清晨好,您是今天最早来到科研通的研友!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您科研之路漫漫前行!

YOLO-SR: An optimized convolutional architecture for robust ship detection in SAR Imagery

遥感 计算机科学 建筑 人工智能 计算机视觉 环境科学 地质学 地理 考古
作者
Chi Kien Ha,Hoanh Nguyen,В. Ван
出处
期刊:Intelligent systems with applications [Elsevier BV]
卷期号:26: 200538-200538 被引量:7
标识
DOI:10.1016/j.iswa.2025.200538
摘要

• A novel one-stage SAR ship detection model, YOLO-SR, is designed to handle speckle noise and large-scale variations while maintaining computational efficiency. • A Balanced Detail Fusion module enhances the integration of fine-grained shallow features with deeper semantic features, improving the detection of small and low-contrast ships. • The C2f-MSDR backbone incorporates Multi-Scale Dilation Residual blocks to expand the receptive field, effectively capturing ships of varying sizes in complex SAR environments. • A data-driven upsampling strategy preserves spatial fidelity, mitigating boundary artifacts and enhancing vessel contour resolution. • A specialized bounding-box regression loss function integrates orientation, distance, and scale factors with a focal mechanism for better detection of occluded ships. • Comprehensive experiments on SAR ship detection dataset demonstrate YOLO-SR’s superiority over state-of-the-art methods, particularly in detecting small and cluttered targets. Accurate and efficient ship detection in synthetic aperture radar (SAR) imagery remains a challenging task due to speckle noise, scale variations, and the low contrast of small vessels. In this work, we present YOLO-SR, an enhanced version of YOLOv10 tailored for SAR ship detection, introducing four key innovations: Balanced Detail Fusion (BDF), C2f‐MSDR, DySample, and the Focaler-SIoU loss. Our BDF module adaptively merges shallow, fine‐grained features with deeper semantic features, preventing subtle ship signatures from being overshadowed by irrelevant clutter. Concurrently, C2f‐MSDR replaces standard bottleneck layers with multi-scale dilation residual blocks, expanding the receptive field to handle wide variations in ship size. To improve spatial resolution and retain boundary details, we incorporate DySample, a data-driven upsampling strategy that counteracts the artifacts of naive interpolation. Finally, Focaler-SIoU refines bounding-box regression by integrating distance, orientation, shape, and a focal-like reweighting, thereby emphasizing difficult, small, or partially occluded ships. Experimental results on SAR ship detection datasets confirm that YOLO-SR outperforms state-of-the-art methods in both precision and recall, while retaining competitive inference speeds. These advances offer a robust framework for real-time maritime surveillance, enhancing the detection of both small and large ships under challenging SAR conditions.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
龙弟弟完成签到 ,获得积分10
5秒前
10秒前
体贴的小霜完成签到,获得积分10
10秒前
魔幻雅柔完成签到,获得积分10
16秒前
Zephyr完成签到 ,获得积分10
17秒前
19秒前
清新的方盒完成签到 ,获得积分10
21秒前
24秒前
Jannatul完成签到,获得积分10
26秒前
29秒前
pericles发布了新的文献求助10
30秒前
39秒前
48秒前
57秒前
ubless607的应助被科研通管家采纳,获得10
59秒前
怕黑的驳完成签到,获得积分10
59秒前
老猫322完成签到 ,获得积分10
1分钟前
1分钟前
英俊的傲珊完成签到,获得积分10
1分钟前
1分钟前
小橙子完成签到 ,获得积分10
1分钟前
qaplay完成签到 ,获得积分0
1分钟前
1分钟前
白嫖论文完成签到 ,获得积分10
1分钟前
1分钟前
CHME完成签到 ,获得积分10
1分钟前
Z小姐完成签到 ,获得积分10
1分钟前
jiayouliyi完成签到 ,获得积分10
1分钟前
1分钟前
玛卡巴卡爱吃饭完成签到 ,获得积分0
1分钟前
1分钟前
炙热迎天完成签到,获得积分10
1分钟前
SMG完成签到 ,获得积分10
1分钟前
qqq完成签到 ,获得积分0
1分钟前
zy完成签到 ,获得积分10
1分钟前
2分钟前
大意的之柔完成签到,获得积分10
2分钟前
2分钟前
2分钟前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8185496
求助须知:如何正确求助?哪些是违规求助? 9574168
关于积分的说明 21834498
捐赠科研通 7783092
什么是DOI,文献DOI怎么找? 3434577
关于科研通互助平台的介绍 2558949
邀请新用户注册赠送积分活动 2481346