Although deep learning-based methods have been successfully applied to polarimetric synthetic aperture radar (PolSAR) image classification tasks, most of the available techniques are not suitable to deal with PolSAR data on irregular domains, e.g., superpixel graphs, because they are naturally designed as grid-based architectures in Euclidean space. To overcome this limitation and achieve robust PolSAR image classification, this article proposes the multiscale evolving weighted graph convolutional network, where weighted graphs based on superpixel technique and Wishart-derived distance are constructed to enable efficient handling of graphical PolSAR data representations. In this article, we derive a new architectural design named graph evolving module that combines pairwise latent feature similarity and kernel diffusion to refine the graph structure in each scale. Finally, we propose a graph integration module based on self-attention to perform robust hierarchical feature extraction and learn an optimal linear combination of various scales to exploit effective feature propagation on multiple graphs. We validate the superiority of proposed approach on classification performance with four real-measured datasets and demonstrate significant improvements compared to state-of-the-art methods. Additionally, the proposed method has shown strong generalization capacity across datasets with similar land covers.

Although graph convolutional networks have found application in polarimetric synthetic aperture radar (PolSAR) image classification tasks, the available approaches cannot operate on multiple graphs, which hinders their potential to generalize effective feature representations across different datasets. To overcome this limitation and achieve robust PolSAR image classification, this paper proposes a novel end-to-end cross-level interaction graph U-Net (CLIGUNet), where weighted max-relative spatial convolution is proposed to enable simultaneous learning of latent features from batch input. Moreover, it integrates weighted adjacency matrices, derived from the symmetric revised Wishart distance, to encode polarimetric similarity into weighted max-relative spatial graph convolution. Employing end-to-end trainable residual transformers with multi-head attention, our proposed cross-level interactions enable the decoder to fuse multi-scale graph feature representations, enhancing effective features from various scales through a deep supervision strategy. Additionally, multi-scale dynamic graphs are introduced to expand the receptive field, enabling trainable adjacency matrices with refined connectivity relationships and edge weights within each resolution. Experiments undertaken on real PolSAR datasets show the superiority of our CLIGUNet with respect to state-of-the-art networks in classification accuracy and robustness in handling unknown imagery with similar land covers.

In the field of target classification, detecting a ground moving target that is easily covered in clutter has been a challenge. In addition, traditional feature extraction techniques and classification methods usually rely on strong subjective factors and prior knowledge, which affect their generalization capacity. Most existing deep-learning-based methods suffer from insufficient feature learning due to the lack of data samples, which makes it difficult for the training process to converge to a steady-state. To overcome these limitations, this paper proposes a Wasserstein generative adversarial network (WGAN) sample enhancement method for ground moving target classification (GMT-WGAN). First, the micro-Doppler characteristics of ground moving targets are analyzed. Next, a WGAN is constructed to generate effective time–frequency images of ground moving targets and thereby enrich the sample database used to train the classification network. Then, image quality evaluation indexes are introduced to evaluate the generated spectrogram samples, with an aim to verify the distribution similarity of generated and real samples. Afterward, by feeding augmented samples to the deep convolutional neural networks with good generalization capacity, the classification performance of the GMT-WGAN is improved. Finally, experiments conducted on different datasets validate the effectiveness and robustness of the proposed method.

Although synthetic aperture radar (SAR) performance and polarimetric data quality are closely related, they represent fundamentally different concepts. This paper delineates their distinctions, investigates their interdependence, and introduces a comprehensive set of technical metrics for evaluating radar system performance and assessing polarimetric data quality. Specifically, radar performance is quantified by seven independent parameters, whereas data quality is characterized by a three-component channel imbalance vector and a twelve-element channel crosstalk matrix. The paper details the measurement methods for these parameters and outlines the associated technical requirements, including calibrator specifications and test-site conditions. To improve operational applicability, an approximate method for data quality assessment is proposed, and its associated errors are analyzed. Special attention is given to the $\gamma$ factor, which is highlighted as a critical and irreplaceable indicator of radar performance. Using field data from the GaoFen-3 (GF-3) satellite, the proposed metrics are applied to evaluate both radar performance and data quality. The results provide insights into the polarimetric characteristics of the system and offer practical guidance for the calibration and application of GF-3 polarimetric SAR data.

Parabolic cylindrical antennas are characterized by their structural simplicity, high radiation efficiency, and low manufacturing costs. Consequently, they are widely used in Earth observation and serve as a viable option for spaceborne Synthetic Aperture Radar (SAR) systems. However, structural errors in the phased array feed and the parabolic cylindrical reflector are inevitable during manufacturing, assembly, and operation. These errors significantly degrade the accuracy of antenna pattern models. To address this issue, this paper proposes a comprehensive radiation pattern model that accounts for structural errors in both the linear feed and the reflector. This approach enables precise pattern prediction and efficient in-orbit calibration. Specifically, the reflected far-field pattern is first calculated using the field superposition principle and the Physical Optics (PO) method. Specifically, the combined phase effects resulting from feed and reflector structural errors are superimposed to establish a direct integration pattern model for the parabolic cylindrical antenna. Given the high computational complexity of the direct integration model, a simplified model based on Fresnel approximation is proposed. This approach significantly reduces integration complexity while preserving the quadratic phase characteristics of the main lobe, thereby substantially improving computational efficiency. Simulation results verify that the simplified model maintains high accuracy in both normalized amplitude and phase. Furthermore, a partitioned calibration method is proposed to compensate for the absolute gain deviation inherent in the simplified model. By integrating weighting relationships derived from sensitivity analysis of individual errors, an empirical parameter is defined to quantify the correlation between total structural errors, antenna performance, and the prediction accuracy of the simplified model. The results indicate that reflector structural errors are the dominant factor affecting the overall performance of the antenna. In contrast, the prediction accuracy of the simplified model is found to be more sensitive to feed structural errors. The simplified model exhibits tolerance to structural errors far exceeding the wavelength, enabling it to effectively replace the direct integration model. This work provides new theoretical foundations and technical methods for tolerance design, performance assurance, in-orbit testing, and calibration of parabolic cylindrical antennas.