Deep learning methods have been widely studied for Polarimetric synthetic aperture radar (PolSAR) land cover classification. The scarcity of PolSAR labeled samples and the small receptive field of the model limit the performance of deep learning methods for land cover classification. In this paper, a vision Transformer (ViT)-based classification method is proposed. The ViT structure can extract features from the global range of images based on a self-attention block. The powerful feature representation capability of the model is equivalent to a flexible receptive field, which is suitable for PolSAR image classification at different resolutions. In addition, because of the lack of labeled data, the Mask Autoencoder method is used to pre-train the proposed model with unlabeled data. Experiments are carried out on the Flevoland dataset acquired by NASA/JPL AIRSAR and the Hainan dataset acquired by the Aerial Remote Sensing System of the Chinese Academy of Sciences. The experimental results on both datasets demonstrate the superiority of the proposed method.

Forests cover approximately one-third of the Earth’s land surface and constitute the core region of the carbon cycle on Earth. The paramount importance and multi-purpose applications of forest monitoring have gained widespread recognition over recent decades. Polarimetric synthetic aperture radar interferometry (PolInSAR) has been demonstrated as a promising technique to retrieve the forest height over large areas with a limited cost. This paper presents an overview of forest height inversion (FHI) techniques based on PolInSAR data. Firstly, we introduce the basic theories of PolInSAR and FHI procedures. Next, we review the established data-based algorithms for single-baseline data and describe innovative techniques related to multi-baseline data. Then, the model-based algorithms are also introduced with their corresponding forest scattering models under multiple data acquisition modes. Subsequently, a case study is presented to demonstrate the applicable scenarios and advantages of different algorithms. Model-based algorithms can provide accurate results when the scene and forest properties are well understood and the model assumptions are valid. Data-based algorithms, on the other hand, can handle complex scattering scenarios and are generally more robust to uncertainties in the input parameters. Finally, the prospect of forest height inversion was analyzed. It is our hope that this review will provide guidelines to future researchers to enhance further FHI algorithmic developments.

Vegetation height estimation plays a pivotal role in forest mapping, which significantly promotes the study of environment and climate. This paper develops a general forest structure model for vegetation height estimation using polarimetric interferometric synthetic aperture radar (PolInSAR) data. In simple terms, the temporal decorrelation factor of the random volume over ground model with volumetric temporal decorrelation (RVoG-vtd) is first modeled by random motions of forest scatterers to solve the problem of ambiguity. Then, a novel four-stage algorithm is proposed to improve accuracy in forest height estimation. In particular, to compensate for the temporal decorrelation mainly caused by changes between multiple observations, one procedure of temporal decorrelation adaptive estimation via Expectation-Maximum (EM) algorithm is added into the novel method. On the other hand, to extract the features of amplitude and phase more effectively, in the proposed method, we also convert Euclidean distance to a generalized distance for the first time. Assessments of different algorithms are given based on the repeat-pass PolInSAR data of Gabon Lope Park acquired in AfriSAR campaign of German Aerospace Center (DLR). The experimental results show that the proposed method presents a significant improvement of vegetation height estimation accuracy with a root mean square error (RMSE) of 6.23 m and a bias of 1.28 m against LiDAR heights, compared to the results of the three-stage method (RMSE: 8.69 m, bias: 4.81 m) and the previous four-stage method (RMSE: 7.72 m, bias: −2.87 m).

Nowadays, polarimetric interferometric synthetic aperture radar (PolInSAR) has attracted increasing attention for the simultaneous acquisition of scattering and terrain information. The scattering mechanism vectors corresponding to the optimal coherences have shown great effect in land cover classification and forest mapping. To extract scattering mechanism vectors, different optimization algorithms are proposed but always accompanied with spatial discontinuities. This letter proposes a method to optimize scattering mechanism vectors based on nonlocal estimations. A pixel‐by‐pixel optimization model is established by calculating patch‐based nonlocal similarity, which effectively reduces the uncertainty of scattering mechanism vector estimation while maintaining high resolution. The phase recovery experiment is conducted on simulated data and the root mean square errors (RMSEs) of three algorithms are compared. The proposed method performs better with a RMSE of 0.22 in the whole image, against 0.25 and 1.27 for other algorithms. The recovery results in the target region (RMSE 0.11 against 0.21 and 0.99) and linear edge region (RMSE 0.26 against 0.49 and 1.52) also indicate strong ability in detail preservation. Then the observed ESAR data is utilized for further assessment. It demonstrates that the proposed method has a significant improvement to reduce singularities and distinguish artificial buildings.

Despite the in-depth understanding of the synthetic aperture-radar (SAR) speckle and its characteristics, despeckling remains an open issue far from being solved. Deep-learning methods with supervised training have made great progress. However, reliable reference images are inconveniently accessible or even non-existent. In this paper, we propose an end-to-end self-supervised method named Speckle2Self for SAR image despeckling, which learns mapping from noisy input to clean output using only the input noisy image itself for training. We formulate the image despeckling as a masked pixel-estimation problem, where a set of masks is carefully designed. The masked pixel values are predicted by the queries of complementary masks indicating the positions of masked pixels through an attention mechanism. Transformer architecture is employed as the network backbone. In addition, a novel loss function is also derived based on the statistics of SAR images, and meanwhile, image downsampling is used to provide guarantees on the white noise assumption involved in our Speckle2Self. We compare the proposed Speckle2Self with reference methods on both synthetic and real images. Experimental results demonstrate that the proposed Speckle2Self achieves comparable despeckling performance with supervised methods, suppressing noise while maintaining structural details. Even compared with self-supervised methods, the proposed Speckle2Self still has significant advantages in SAR image-despeckling metrics.

Advanced deception countermeasures now enable adversaries to inject false targets into synthetic-aperture-radar (SAR) imagery, generating electromagnetic signatures virtually indistinguishable from genuine targets, thus destroying the separability essential for conventional recognition algorithms. To address this problem, we propose a versatile front-end Multi-Feature Extraction and Spatial Transformation Network (MFE-STN), specifically designed for the task of discriminating between true targets and deceptive false targets created by SAR jamming, which can be seamlessly integrated with existing CNN backbones without architecture modification. MFE-STN integrates three complementary operations: (i) wavelet decomposition to extract the overall geometric features and scattering distribution of the target, (ii) a manifold transformation module for non-linear alignment of heterogeneous feature spaces, and (iii) a lightweight deformable spatial transformer that compensates for local geometric distortions introduced by deceptive jamming. By analyzing seven typical parameter-mismatch effects, we construct a simulated dataset containing six representative classes—four known classes and two unseen classes. Experimental results demonstrate that inserting MFE-STN boosts the average F1-score of known targets by 12.19% and significantly improves identification accuracy for unseen targets. This confirms the module’s capability to capture discriminative signatures to distinguish genuine targets from deceptive ones while exhibiting strong cross-domain generalization capabilities.