测绘通报2024,Issue(12) :18-23.DOI:10.13474/j.cnki.11-2246.2024.1204

改进YOLOv8的SAR影像船舰目标检测模型

Improving the SAR image ship target detection model of YOLOv8

杨明秋 陈国坤 左小清 董燕
测绘通报2024,Issue(12) :18-23.DOI:10.13474/j.cnki.11-2246.2024.1204

改进YOLOv8的SAR影像船舰目标检测模型

Improving the SAR image ship target detection model of YOLOv8

杨明秋 1陈国坤 1左小清 1董燕1
扫码查看

作者信息

  • 1. 昆明理工大学国土资源工程学院,云南 昆明 650093
  • 折叠

摘要

在SAR影像船舰目标检测任务中,受近海岸区域背景复杂和船舰目标多尺度等因素影响,船舰目标在检测过程中出现检测精度不高、漏检的问题.针对上述问题,本文提出了一种基于YOLOv8s改进的SAR影像船舰目标检测模型,并在SSDD和HRSID数据集上进行试验验证,效果优于其他经典算法.

Abstract

In the SAR image ship target detection task,due to factors such as complex coastal background and multi-scale ship targets,ship targets may have low detection accuracy and missed detections during the detection process.In response to the above issues,this article proposes an improved SAR image ship target detection model based on YOLOv8s,and conducts experimental verification by SSDD and HRSID datasets,with better performance than other classical algorithms.

关键词

船舰目标检测/SAR影像/残差增强/可变形卷积/动态稀疏注意力

Key words

ship target detection/SAR imaging/residual enhancement/deformable convolution/dynamic sparse attention

引用本文复制引用

出版年

2024
测绘通报
测绘出版社

测绘通报

CSTPCDCSCD北大核心
影响因子:1.027
ISSN:0494-0911
段落导航相关论文