At present, the majority of deep learning-based ship object detection algorithms concentrate predominantly on enhancing recognition accuracy, often overlooking the complexity of the algorithm. These complex algorithms demand significant computational resources, making them unsuitable for deployment on resource-constrained edge devices, such as airborne and spaceborne platforms, thereby limiting their practicality. With the purpose of alleviating this problem, a lightweight and high-accuracy synthetic aperture radar (SAR) ship image detection network (LHSDNet) is proposed. Initially, GhostHGNetV2 was utilized as the feature extraction network, and the calculation amount of the network was reduced by GhostConv. Next, a lightweight feature fusion network was designed to combine shallow and deep features through lightweight convolutions, effectively preserving more information while minimizing computational requirements. Lastly, the feature extraction module was integrated through parameter sharing, and the detection head was lightweight to save computing resources further. The results from our experiments demonstrate that the proposed LHSDNet model increases mAP50 by 0.7% in comparison to the baseline model. Additionally, it illustrates a pronounced decrease in parameter count, computational demand, and model file size by 48.33%, 51.85%, and 41.26%, respectively, when contrasted with the baseline model. LHSDNet achieves a balance between precision and computing resources, rendering it more appropriate for edge device implementation.

In this article, we have improved the quantitative inversion performance of the cross-correlated contrast source inversion (CC-CSI) method by incorporating the subspace optimization strategy. The proposed method is called the cross-correlated subspace optimization method (CC-SOM). Meanwhile, multifrequency data are used to improve the inversion performance of high-contrast scatterers, where the L-curve method is introduced to select the regularization parameters of each frequency point without relying on experience. Finally, a fast algorithm is implemented by using the property of singular value decomposition (SVD) to simplify the large-scale matrix, and the fast Fourier transform (FFT) to accelerate the calculation. Synthetic and experimental inversion results demonstrate that both CC-SOM and CC-CSI show better robustness than SOM. In comparison to CC-CSI, CC-SOM is superior in terms of inversion accuracy when the regularization parameters have been appropriately selected. However, these advantages come at the cost of higher computational complexity.

Synthetic Aperture Radar (SAR) is a high-resolution imaging sensor commonly mounted on platforms such as airplanes and satellites for widespread use. In complex electromagnetic environments, radio frequency interference (RFI) severely degrades the quality of SAR images due to its widely varying bandwidth and numerous unknown emission sources. Although traditional deep learning-based methods have achieved remarkable results by directly processing SAR images as visual ones, there is still considerable room for improvement in their performance due to the wide coverage and high intensity of RFI. To address these issues, this paper proposes the fusion of segmentation and inpainting networks (FuSINet) to suppress SAR RFI in the time-frequency domain. Firstly, to weaken the dominance of RFI in SAR images caused by high-intensity interference, a simple CCN-based network is employed to learn and segment the RFI. This results in the removal of most of the original interference, leaving blanks that allow the targets to regain dominance in the overall image. Secondly, considering the wide coverage characteristic of RFI, a U-former network with global information capture capabilities is utilized to learn the content covered by the interference and fill in the blanks created by the segmentation network. Compared to the traditional Transformer, this paper enhances its global information capture capabilities through shift-windows and down-sampling layers. Finally, the segmentation and inpainting networks are fused together through a weighted parameter for joint training. This not only accelerates the learning speed but also enables better coordination between the two networks, leading to improved RFI suppression performance. Extensive experimental results demonstrate the substantial performance enhancement of the proposed FuSINet. Compared to the PISNet+, the proposed attention mechanism achieves a 2.49 dB improvement in peak signal-to-noise ratio (PSNR). Furthermore, compared to Uformer, the FuSINet achieves an additional 4.16 dB improvement in PSNR.

Polarimetric synthetic aperture radar (PolSAR) building extraction plays an important role in urban planning, disaster management, etc. In this paper, a building extraction method using refined model-based decomposition and robust scattering feature is proposed. On the one hand, the newly proposed refined five-component decomposition and its derived scattering powers are applied to detect the buildings. On the other hand, by combining the matrix elements and co-polarization correlation coefficient, a robust feature is proposed to discriminate buildings and non-buildings. Both these two preliminary extraction results are obtained through thresholding segmentation. Finally, they are fused via the HX Markov random fields so as to further improve the extraction accuracy. The performance of the proposed method is demonstrated and evaluated with Gaofen-3 and uninhabited aerial vehicle SAR full PolSAR data over different test sites. Outputs show that the proposed method outperforms other state-of-the-art methods and provides an overall accuracy of over 90%.

Due to incomprehensive and inaccurate scattering modeling, the state-of-the-art polarimetric synthetic aperture radar (PolSAR) model-based target decompositions are incapable of effectively depicting the scattering mechanism of obliquely oriented urban areas. In this paper, a seven-component model-based decomposition scheme is proposed by constructing several sophisticated scattering models. First, an eigenvalue-based obliquely-oriented dihedral scattering model is presented to reasonably distribute the co-polarization and cross-polarization scattering powers in obliquely oriented urban areas, thus accurately characterizing the urban scattering. Second, the ±45° oriented dipole and ±45° quarter-wave reflector scattering models are incorporated for the purpose of accounting for the real and imaginary components of the $T_{13}$ element in the coherency matrix so as to fully utilize polarimetric information. Finally, according to their mathematical forms, several strategies for model parameter solutions are designed, and the seven-component decomposition is fulfilled. Experimental results conducted on different PolSAR data demonstrate that the proposed method considerably improves the PolSAR scattering interpretation in a more physical manner compared to other existing model-based decomposition, which can be applied for urban area detection, classification, and other urban planning applications.

Unmanned aerial vehicles (UAV) have become vital targets in civilian and military fields. However, the polarization characteristics are rarely studied. This paper studies the polarization property of UAVs via the fusion of three polarimetric decomposition methods. A novel algorithm is presented to classify and recognize UAVs automatically which includes a clustering method proposed in “Science”, one of the top journals in academia. Firstly, the selection of the imaging algorithm ensures the quality of the radar images. Secondly, local geometrical structures of UAVs can be extracted based on Pauli, Krogager, and Cameron polarimetric decomposition. Finally, the proposed algorithm with clustering by fast search and find of density peaks (CFSFDP) has been demonstrated to be better than the original methods under the various noise conditions with the fusion of three polarimetric decomposition methods.