MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SAR

Existing convolution neural network (CNN)-based video synthetic aperture radar (SAR) moving target shadow detectors are difficult to model long-range dependencies, while transformer-based ones often suffer from greater complexity. To handle these issues, this paper proposes MambaShadowDet, a novel l...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaowo Xu, Tianwen Zhang, Xiaoling Zhang, Wensi Zhang, Xiao Ke, Tianjiao Zeng
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/2/214
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832587566157660160
author Xiaowo Xu
Tianwen Zhang
Xiaoling Zhang
Wensi Zhang
Xiao Ke
Tianjiao Zeng
author_facet Xiaowo Xu
Tianwen Zhang
Xiaoling Zhang
Wensi Zhang
Xiao Ke
Tianjiao Zeng
author_sort Xiaowo Xu
collection DOAJ
description Existing convolution neural network (CNN)-based video synthetic aperture radar (SAR) moving target shadow detectors are difficult to model long-range dependencies, while transformer-based ones often suffer from greater complexity. To handle these issues, this paper proposes MambaShadowDet, a novel lightweight deep learning (DL) detector based on a state space model (SSM), dedicated to high-speed and high-accuracy moving target shadow detection in video SAR images. By introducing SSM with the linear complexity into YOLOv8, MambaShadowDet effectively captures the global feature dependencies while relieving computational load. Specifically, it designs Mamba-Backbone, combining SSM and CNN to effectively extract both global contextual and local spatial information, as well as a slim path aggregation feature pyramid network (Slim-PAFPN) to enhance multi-level feature extraction and further reduce complexity. Abundant experiments on the Sandia National Laboratories (SNL) video SAR data show that MambaShadowDet achieves superior moving target shadow detection performance with a detection accuracy of 80.32% F1 score and an inference speed of 44.44 frames per second (FPS), outperforming existing models in both accuracy and speed.
format Article
id doaj-art-4b9e58a038344e438b7a06565dfbb89c
institution Kabale University
issn 2072-4292
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-4b9e58a038344e438b7a06565dfbb89c2025-01-24T13:47:45ZengMDPI AGRemote Sensing2072-42922025-01-0117221410.3390/rs17020214MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SARXiaowo Xu0Tianwen Zhang1Xiaoling Zhang2Wensi Zhang3Xiao Ke4Tianjiao Zeng5School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Aeronautics and Astronautics, University of Electronic Science and Technology of China, Chengdu 611731, ChinaExisting convolution neural network (CNN)-based video synthetic aperture radar (SAR) moving target shadow detectors are difficult to model long-range dependencies, while transformer-based ones often suffer from greater complexity. To handle these issues, this paper proposes MambaShadowDet, a novel lightweight deep learning (DL) detector based on a state space model (SSM), dedicated to high-speed and high-accuracy moving target shadow detection in video SAR images. By introducing SSM with the linear complexity into YOLOv8, MambaShadowDet effectively captures the global feature dependencies while relieving computational load. Specifically, it designs Mamba-Backbone, combining SSM and CNN to effectively extract both global contextual and local spatial information, as well as a slim path aggregation feature pyramid network (Slim-PAFPN) to enhance multi-level feature extraction and further reduce complexity. Abundant experiments on the Sandia National Laboratories (SNL) video SAR data show that MambaShadowDet achieves superior moving target shadow detection performance with a detection accuracy of 80.32% F1 score and an inference speed of 44.44 frames per second (FPS), outperforming existing models in both accuracy and speed.https://www.mdpi.com/2072-4292/17/2/214video synthetic aperture radar (SAR)moving target shadow detectionstate space model (SSM)
spellingShingle Xiaowo Xu
Tianwen Zhang
Xiaoling Zhang
Wensi Zhang
Xiao Ke
Tianjiao Zeng
MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SAR
Remote Sensing
video synthetic aperture radar (SAR)
moving target shadow detection
state space model (SSM)
title MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SAR
title_full MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SAR
title_fullStr MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SAR
title_full_unstemmed MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SAR
title_short MambaShadowDet: A High-Speed and High-Accuracy Moving Target Shadow Detection Network for Video SAR
title_sort mambashadowdet a high speed and high accuracy moving target shadow detection network for video sar
topic video synthetic aperture radar (SAR)
moving target shadow detection
state space model (SSM)
url https://www.mdpi.com/2072-4292/17/2/214
work_keys_str_mv AT xiaowoxu mambashadowdetahighspeedandhighaccuracymovingtargetshadowdetectionnetworkforvideosar
AT tianwenzhang mambashadowdetahighspeedandhighaccuracymovingtargetshadowdetectionnetworkforvideosar
AT xiaolingzhang mambashadowdetahighspeedandhighaccuracymovingtargetshadowdetectionnetworkforvideosar
AT wensizhang mambashadowdetahighspeedandhighaccuracymovingtargetshadowdetectionnetworkforvideosar
AT xiaoke mambashadowdetahighspeedandhighaccuracymovingtargetshadowdetectionnetworkforvideosar
AT tianjiaozeng mambashadowdetahighspeedandhighaccuracymovingtargetshadowdetectionnetworkforvideosar