With the rapid development of deep learning, convolutional neural networks have achieved milestones in synthetic aperture radar (SAR) image object detection. However, object detection in SAR images is still a great challenge due to the difficulty in distinguishing targets from complex backgrounds. At the same time, most of the targets in SAR images are small and unevenly distributed, which makes it challenging to extract sufficient feature information. To solve these issues mentioned above, an efficient object detection network for SAR images based on Swin transformer and YOLOv7 is proposed in this article. First, we design a novel feature aggregation module Petty-specialized feature aggregation (PS-FPN) to enrich small targets’ semantic and spatial features while keeping the model lightweight. PS-FPN module uses the fusion of deep and shallow features by using cross-layer feature aggregation and single-branch feature aggregation to enhance the detection of small targets. Second, a novel attention mechanism strategy mix-attention is proposed to find more attention regions. Finally, we add one more prediction head to extract shallow features that effectively preserve small targets’ feature information. To verify the effectiveness of the proposed algorithm, extensive experiments are carried out on several challenging SAR image datasets. The results show that, compared with other state-of-the-art detectors, the proposed method can achieve significant performance based on lightweight detection.

Road extraction from remote sensing images is significant for urban planning, intelligent transportation, and vehicle navigation. However, it is challenging to automatically extract roads from remote sensing images because the scale difference of roads in remote sensing images varies greatly, and slender roads are difficult to identify. Moreover, the road in the image is often blocked by the shadows of trees and buildings, which results in discontinuous and incomplete extraction results. To solve the above problems, this paper proposes a multiscale feature encoding and long-range context-aware network (MECA-Net) for road extraction. MECA-Net adopts an encoder–decoder structure and contains two core modules. One is the multiscale feature encoding module, which aggregates multiscale road features to improve the recognition ability of slender roads. The other is the long-range context-aware module, which consists of the channel attention module and the strip pooling module, and is used to obtain sufficient long-range context information from the channel dimension and spatial dimension to alleviate road occlusion. Experimental results on the open DeepGlobe road dataset and Massachusetts road dataset indicate that the proposed MECA-Net outperforms the other eight mainstream networks, which verifies the effectiveness of the proposed method.

Compared with large scale ships, small scale ships occupy few pixels and have low contrast, so it poses a great challenge to detect multi-scale ships in SAR images. In order to improve the accuracy of multi-scale ship detection in SAR images, this paper designs a general multi-scale pyramid attention module (MPAM), which is a plug-and-play lightweight module that can adapt to many ship detection networks. In MPAM, a deep feature extraction sub-module (DFES) is first designed to use the multi-scale pyramid structure to divide the feature map into different levels, extracting rich features with resolution and semantic information for multi-scale ship detection. The channel multi-layer attention fusion sub-module (CMAFS) and spatial multi-layer attention fusion sub-module (SMAFS) are then designed to fuse the channel and spatial attention blocks on different level feature maps, which could better learn the dependent features from the channel and spatial dimensions, to enhance the feature representation. Finally, the fused feature map is input into the existing ship detection networks to obtain the detection result. Experiments on SAR datasets containing multi-scale ships show that the effectiveness of MPAM in improving the accuracy of the existing ship detection networks.

Remote sensing image change detection refers to the automatic acquisition of the location and category of object changes using images of different phases. The rapid development of optical remote sensing and artificial intelligence technology has made deep learning a mainstream method for remote sensing change detection. In current research, deep learning-based remote sensing change detection mainly focuses on determining whether changes have occurred at the pixel level, and this pixel-level change detection is susceptible to pseudo-changes due to factors such as differences in imaging angles of optical cameras, changes in imaging seasons, and occlusion. In addition, in some specific scenarios, such as the change analysis task for movable objects with regular geometries (e.g., airplane, vehicles, etc.) is more concerned with the change of the object as a whole than the change of the pixels inside the object, in which case pixel-level methods are no longer the best choice and object-level change detection methods should be adopted. To address the above problems, we propose spatial-frequency difference attention network for object-level change detection (SFDA-OCDet). Firstly, we use a Siamese backbone with shared weight parameters to extract the image features of different stages from the input dual-temporal images. Secondly, we design the spatial difference attention (SDA) module, which obtains the attention weights from the local context and the global context to enhance the learning ability of the difference features of different time images from the spatial domain. Then, we propose the wavelet frequency difference attention (WFDA) module, which utilizes the discrete wavelet transform and the attention mechanism to enhance the difference feature learning capability from the frequency domain. Finally, we present the wavelet balanced feature pyramid (WBFP) module, which introduces the discrete wavelet transform and its inverse transformation to implement the downsampling and upsampling process to avoid feature information loss. The change detection head outputs change detection results in the form of a bounding box. We construct four object-level change detection datasets, Airplane-CD, Vehicle-CD, Building-CD and Wind turbine-CD, to carry out experimental validation. Numerous experiments have shown that the proposed SFDA-OCDet outperforms other comparative methods, proving the validity of the method.

Infrared small target detection has long faced significant challenges due to the extremely small size of targets, low contrast, and the scarcity of annotated data. To tackle these issues, we propose a wavelet-recalibrated semi-supervised network (WRSSNet) that integrates synthetic data augmentation, feature reconstruction, and semi-supervised learning, aiming to fully exploit the potential of unlabeled infrared images under limited supervision. We construct a dataset containing 843 visible-light small target images and employ an improved CycleGAN model to convert them into high-quality pseudo-infrared images, effectively expanding the scale of training data for infrared small target detection. In addition, we design a lightweight wavelet-enhanced channel recalibration and fusion (WECRF) module, which integrates wavelet decomposition with both channel and spatial attention mechanisms. This module enables adaptive reweighting and efficient fusion of multi-scale features, highlighting high-frequency details and weak target responses. Extensive experiments on two public infrared small target datasets, NUAA-SIRST and IRSTD-1K, demonstrate that WRSSNet achieves superior detection accuracy and lower false alarm rates compared to several state-of-the-art methods, while maintaining low computational complexity.