Polarimetric Synthetic Aperture Radar Data Research Articles

Composite Sequential Network With POA Attention for PolSAR Image Analysis

The scattering response of polarimetric synthetic aperture radar (PolSAR) data is strongly target orientation-dependent. Formulating the polarimetric matrix as sequential data by rotating the polarimetric matrix along the radar line of sight would provide rich information about land-cover properties. In this work, we propose a composite sequential network (CSN) with polarization orientation angle (POA) attention to model the polarimetric coherency matrix sequence and explore target scattering orientation diversity features. Three major factors strengthen the proposed method for PolSAR image analysis. First, CSN improves the feature comprehensiveness by extending the interpretation mode of PolSAR data from spatial polarization to spatial polarization orientation. In this way, CSN could describe polarimetric response dynamics at different orientations. Second, a two-stream composite network with both real- and complex-valued convolutional long short-term memory (ConvLSTM) network is proposed to process the diagonal and off-diagonal elements of the coherency matrix sequence, respectively. Compared to existing real-/complex-valued networks, the CSN explores the significant phase information of the off-diagonal elements by operations in the complex domain. Meanwhile, CSN prevents padding 0 meaninglessly in the imaginary part of the real-valued diagonal elements. Third, during the sequential modeling of the polarimetric matrix, a POA attention mechanism is proposed. Equipped with POA-sensitive decomposition loss, the CSN attends to substantial POA range derived by targets' physical scattering mechanism and learns features closely related to the scattering mechanism. Extensive experiments and analysis on land-cover classification demonstrate the proposed method's robustness and excellence.

Exploring PolSAR Images Representation via Self-Supervised Learning and Its Application on Few-Shot Classification

Deep learning methods have attracted much attention in the field of polarimetric synthetic aperture radar (PolSAR) image classification over the past few years. However, for supervised learning based methods, it is quite difficult to get large amounts of high-quality and labeled PolSAR data in real applications. In addition, there is a problem of poor generalization for the method of specific supervision labels. To solve the above issue, we explore how to learn representations from unlabeled data from a new perspective. In this letter, a self-supervised PolSAR representation learning (SSPRL) method is proposed. Different from supervised learning based methods, SSPRL aims to learn PolSAR image representations via unsupervised learning approach. Specifically, a self-supervised learning (SSL) method without negative samples is explored and a positive sample generation approach and a novel encoder architecture designed for PolSAR images are proposed. Moreover, mixup is implemented as a regularization strategy. Further the convolutional encoder is utilized to transfer the feature representation from the unlabeled PolSAR data to the downstream task, that is, to achieve the few-shot PolSAR classification. Comparative experimental results on two widely-used PolSAR benchmark datasets verify the effectiveness of proposed method and demonstrate that SSPRL produces impressive performance on few-shot classification task compared with state-of-the-art algorithms.

Polarimetric SAR Despeckling With Convolutional Neural Networks

Coherent imaging systems such as synthetic aperture radar (SAR) are subject to speckle, the reduction of which is an active area of study. Methods based on deep convolutional neural networks (CNNs) have recently demonstrated state-of-the-art performance in the removal of additive noise from natural images and speckle from single-channel SAR images. The application of deep learning to multichannel SAR modalities such as polarimetric SAR (PolSAR) is complicated in part by the nature of the data as images of complex-valued covariance matrices. In this article, we propose a CNN-based PolSAR despeckling approach that uses an invertible transformation involving a matrix logarithm to facilitate CNN processing of the PolSAR data. A residual learning strategy is adopted, in which the CNN is trained to identify the speckle component which is then removed from the corrupted image. The experimental results on simulated and measured PolSAR data show the proposed approach to markedly reduce speckle and preserve scene features.

PolSAR Ship Detection Based on Azimuth Sublook Polarimetric Covariance Matrix

Polarimetric synthetic aperture radar (PolSAR) is an important tool for marine remote sensing. In the past two decades, subaperture decomposition is considered as an alternative method to extract ship targets, especially in complex conditions where the detection based on intensity performs not well. The introduction of polarization information can enhance the detection ability of radar. The combination of sublook techniques and polarization will lead to more precise target extraction. The first methods jointly using information of polarization and sublook is the polarimetric internal Hermitian product (Pol-IHP), which nowadays has been corrected as polarimetric Hermitian internal product (Pol-HIP). However, it is limited by the number of sublooks and recently developed as generalized multi-sublooks correlation (GMC). In this study, a novel sublook polarimetric covariance matrix (SPCM) is established, based on which a sublook-polarimetric whitening filter (Sub-PWF) is further proposed for ship detection. The SPCM represents a given pixel using a high-dimensional covariance matrix which utilizes the spectral correlation in imaging progress and the scattering characteristics of different polarization channels. The superiority of Sub-PWF based on SPCM is validated via the measured single look complex (SLC) PolSAR data from RadarSAT-2 (RS-2) and GaoFen-3 (GF-3). Results show this new method can improve the target to clutter ratio (TCR) and reduce the clutter fluctuation. The ROC curves in both small and large scenes proves that Sub-PWF can reach ideal detection performance.

Polarimetric Decomposition-Based Unified Manmade Target Scattering Characterization With Mathematical Programming Strategies

Due to the orientation variation and structural complexity, designing generalized and unified features to highlight manmade target scattering from polarimetric synthetic aperture radar (PolSAR) data is challenging. Inspired by the thought of mathematical programming (MP) and coupled with the model-based decomposition, this article proposes two polarimetric features: scattering contribution combiner (SCC) and scattering contribution angle (SCA) for unified scattering characterization of manmade targets. To this end, a rotated dihedral scattering model is first constructed concerning the analogous difference reciprocal and sigmoid function transformations, which adequately reflects the transition of co- and cross-pol responses caused by the orientation variation. Along with the dipole-like compound scattering models and through designing a discriminant-based model solution method, a fine eight-component decomposition using full polarimetric information is proposed. Through skillfully employing the MP strategies, the proposed decomposition achieves the physical optimization of scattering modeling and reasonable inversion of model parameters. Thus, it can accurately describe the local structure scattering and remarkably improve the overestimation of volume scattering. Subsequently, by analyzing the significance distribution on targets of different scattering mechanisms, the SCC is constructed via the linear/nonlinear combination of scattering contributions on the one hand. On the other hand, by further mining the information implied in the scattering contributions, the SCA is proposed with the strategy of trigonometric function transformation. Experimental results conducted on real PolSAR data not only demonstrate the effectiveness and superiority of the constructed features but also exhibit a clear advantage of fine polarimetric decomposition in scattering understanding, which encourages the use of them for further applications.

Natural Embedding of the Stokes Parameters of Polarimetric Synthetic Aperture Radar Images in a Gate-Based Quantum Computer

Quantum algorithms are designed to process quantum data (quantum bits) in a gate-based quantum computer. They are proven rigorously that they reveal quantum advantages over conventional algorithms when their inputs are certain quantum data or some classical data mapped to quantum data. However, in a practical domain, data are classical in nature, and they are very big in dimension, size, and so on. Hence, there is a challenge to map (embed) classical data to quantum data, and even no quantum advantages of quantum algorithms are demonstrated over conventional ones when one processes the mapped classical data in a gate-based quantum computer. For the practical domain of earth observation (EO), due to the different sensors on remote-sensing platforms, we can map directly some types of EO data to quantum data. In particular, we have polarimetric synthetic aperture radar (PolSAR) images characterized by polarized beams. A polarized state of the polarized beam and a quantum bit are the Doppelganger of a physical state. We map them to each other, and we name this direct mapping a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">natural embedding</i> , otherwise an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">artificial embedding</i> . Furthermore, we process our <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">naturally embedded</i> data in a gate-based quantum computer by using a quantum algorithm regardless of its quantum advantages over conventional techniques; namely, we use the QML network as a quantum algorithm to prove that we <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">naturally embedded</i> our data in input qubits of a gate-based quantum computer. Therefore, we employed and directly processed PolSAR images in a QML network. Furthermore, we designed and provided a QML network with an additional layer of a neural network, namely, a hybrid quantum-classical network, and demonstrate how to program (via optimization and backpropagation) this hybrid quantum-classical network when employing and processing PolSAR images. In this work, we used a gate-based quantum computer offered by an IBM Quantum and a classical simulator for a gate-based quantum computer. Our contribution is that we provided very specific EO data with a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">natural embedding</i> feature, the Doppelganger of quantum bits, and processed them in a hybrid quantum-classical network. More importantly, in the future, these PolSAR data can be processed by future quantum algorithms and future quantum computing platforms to obtain (or demonstrate) some quantum advantages over conventional techniques for EO problems.

Polarimetric Scattering Spectrum Analysis for Target Characterization

This study proposes the spectrum analysis of the scattering-type parameter, θ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FP</sub> , to characterize different land cover targets. Many orthonormal projections of the scattering information onto distinct polarization bases have been proposed in the literature. However, these conventional orthonormal basis projection techniques often provide various target characterization ambiguities and classification. In this work, we propose projecting the target coherency matrix onto several random realizations of the normalized scattering configuration without restriction to the orthogonality constraint. This unique approach helps enhanced target characterization from polarimetric SAR data. We show the efficacy of our proposed technique over different land cover types using the C-band RADARSAT-2 data over SF and the L-band ALOS-2 data over Mumbai.

Nonlocal Model-Free Denoising Algorithm for Single- and Multichannel SAR Data

Among the large number of synthetic aperture radar (SAR) image despeckling approaches existing in literature, nonlocal (NL) filters have received a desirable boost. However, often NL approaches define the similarity criterion based on model assumptions, such as a fully developed speckle model. This assumption may not be verified in high-resolution images of urban environments. To address this issue, a standalone model-free despeckling framework is proposed in this article. The presented approach provides a generic framework for denoising a variety of SAR products, from a single-intensity/amplitude image to polarimetric and interferometric SAR data. In particular, the method is based on the empirical distributional similarity between the patch containing the pixel to be recovered and the patch containing a similar candidate pixel. To decide whether the patches follow a similar distribution, the Kolmogorov–Smirnov test is adapted. Finally, the restoration process aggregates the selected similar pixels based on their relative importance derived from their distribution similarities. To mitigate the blurring effect and preserve the resolution, the inhomogeneity of the ratio image is used to perform the bias reduction step. The designed generic despeckling filter was tested on different products of SAR data. The results show that the method proves to be an unbiased restoration approach and is able to preserve structures and textures. It works fully automatically and efficiently with single and multilook (and multichannel) images.

Hybrid Three-Component Scattering Power Characterization From Polarimetric SAR Data Isolating Dominant Scattering Mechanisms

Rapid advancements have been made in model-based decomposition techniques for polarimetric Synthetic Aperture Radar (PolSAR) data. Improvements have been primarily driven by including additional scattering models to the three-component model-based method first introduced by Freeman and Durden. Nevertheless, the three-component method is still extensively used due to its simplicity and ease of interpretability. Recently, the paradigm of the decomposition strategy has been changed to non-model types with notable success. Thus utilizing this new approach, we propose a hybrid (i.e., combining non-model and model-based) three-component methodology in this work. The proposed method primarily involves three steps: (i) the generalized eigendecomposition technique is first used to determine the optimum volume scattering power, (ii) the residual rank-2 coherency matrix (i.e., volume scattering model deducted) is appropriately transformed using two unitary transformations to decorrelate the odd and even bounce scattering components, and (iii) compute the odd and even bounce scattering power contributions using the newly developed scattering-type parameter obtained from the rank-2 matrix. Each step carries relevant physical significance that is appropriately addressed in this work. The proposed methodology is first demonstrated using some specific coherency matrices from canonical targets and a few matrices extracted from different landcover types from full-polarimetric SAR images. We then apply the proposed method over diverse landcover types using two full-polarimetric SAR images. We compare the results with the state-of-the-art three-component model-based decomposition techniques to validate the effectiveness of the proposed method that deals with the existing challenges of model-based decomposition methods.

Man-Made Target Structure Recognition With Polarimetric Correlation Pattern and Roll-Invariant Feature Coding

Man-made target recognition is of great significance for many applications within microwave remote sensing. The scattering diversity of various man-made target structures makes radar target identification a difficult task. This work aims at mitigating this issue by mining and utilization of man-made target scattering diversity in polarimetric rotation domain with the interpretation tool of polarimetric correlation pattern. The optimal polarimetric roll-invariant feature set is collected from polarimetric correlation pattern. Then, a polarimetric roll-invariant feature coding scheme is developed for man-made target structure recognition. Moreover, polarimetric radar measurement errors in terms of channel coupling and imbalance are also considered. Experimental studies with electromagnetic computation datasets including canonical structures and an unmanned aerial vehicle (UAV) target and real spaceborne polarimetric synthetic aperture radar (PolSAR) data of a ship target are carried out. Compared with the Cameron decomposition, the proposed method exhibits better recognition performance and stronger robustness, especially for oriented man-made structures.

Change Detection in Dual Polarization Sentinel-1 Data With Wilks’ Lambda

When the covariance matrix formulation is used for multilook polarimetric synthetic aperture radar (SAR) data, Wilks' Lambda can be used for change detection between acquisitions at two time points. We briefly describe the theory for dual polarization, diagonal only data. A case study illustrates the technique on Sentinel-1 C-band data covering the international Frankfurt Airport. We successfully detect change including direction of change in the sense that Wilks' Lambda shows whether radar signal increases or decreases over time.

Three-Dimension Polarimetric Correlation Pattern Interpretation Tool and its Application

Polarimetric radar can acquire complete polarization information and is widely used in many applications. However, target orientation relative to the radar line of sight usually exhibits significant influences on the scattering mechanisms. Recently, such target scattering diversity has been successfully characterized and utilized with the polarimetric rotation domain interpretation techniques. In radar polarimetry, target scattering responses are affected by both polarization orientation angle and polarization ellipticity angle. In this vein, this work aims at exploring and utilizing the complete target scattering diversity by extending polarimetric rotation domain techniques to the polarization ellipticity angle dimension. The main idea is to develop a three-dimension polarimetric correlation pattern (3-D PCP) interpretation tool, which can visualize and exhibit targets’ polarimetric rotation domain properties in terms of both the polarimetric orientation and ellipticity angles. Then, a set of global, local, and mutual polarimetric features are proposed to characterize the responses of a 3-D PCP interpretation tool. Especially, the curvatures in differential geometry are first introduced to describe the properties of the 3-D surface. The performance of these new polarimetric features is investigated with spaceborne polarimetric synthetic aperture radar (PolSAR) data. Experimental results demonstrate the advantage of the proposed 3-D PCP features in enhancing the target clutter ratio (TCR), especially for the weak ship area. Based on this, a non-local superpixel-level contrast measure (NSLCM) method for ship detection is proposed. Pure sea samples can be determined adaptively for salient map construction. Comparison results demonstrate better detection performance for both the inshore dense ship area and weak ship area.

Soil Permittivity Estimation Over Croplands Using Full and Compact Polarimetric SAR Data

Soil permittivity estimation using Polarimetric Synthetic Aperture Radar (PolSAR) data has been an extensively researched area. Nonetheless, it provides ample scope for further improvements. The vegetation cover over the soil surface leads to a complex interaction of the incident polarized wave with the canopy and subsequently with the underlying soil surface. This paper introduces a novel methodology to estimate soil permittivity over croplands with vegetation cover using the full and compact polarimetric modes. The proposed method utilizes the full and compact polarimetric scattering-type parameters, θ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FP</sub> and θ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">CP</sub> , respectively. These scattering type parameters are a function of the soil permittivity and the Barakat degree of polarization. The method considers the X-Bragg scattering model for the soil surface. In particular, these scattering-type parameters explicitly account for the depolarizing structure of the scattered wave while characterizing targets. Thus, the depolarization information in terms of surface roughness in the X-Bragg model gets inherent importance while using θ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FP</sub> and θ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">CP</sub> , unlike existing scattering-type parameters. Therefore, the proposed technique enhances the expected value of the inversion accuracies. This study validated the major phenology stages of four crops using the UAVSAR full-pol and simulated compact pol SAR data and the ground truth data collected during the SMAPVEX12 campaign over Manitoba, Canada. The proposed method estimated permittivity with an RMSE of 2.2 to 4.69 for FP and 3.28 to 5.45 for CP SAR data along with a Pearson coefficient, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">r</i> ≥ 0.62.

The analysis on uncertainty resulting from method and wavelength selecting in forest height inversion using simulated polarimetric interferometric SAR data

The analysis on uncertainty resulting from method and wavelength selecting in forest height inversion using simulated polarimetric interferometric SAR data

A Novel Unsupervised Object-Level Crop Rotation Detection With Time-Series Dual-Polarimetric SAR Data

Crop rotation is subsidized by the government because of its many advantages. Monitoring whether crop rotation is beneficial for agricultural management, and can also provide a reference for government subsidy policies for crop rotation. In this paper, we propose an unsupervised object-oriented crop rotation detection method using time-series polarimetric SAR (PolSAR) data. On the one hand, we construct the change detection matrix based on the likelihood ratio test (LRT) distance to perform temporal filtering. Then, the pixel-level temporal change image is generated using Shannon entropy and maximum between-class variance (OTSU). On the other hand, we perform temporal segmentation on time-series PolSAR images to obtain superpixel results. Finally, the object-level crop rotation results are obtained with the probabilistic label relaxation (PLR) model. 42 Sentinel-1 dual-polarization SAR datasets during 2018 and 2019 are selected for detecting crop rotation changes on farms within Jinchang, China. Experimental results show that the crop rotation detection accuracy and <i>Kappa</i> coefficients of this method can reach 96.21% and 0.8989, respectively.

Sensitivity of Multifrequency Polarimetric SAR Data to Postfire Permafrost Changes and Recovery Processes in Arctic Tundra

We used full-polarimetric L-band and P-band synthetic aperture radar (SAR) data collected from the recent NASA Arctic Boreal Vulnerability Experiment (ABoVE) airborne campaign and Sentinel-1 C-band dual-polarization data to understand the sensitivity of radar backscatter intensity and phase to fire-induced changes in the surface and subsurface soil processes in Arctic tundra underlain by permafrost. The 2007 Anaktuvuk River fire on the Alaska North Slope was used as a case study. At ~10-year postfire, we observed a strong increase (>~3–4 dB) in the low-frequency radar backscatter in severely burned areas during the thaw season, in contrast to limited (< ~0.5 dB) C-band backscatter differences (VV and VH) between burned and unburned areas. However, C-band winter backscatter is generally higher (>1 dB) in burned areas than the adjacent unburned areas. Polarimetric decomposition analysis indicated a general trend toward more random surface scattering, and strong increases in double-bounce scattering and volume scattering power at both P- and L-band in the burned areas. The ice-rich yedoma region shows the largest backscatter increases in burned areas and the highest correlation with burn severity and microtopography changes. The above backscatter changes are attributed to increasing surface roughness and microtopography due to ice-wedge degradation and thermokarst development and increasing subsurface scattering due to an overall drier and deeper active layer in burned areas. Among all frequencies, P-band shows consistently larger contrast in backscatter power and phase between burned and unburned areas, which makes it potentially more useful to study fire–permafrost interactions in the Arctic over decadal time scales.

TS-SHES: Terrain Segmentation in Complex-Valued PolSAR Images Via Scattering Harmonization and Explicit Supervision

Convolutional neural network (CNN) has attracted extensive attention in the research field of polarimetric synthetic aperture radar (PolSAR) terrain segmentation. However, directly using CNN in PolSAR terrain segmentation while ignoring the characteristics of PolSAR images has become the main factor restricting the performance of algorithms. In this article, we propose an efficient PolSAR terrain segmentation algorithm called TS-SHES, which integrates the polarization scattering characteristics of PolSAR images and the CNN learning process into a unified architecture. First, considering the intrinsic structure of complex-valued PolSAR data, TS-SHES transforms the scattering matrix into the form of amplitude and phase components, which preserves the original information maximally. Then, TS-SHES introduces a scattering harmonized encoding method (SH-Enc) to balance the feature contributions of weak and strong scattering regions as well as map the two components into the same representation space. Through the above scattering harmonization operations, the segmentation performance of CNN on weak scattering regions can be improved, and the feature imbalance in amplitude and phase can be alleviated. Furthermore, in view of the implicit states of CNN feature construction, a scattering explicit learning network (SEL-Net) is presented to collect the scattering features of amplitude and phase. Via explicit supervision, SEL-Net avoids the incomplete collection of scattering information caused by implicit feature construction, thereby improving the segmentation accuracy. Abundant experiments are conducted on two PolSAR images acquired by the GaoFen-3 satellite, which demonstrates the superiority of our proposed algorithm.

A Novel Crop Classification Method Based on the Tensor-GCN for Time-Series PolSAR Data

Time-series polarimetric synthetic aperture radar (PolSAR) has been proven to be an effective technique for crop classification and agricultural activity monitoring. However, the characterization and utilization of time-series PolSAR data by existing methods are still inadequate. They are unable to extract and utilize time-varying features, which can describe the dynamic changes of crop polarimetric information. In this paper, we propose a tensor form to comprehensively describe the information of time-series PolSAR data, including spatial context information, polarimetric scattering information, and temporal context information. And we define a novel similarity value for the tensors (TSV), which can simultaneously consider distance and shape similarity of tensors. Then, we construct a tensor-based graph representation to capture the global similarity information of time-series PolSAR data. Finally, we propose a tensor-based graph convolutional network (Tensor-GCN) to extract deep features of graph node tensors for crop classification. Experimental results and analysis on two time-series PolSAR data firmly demonstrate the superiority of the proposed Tensor-GCN to other state-of-the-art methods.

A Statistical-Spatial Feature Learning Network for PolSAR Image Classification

Polarimetric synthetic aperture radar (PolSAR) image classification plays an important role in the development of remote sensing image interpretation for the rich polarization information. Generative methods learn the statistical distribution characteristic of the scattered echoes data with heavy noise. However, it is tedious to design and solve different generative likelihood functions. Discriminative methods learn image features and classifier in an end-to-end framework whose classification performance is also limited without polarization statistical characteristics. In order to make full use of the statistical characteristic of PolSAR echoes and spatial feature of PolSAR image, a novel real-value hybrid generative/discriminative (HGD) deep network is proposed to learn statistical-spatial feature for PolSAR image classification with the data expressing ability of the generative term and the end-to-end learning ability of the discriminative term. First, a derived replacement form is obtained from the typical Wishart distribution of PolSAR data with an eigenvalue generative term. In this way, the complex-value form of PolSAR data converts into real-value expression, which makes the PolSAR image classification easy to understand and implement in the deep network. Finally, these two terms are integrated from the variational Bayesian theory with an alternative optimization equation derived for PolSAR image classification. It provides a normal framework for PolSAR multifeature learning. Experiments are tested on different PolSAR datasets compared with several state-of-the-art individual generative or discriminative methods. All visual experimental results and accuracy values demonstrate the superiority of the proposed method.

Pol-NAS: A Neural Architecture Search Method With Feature Selection for PolSAR Image Classification

With the development of deep learning, more and more neural networks have been used in polarimetric synthetic aperture radar (PolSAR) image classification and obtain good results. As we all know, the performances of neural networks highly depend on well-designed neural architectures. Besides, the features input to neural networks also have a huge impact on the classification results. Both architecture design and feature selection are time-consuming and require human expertise. So, in this paper, we propose a neural architecture search method with feature selection (Pol-NAS) for PolSAR image classification. It can automatically search and obtain a good architecture including intra-cell, inter-cell structure and the number of layers in the search stage. Meanwhile, all the features commonly used in PolSAR data interpretation, rather than part of them, are input to the model in order to avoid selecting the size of optimal feature subset, which is a hyper-parameter and usually different for different models. Then, we propose the Feature Attention block (FA block) and redesign the stem layers by combining the FA block and the original stem layers. Thus, Pol-NAS can adaptively find the importance of each feature in the training stage by using the redesigned stem layers. With the help of Pol-NAS, we only need to prepare the data and wait for the classification results. Experimental results on three real PolSAR datasets show that the performance of Pol-NAS is better than that of state-of-the-art PolSAR image classification models. The code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/guangyuanLiu/Pol-NAS</uri> .