of fusion classification accuracy. In this paper, a new method of hyperspectral image classification is studied, which is realized by two key steps: division of data sources based on adaptive subspace decomposition (ASD) and fusion classification based on consensus theory. In order to testify the effectiveness of the proposed method, computer simulations are conducted on AVIRIS data. The experiment investigation shows that the classification result in our new method is improved compared with both the equal weights and conventional approach in the full data space.