顾及相干性与空间分布特征的SAR图像海上风电平台检测方法  

SAR Image Detection Method for Offshore Wind Turbines Considering Correlation and Spatial Distribution Characteristics

作　　者：王佳俊 靳国旺[1] 熊新[1,3] 李佳豪 叶豪 秦永志 WANG Jiajun;JIN Guowang;XIONG Xin;LI Jiahao;YE Hao;QIN Yongzhi(Information Engineering University,Zhengzhou 450001,China;92556 Troops,Ningbo 315000,China;Key Laboratory of Smart Earth,Beijing 100094,China;61206 Troops,Beijing 100042,China)

机构地区：[1]信息工程大学,河南郑州450001 [2]92556部队,浙江宁波315000 [3]智慧地球重点实验室,北京100094 [4]61206部队,北京100042

出　　处：《测绘科学技术学报》2024年第5期484-490,共7页Journal of Geomatics Science and Technology

基　　金：国家自然科学基金项目(42201492)。

摘　　要：针对现有风电平台检测方法依赖长时间序列影像的问题,设计了一种顾及相干性与空间分布特征的SAR图像海上风电平台检测方法。该方法充分考虑风电平台呈规则矩阵型、梅花型的空间分布特征与高相干性,利用自适应参数的DBSCAN算法对相干系数图中高相干像素进行聚类,通过将聚类结果的最小外接矩形作为风电场边界来排除船舶干扰,再采用NCC算法将风电平台模板与风电场SAR正射影像进行匹配,实现风电平台的检测。采用覆盖中国南海、欧洲北海两个试验区的Sentinel-1A影像进行了实验,验证了本文方法的有效性。In view of the problem that existing wind turbine detection methods rely on long-term sequence images,a SAR image offshore wind turbine detection method considering coherence and spatial distribution characteristics is designed in this paper.In this method,the spatial distribution characteristics of wind turbines in a regular matrix and plum blossom shape,as well as high coherence,are fully considered.The DBSCAN algorithm with adaptive parameters is used to cluster high coherence pixels in the coherence coefficient map.The minimum bounding rectangle of the clustering results is used as the boundary of the wind farm to eliminate ship interference.Then,the NCC algorithm is used to match the wind turbine template with the wind farm SAR orthoimage to achieve wind turbine detection.Experiments were conducted using Sentinel-1A images covering the South China Sea and the North Sea of Europe,verifying the effectiveness of the proposed method.

关 键 词：合成孔径雷达 海上风电平台 目标检测 相干系数 影像匹配 

分 类 号：P237[天文地球—摄影测量与遥感]