Synthetic aperture radar (SAR) can operate around the clock and in all weather, and therefore high-resolution SAR images have been frequently applied for ship inspection. However, current ship target detection and identification methods have limited detection accuracy and lead to missing detection of small targets due to speckle noise caused by the imaging principle of SAR imagery and complex nearshore interference. Therefore, this paper proposes an improved YOLOv5 method to address the problem of low accuracy in multi-ship target detection tasks in complex scenes. The developed scheme enhances the ship target detection performance while reducing the number of parameters. Specifically, first, we improve the size of the input SAR images and optimize the anchor frames of ship targets in the training dataset to locate small target ships more accurately. Then, symmetric pyramidal non-local block(APNB) and Sim attention mechanism (SimAM) are introduced to reduce near-shore background interference. Additionally, to make the C3 module output richer and with more feature information, channel shuffling is performed after the C3 output to enhance the information exchange between channels. Finally, to reduce the number of parameters and computational cost during model training, the normal convolution in the NECK part is replaced with Ghost convolution. The index F1 of the proposed method on the High-Resolution SAR Image Dataset (HRSID) and SAR Ship Detection Dataset (SSDD) reaches the highest of 91.3% and 95.8%, respectively. MAPs (0.5:0.95) for the two datasets are both the highest, which are at least 2% higher than the suboptimal method. In selected specific inshore scenes, the ship detection performance of the proposed method outperforms current advanced methods for multiscale ships. It is shown that the proposed method can extract ship features effectively in complex scenes and its effectiveness is further validated on the large scene AIR-SARShip-1 dataset.

To effectively mitigate disaster damage, it is crucial to obtain landslide information quickly and accurately with the abundant remote sensing images. Although related landslide detection research has been carried out a lot in recent years, all of the research articles are still based on single-label image classification. Meanwhile, the existing methods face challenges in balancing global and local information effectively, interpreting and modeling intricate interlabel dependencies. To solve the problems above, we perform multilabel annotation on the Bijie landslide dataset and propose a novel multilabel classification network for landslide detection. The proposed network consists of a feature representation module and a category relation parsing module. The first module is designed to extract feature maps that represent high-level semantic information and convert them into feature sequences. This can fully utilize the capability of the bidirectional long short-term memory network to model label dependencies. The second module is used to model interlabel dependencies to enhance the ability of the model for efficiently detecting landslides. Compared with other recent landslide detecting methods, the results show that the proposed scheme is more effective. F ₁ score on landslide detection reaches up to 97.56% and the attention area of categories is more precise in the proposed method. Therefore, the proposed model can provide reliable and effective decision support for disaster emergency response.

Synthetic aperture radar (SAR) target recognition is essential for SAR image interpretation. It has been widely used in national defense and national economy. At present, the SAR image detection and recognition methods based on convolutional neural network (CNN) have problems such as insufficient extraction of the feature information of SAR image targets, false targets caused by the interference of complex backgrounds, and low detection performance. The main reason is that the feature extraction of CNN is a local operation in space and time, which ignores the correlation between pixels and regions and the dependencies between channels in SAR images. In this paper, a non-local channel attention network (NLCANet) SAR image target recognition method is proposed based on the GoogLeNet structure combined with asymmetric pyramid non-local block (APNB) and squeeze-and-excitation block (SEB). APNB is added to the GoogLeNet framework to capture more context information and enhance the correlation between pixels and regions. SEB is added to the Inception structure to become Inception-SEB (ISEB), through which channel dependencies based on the fusion of different scale features can be obtained. The experimental results based on the moving and stationary target acquisition and recognition (MSTAR) dataset and the SAR ship detection dataset (SSDD) show that the proposed method improves the detection ability of targets in complex backgrounds and achieves better land-sea target recognition performance.

For radar target detection, the clutter distribution model needs to be identified first. The goodness of fit (GoF) between the original data and the assumed distribution can be used to choose the proper distribution model. Generally, the GoF is obtained using data histogram and theoretical distribution curve, and then the distribution model is judged via GoF. However, when the sample number is small, the histogram is rough and fluctuating, affecting the analysis of GoF. For the small sample, the graphical characteristic is obtained with the sample data to choose the most fitting distribution to the data in this paper. The graphical characteristic is acquired by a simpler process, that is, the original data are directly set as the test statistics, avoiding computing and sorting of other statistics. In this paper, the real airborne circular synthetic aperture radar data under different scan angles are analyzed using the GoF corresponding to histogram and graphical GoF, respectively. The results show that when the sea-clutter data histogram is close to two distributions, a more fitting distribution model may not be obtained by traditional GoF, but can be acquired by graphical representation. In addition, the sea data with different sight angles have different match properties. It is seen that the sea data are closer to the Rayleigh distribution in side-looking mode than that in big squint-angle mode, while the Weibull distribution and K distribution show equal fitting performance to sea clutter under variant radar sight angles.

Recent advances in deep learning have driven significant success in synthetic aperture radar (SAR) automatic target recognition (ATR), particularly through convolutional neural network (CNN)-based classification algorithms. However, SAR images possess distinctive physical scattering properties owing to their unique imaging mechanism. Many deep learning algorithms rely solely on amplitude information, ignoring phase information, which may result in the loss of information in the original complex-valued SAR image and suboptimal performance. To tackle these problems, this article introduces a SAR target recognition approach based on complex-valued operations, designated as Complex-valued Residual Mish activation and Convolution Block Attention Module (CBAM) net (CRMC-Net). The CRMC-Net effectively utilizes the amplitude and phase information in complex-valued SAR data. Specifically, first, the elements of CNN, including the input and output layers, the convolution layers, the activation functions, and the pooling layers, are extended to the complex-valued domain. Secondly, in order to further enhance the representation ability of the model, multiscale information of the target is extracted through different convolution kernel sizes. Finally, the network constructs many complex-valued operation blocks to enhance the robustness of the designed network, including the complex-valued residual block (CReLBlock), complex-valued Mish (CMish) activation function, and complex-valued convolution block attention module (CCBAM). The experimental results obtained from the Moving and Stationary Target Capture and Recognition dataset (MSTAR) and OpenSARShip2.0 (OPS 2.0) dataset demonstrate that the proposed network model outperforms the traditional real-valued models. It can further reduce the classification error and enhance performance.