2026· IEEE Transactions on Aerospace and Electronic Systems· Vol 62, pp. 15773-15792· 0 citations· 70 references
Computer Science
Abstract
In conventional single-view strip synthetic aperture radar (SAR) images, the occlusion effect and scattering interference between adjacent targets lead to blurred edges of small targets, thereby compromising the localization accuracy of detection boxes. Moreover, complex backgrounds can obscure the scattering characteristics of targets, significantly increasing the difficulty of detecting dense small targets. In contrast, airborne wide-angle staring SAR (WasSAR) conducts long-duration, multiview observations on regions of interest, enabling rapid acquisition of multiview scattering characteristics of targets, which holds promise for addressing the challenge of detecting dense small targets. This article leverages the advantages of multiview observations from airborne WasSAR and proposes a two-stage detection framework for dense small targets. In the first stage, background suppression is performed on WasSAR images by modeling the multiview SAR images on a per-pixel basis using the Gaussian mixture model, which suppresses the background while preserving the structural integrity and saliency of small targets. In the second stage, a target detection network with multiview inputs is introduced. This network comprises a local structure enhancement shared branch and a multibranch module. Then, a feature fusion module is employed to strengthen the information interaction among multiview SAR images, thereby forming a robust feature representation for both the regression and classification of target bounding boxes. Extensive experiments conducted on two public and team-collected WasSAR datasets demonstrate that the proposed method achieves optimal performance in small target detection tasks compared to ten various existing detectors. This article also analyzes the impact of different views on detection performance, providing an effective solution for SAR small target detection.
Synthetic aperture radar (SAR) target detection is a fundamental task in intelligent SAR image interpretation and supports a wide range of remote sensing applications. However, existing methods remain limited in handling sparse and discontinuous scattering responses, substantial variations in target scale, speckle nois...
An Chang, Yu Zhang, Chang Liu· Remote Sensing· 0 citations
Robust detection of infrared dim small targets is critical in long-range detection and precision guidance systems. Targets in complex backgrounds usually occupy only a few pixels with extremely low contrast, and are often buried in structured clutter such as clouds, sea-sky lines and strong noise, rendering single-fram...
Bo-Wen Zhang, Jing Gao, Yi-Lin Liu et al.· International Conference on...· 0 citations
Synthetic aperture radar (SAR) ship detection is essential for maritime surveillance, yet coherent speckle can obscure tiny offshore vessels, while high-response coastal structures can produce target-like responses. Repeated downsampling further weakens sparse vessel signatures, whereas a separate high-resolution predi...
Infrared small target detection is important for remote sensing surveillance, precision guidance, and early warning, but reliable detection in complex scenes remains difficult because small targets usually occupy only a few pixels and exhibit weak contrast and low signal-to-noise ratio. Complex background clutter, incl...
Yi-Wen Wang, Wei Zheng, Yi-Ming Zhao et al.· Global Intelligent Industry...· 0 citations
Aircraft detection in synthetic aperture radar (SAR) images is essential for airport surveillance, airspace security, and military reconnaissance. However, the coherent imaging mechanism of SAR, varying observation angles, and complex airport backgrounds make aircraft detection challenging. Aircraft targets may suffer...
Jian Zhuge, Jia-Bin Hu, Ren-Xiu Liao et al.· IEEE Geoscience and Remote S...· 0 citations
The results show that edge texture enhancement cannot be directly equated with detection benefits; enhancement evaluation should focus on full-image fidelity, target area indicators, and authentic detection feedback.
Run-Xing Song· Academic Journal of Science...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.