Youming Wu

Synthetic Aperture Radar (SAR) aircraft detection and recognition play a crucial role in tasks such as flight scheduling and analysis of airport traffic. However, SAR images suffer from inherent sidelobe interference and speckle noise. These factors degrade the integrity of aircraft scattering features, thereby limiting the accuracy of fine-grained detection and recognition. To address these challenges, a discriminative feature enhancement network (DFE-Net) is proposed. The key lies in constructing a discriminative feature from SAR complex-valued images, where scattering invariants and characteristic variables are effectively decoupled and enhanced. A discriminative channel optimization (DCO) module is proposed to obtain a tri-channel representation for sidelobe and speckle suppression as well as target feature enhancement. DCO learns tri-channel interactions and adaptively fuses shared and specific cues. This mechanism expands inter-class separability and provides a robust feature space for fine-grained tasks. Furthermore, a contour-prior guidance (CPG) module fuses contour priors with deep features. This highlighting of critical contours enhances geometric consistency while reducing false alarms. By delivering stable and semantically enriched features, DFE-Net significantly improves detection and recognition performance under limited image quality. Extensive experiments on the SAR-CARD dataset from Gaofen-3 demonstrate that DFE-Net consistently outperforms existing baselines in terms of both accuracy and robustness.

Synthetic Aperture Radar (SAR) aircraft detection plays a crucial role in various civilian applications. Benefiting from the powerful capacity of feature extraction and analysis of deep learning, aircraft detection performance has been improved by most traditional general-purpose visual intelligence methods. However, the inherent imaging mechanisms of SAR fundamentally differ from optical images, which poses challenges for SAR aircraft detection. Aircraft targets in SAR imagery typically exhibit indistinct details, discrete features, and weak contextual associations and are prone to non-target interference, which makes it difficult for existing visual detectors to capture critical features of aircraft, limiting further optimization of their performance. To address these issues, we propose the scattering feature guidance network (SFG-Net), which integrates feature extraction, global feature fusion, and label assignment with essential scattering distribution of targets. This enables the network to focus on critical discriminative features and leverage robust scattering features as guidance to enhance detection accuracy while suppressing interference. The core components of the proposed method include the detail feature supplement (DFS) module and the context-aware scattering feature enhancement (CAFE) module. The former integrates low-level texture and contour features to mitigate detail ambiguity and noise interference, while the latter leverages global context of strong scattering information to generate more discriminative feature representations, guiding the network to focus on critical scattering regions and improving learning of essential features. Additionally, a feature scattering center-based label assignment (FLA) strategy is introduced, which utilizes the spatial distribution of scattering information to adaptively adjust the sample coverage and ensure that strong scattering regions are prioritized during training. A series of experiments was conducted on the CSAR-AC dataset to validate the effectiveness and generalizability of the proposed method.

In the remote sensing field, resolution-degraded (RD) images are often captured under poor imaging conditions, leading to missing texture information for segmentation. To tackle these challenges, a parallel super-resolution branch can be incorporated into the segmentation pipeline to restore detailed features through a feature supplementation operation. However, existing methods based on the structure are mainly limited by the direct supplementation operation between the two tasks with different feature distributions, and the original semantic distribution may be damaged. Therefore, a novel learning method called Generalized Feature Affinity (GFA) is proposed. It is expected to realize more comprehensive feature supplementation in a consistent feature distribution space between the two tasks. Specifically, a Global Distribution Affinity (GDA) module is developed to encourage uniformity in the intra-class distribution between restored and semantic features. Then a Local Texture Affinity (LTA) module is designed to better transfer detailed information by enhancing the texture prominence in feature maps and exploring supplementation in spatial and channel aspects. To guide the supplementation direction more effectively, a Results-oriented Hard Sampling (ROHS) strategy which integrates auxiliary spatial attention is further proposed to enhance the performance of the aforementioned modules. Extensive experiments are conducted on the widely recognized benchmarks ISPRS Potsdam and Vaihingen, and LoveDA datasets. The proposed method results in an improvement of about 2% mIoU across many commonly used segmentation models, demonstrating the effectiveness and generalization of GFA.

Aircraft detection in synthetic aperture radar (SAR) images plays a crucial role in supporting essential tasks like airport management and airspace monitoring. Most existing SAR aircraft detection algorithms are predominantly designed based on the scattering characteristics of full-aperture images, which provide high-resolution and rich detail information. However, in full-aperture images, responses from different angles are aliased, making it challenging to disentangle angular-specific details for anisotropic aircraft. Furthermore, the extended synthetic aperture time exacerbates the defocusing of slow-moving aircraft, increasing the likelihood of missed detections. Considering the above challenges, for targets exhibiting anisotropy and dispersion under high resolution, their non-stationary characteristics can be extracted by decomposing sub-apertures and employed as additional information to enhance detection accuracy. Therefore, we propose a dual-aperture feature fusion network (DAF-Net) designed to improve aircraft detection by mining, enhancing, and fusing features from both full-aperture and sub-aperture images.Specifically, a siamese backbone network with position-consistent weight supervision is introduced to efficiently and accurately extract features from the parallel input of full-aperture and sub-aperture images.Subsequently, an aperture-based feature enhancement module (AFEM) is proposed to leverage the scattering characteristics of aircraft and their context, thereby reinforcing the unique semantic information present in both full-aperture and sub-aperture images. Finally, a region-aware cross-attention fusion module (RCFM) is developed, which adaptively associates and fuses full-aperture and sub-aperture features based on spatial mapping relationships. Furthermore, experiments conducted on Complex-valued SAR Aircraft dataset (CSAR-A) demonstrate the effectiveness of the proposed method.DAF-Net improves AP50 and AP75 by 5.8% and 5.3% compared to the baseline, respectively, and achieves optimal performance compared to other detectors.

Aircraft detection in synthetic aperture radar (SAR) imagery is significant due to its critical role in various applications, including surveillance, reconnaissance, and security. However, given the background interference and discreteness of aircraft scattering, detectors are prone to acquire unremarkable aircraft features. These factors lead to false alarms and present difficulties in locating aircraft accurately. This article proposes an innovative scattering topology cue-based network (STC-Net), which enhances aircraft discriminability and more accurately evaluates the quality of the prediction results. We model the aircraft with the star topology (ST), which not only emphasizes critical components like the nose and wings but also explicitly links them as a cohesive unit. Based on the cue of ST, the star topology space fusion module (ST-SFM) and the star topology channel attention module (ST-CAM) are designed. The former integrates discrete components to reestablish the aircraft features based on neighboring information of ST, while the latter suppresses background interference to highlight the aircraft by exploiting node information of ST. In addition, completeness and consistency loss (CCLoss) function that includes the completeness-aware label and the positive sample weighting function is introduced. The completeness-aware label describe the localization accuracy by incorporating the degree of overlap of predicted results on ST, while the positive sample weighting function enhances the consistency of the classification and localization branches. Furthermore, experiments conducted on the Gaofen-3 SAR Aircraft Detection Dataset (GF3ADD) and the publicly available SAR-AIRcraft-1.0 dataset demonstrate the effectiveness and generalizability of STC-Net, with our method achieving state-of-the-art performance.

Object detection and recognition (OD&R) based on deep learning is a hot topic in the application of synthetic aperture radar (SAR). These methodologies based on deep learning are inherently data-driven, which means that their performance is subjected to the corresponding datasets. Although existing datasets have included some common targets collected from real-valued intensity SAR images, they still exist some limitations in terms of quantity, categories, diversities, data domain. Hence, it is urgent to establish a large-quantity benchmark for fine-grained OD&R on complex-valued SAR images, which contains rich signal-domain features well coupled with classical physical modeling. Additionally, considering the unique imaging characteristics and diverse imaging conditions, some important attribute information, such as incidence and attitude angles, is necessary to be attached. In this paper, we propose a novel benchmark dataset with more than 340k instances for Fine-grAIned object detection and Recognition based on single look Complex SAR images, which is named as FAIR-CSAR. We collected complex-valued SAR images with a resolution of 1m to 5m from 175 entire images of Gaofen-3 covering 32 cities and multiple sea areas worldwide. All instances in the FAIR-CSAR are annotated by oriented bounding boxes (OBB), covering 5 major categories and 22 subcategories. Compared with existing datasets dedicated to OD&R, the FAIR-CSAR dataset has four particular advantages: (1) it contains complex-valued SAR images from various acquisition modes and polarization modes, including full-scale signal-domain features for object recognition; (2) it is much larger than other existing OD&R datasets in terms of quantity of instances; (3) it provides more fine-grained category annotation, and more detailed attribute information; (4) it provides more challenging images with some common imaging phenomena, such as speckle noise, azimuth ambiguities and so on. To establish a baseline adapted for single look complex SAR images, a multi-domain feature extraction and fusion network (MDNet) is proposed as a novel framework to mine detailed information underlying in various domains. And a series of state-of-the-art (SOTA) algorithms are applied on the FAIR-CSAR to build the fine-grained OD&R benchmark. Experimental results indicate that FAIR-CSAR is closer to practical application and more challenging than existing datasets for SAR images.

Sidelobes commonly disturb SAR image understanding and interpretation. Traditional spatially variant filtering algorithms achieve superior tradeoff between sidelobe suppression and resolution preservation, by means of adaptively calculate filtering parameters under some specific restrictions, such as filter design restriction and minimum amplitude constraint (MAC). These restriction aims to obtain an efficient analytical solution of filters that is easy to calculate under unsupervised conditions. However, the restriction scope is so narrow that the suppression performance achieved by these filters is limited. Also, since the unsupervised optimization based on MAC indiscriminately minimizes amplitude, the mainlobe loss is unavoidable. To further improve the performance, a spatially variant convolution neural network (SVNN) is proposed, which consists of two core modules. One is Spatially Variant Filter Generation (SVFG) Module, adaptively generating superior spatially variant filters under more relaxed restrictions. The other is Paralleled Shifted Convolution (PSC) Module, converting the signal format to achieve fast and parallel spatially variant filtering process. Benefiting from more relaxed filter restrictions, the novel network successfully achieves better performance on sidelobe suppression. Additionally, with supervised optimization based on another more accurate restriction, namely minimum error constraint (MEC), the proposed algorithm also achieves superior mainlobe maintenance. All of them are validated by comparative experiments based on satellite data from GaoFen-3 and TerraSAR-X, and our proposed method achieve state-of-the-art performance. The entire project is available at https://github.com/suoyuxi/SVNN.

Synthetic aperture radar (SAR) possesses significant advantages in aircraft detection due to its all-day and all-weather monitoring capability, but some unique problems in SAR images decrease the performance of aircraft detection. The speckle effect and excessive dynamic range are the most common problems that interfere with the visual features in SAR images and deteriorate detection performance. However, there lacks a detection-oriented image enhancement algorithm to collaboratively solve these two problems. An adaptive image enhancement algorithm is proposed to the improve the performance of aircraft detection in SAR images. The proposed image enhancement algorithm provides a pseudo-color image through speckle suppression and channel combination, which consists of the speckle noise suppression channel, strong scattering feature enhancement channel, and weak scattering feature enhancement channel. The speckle noise suppression is achieved by a despeckle network and the radiational feature enhancement channels are derived from an adaptive quantization method based on the characteristics of amplitude distribution. By optimizing the quality of the input image, the proposed image enhancement algorithm improves the performance of aircraft detection. Experiments based on datasets acquired by GaoFen-3 satellites indicate that the proposed algorithms significantly improve the detection performance of various types of detectors. The source project is available at https://github.com/suoyuxi/ChannelEnhancement.