With the development of artificial intelligence (AI) and synthetic aperture radar (SAR) technology, SAR ship target automatic detection has made significant progress. However, due to the complex structure and various sizes of ship, detecting multi-scale ships under complex backgrounds is still challenging. The existing convolution neural networks (CNNs)-based SAR ships detection methods suffer from insufficient feature extraction capabilities and poor feature reusability. To address the issues, an SAR ship detection method based on multi-feature fusion and non-local channel attention mechanism (MFNL-Net) is proposed in this article. MFNL-Net is a one-stage detector and designed to improve the performance of detecting multi-scale ships by enhancing the feature extraction ability and nonlinear relationship between different features. Specially, the non-local channel attention block (NLCA-Block) is employed to enhance the nonlinear relationship between different channel features and the ability of local feature expression; the pooling feature fusion (PFF) module and the deconvolution feature fusion (DFF) module are reasonably combined to obtain abundant ships target feature information and suppress interference caused by surroundings; the multi-scale spatial pyramid pooling (MSPP) is used to boost feature information interaction and reduce the loss of significant feature; the focus classification loss is used to alleviate influences of the positive-negative samples imbalance. To evaluate the effectiveness of MFNL-Net, experiments are conducted on real SAR ship datasets SSDD and GS-Ship. Experimental results show that our method obtains 9.10%, 8.61%, 7.23%, and 4.44% gains in mean average precision (mAP) compared with the four state-of-the-art CNNs-based methods, respectively.

Deep learning (DL) models have recently been widely used in UAV aerial image semantic segmentation tasks and have achieved excellent performance. However, DL models are vulnerable to adversarial examples, which bring significant security risks to safety-critical systems. Existing research mainly focuses on solving digital attacks for aerial image semantic segmentation, but adversarial patches with physical attack attributes are more threatening than digital attacks. In this article, we systematically evaluate the threat of adversarial patches on the aerial image semantic segmentation task for the first time. To defend against adversarial patch attacks and obtain accurate semantic segmentation results, we construct a novel robust feature extraction network (RFENet). Based on the characteristics of aerial images and adversarial patches, RFENet designs a limited receptive field mechanism (LRFM), a spatial semantic enhancement module (SSEM), a boundary feature perception module (BFPM) and a global correlation encoder module (GCEM), respectively, to solve adversarial patch attacks from the DL model architecture design level. We discover that semantic features, shape features and global features contained in aerial images can significantly enhance the robustness of the DL model against patch attacks. Extensive experiments on three aerial image benchmark datasets demonstrate that the proposed RFENet has strong resistance to adversarial patch attacks compared with the existing state-of-the-art methods.

Aerial Image Semantic segmentation based on convolution neural networks (CNNs) has made significant process in recent years. Nevertheless, their vulnerability to adversarial example attacks could not be neglected. Existing studies typically focus on adversarial attacks for image classification, ignoring the negative effect of adversarial examples on semantic segmentation. In this article, we systematically assess and verify the influence of adversarial attacks on aerial image semantic segmentation. Meanwhile, based on the robust characteristics of global features, we construct a novel global feature attention network (GFANet) for aerial image semantic segmentation to solve the threat of adversarial attacks. GFANet uses the global context encoder (GCE) to obtain the context dependencies of global features, introduces the global coordinate attention mechanism (GCAM) to enhance the global feature representation to suppress adversarial noise, and the feature consistency alignment (FCA) is used for feature calibration. In addition, we construct a universal adversarial training strategy to improve the robustness of the semantic segmentation model against adversarial example attacks. Extensive experiments on three aerial image datasets demonstrate that GFANet is more robust against adversarial attacks than existing state-of-the-art semantic segmentation models.

Building extraction from high-resolution remote sensing images is of great importance for urban planning, disaster assessment, and geography mapping. In recent years, convolutional neural networks have made outstanding achievements in improving the precision of building extraction. However, most existing approaches have some problems, such as insufficient detailed feature extraction and ignorance of the relationship between different features. In this study, we propose a novel multi-channel recurrent attention network (MCANet) for building extraction. Firstly, the multi-scale channel attention mechanism is used to expand the convolution kernel receptive field, making the model can extract rich building region feature information. Secondly, we use the spatial pyramid recurrent block to establish long-range dependencies over space, channel, and layer of different convolutions. Finally, the multi-channel feature fusion block is used to fuse the multi-scale channel features information, and improve the building extraction precision. Experimental results show that the proposed MCANet achieves better results (recall, precision, intersection-over-union, and F ₁ _score on the Inria Aerial Image Labeling Dataset are 89.82%, 94.38%, 87.42%, and 88.25%, respectively), and outperforms the other state-of-the-art approaches.

The outstanding performance of deep neural networks (DNNs) in multiple computer vision in recent years has promoted its widespread use in aerial image semantic segmentation. Nonetheless, prior research has demonstrated the high susceptibility of DNNs to adversarial attacks. This poses significant security risks when applying DNNs to safety-critical earth observation missions. As an essential means of attacking DNNs, data poisoning attacks destroy model performance by contaminating model training data, allowing attackers to control prediction results by carefully crafting poisoning samples. Toward building a more robust DNNs-based aerial image semantic segmentation model, in this study, we proposed a robust invariant feature enhancement network (RIFENet) that can resist data poisoning attacks and has superior semantic segmentation performance. The constructed RIFENet improves the resistance to poisoning attacks by extracting and enhancing robust invariant features. Specifically, RIFENet uses a texture feature enhancement module (T-FEM), structural feature enhancement module (S-FEM), global feature enhancement module (G-FEM), and multi-resolution feature fusion module (MR-FFM) to enhance the representation of different robust features in the feature extraction process to suppress the interference of poisoning samples. Experiments on several benchmark aerial image datasets demonstrate that the proposed method is more robust and exhibits better generalization than other state-of-the-art methods.

Profiting from the powerful feature extraction and representation capabilities of deep learning (DL), aerial image semantic segmentation based on deep neural networks (DNNs) has achieved remarkable success in recent years. Nevertheless, the security and robustness of DNNs deserve attention when dealing with safety-critical earth observation tasks. As a typical attack pattern in adversarial machine learning (AML), backdoor attacks intend to embed hidden triggers in DNNs by poisoning training data. The attacked DNNs behave normally on benign samples, but when the hidden trigger is activated, its prediction is modified to a specified target label. In this article, we systematically assess the threat of backdoor attacks to aerial image semantic segmentation tasks. To defend against backdoor attacks and maintain better semantic segmentation accuracy, we construct a novel robust generative adversarial network (RFGAN). Motivated by the sensitivity of human visual systems to global and edge information in images, RFGAN designs the robust global feature extractor (RobGF) and the robust edge feature extractor (RobEF) that force DNNs to learn global and edge features. Then, RFGAN uses robust global and edge features as guidance to obtain benign samples by the constructed generator, and the discriminator to obtain semantic segmentation results. Our method is the first attempt to address the backdoor threat to aerial image semantic segmentation by constructing the robust DNNs model architecture. Extensive experiments on real-world scenes aerial image benchmark datasets demonstrate that the constructed RFGAN can effectively defend against backdoor attacks and achieve better semantic segmentation results compared with the existing state-of-the-art methods.