SAR Image Detection Method for Offshore Wind Turbines Considering Correlation and Spatial Distribution Characteristics
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 rec-tangle 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 tur-bine 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.
万方数据
维普
