基于SAR图像的海洋中尺度涡自动检测算法  

Automatic Detection Algorithm of Ocean Mesoscale Eddies Based on SAR Images

在线阅读下载全文

作  者:蔡泽坤 陈永强 CAI Zekun;CHEN Yongqiang(Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China)

机构地区:[1]中国科学院空天信息创新研究院,北京100094

出  处:《自动化应用》2024年第16期201-206,211,共7页Automation Application

摘  要:海洋中尺度涡是一种常见且重要的海洋现象,目前采用合成孔径雷达技术(对海面高分辨率成像)为观测和研究海洋中的尺度涡提供了条件。由于中尺度涡形态的复杂性和多样性导致中尺度涡的检测十分困难,为解决中尺度涡的高效检测的问题,提出了一种海洋中尺度涡自动检测算法,利用“巡线提取”提取和聚类高分辨率海面图像的纹理,使图像中的纹理特征转化为描述海洋流场方向的线段集合,然后利用线段的几何特征结合漩涡的规律计算中尺度涡特征,再根据阈值筛选实现中尺度涡涡心的检测。该算法具有极强的灵活性和丰富的功能,能帮助研究人员获得足够的中尺度涡研究样本,具有较高的应用价值。选用GF-3号和哨兵1号卫星图像进行实验,结果表明,所提检测算法能准确地检测中尺度涡并进行可视化呈现。Mesoscale eddies in the ocean are a common and important marine phenomenon,and currently synthetic aperture radar technology(high-resolution imaging of the sea surface)is used to provide conditions for observing and studying scale eddies in the ocean.Due to the complexity and diversity of mesoscale vortex morphology,it is very difficult to detect mesoscale eddies.To solve the problem of efficient detection of mesoscale eddies,an automatic detection algorithm for mesoscale eddies in the ocean is proposed.The algorithm uses"patrol line extraction"to extract and cluster the texture of high-resolution sea surface images,transforming the texture features in the images into a set of line segments that describe the direction of the ocean flow field.Then,the geometric features of the line segments are combined with the laws of eddies to calculate the mesoscale vortex features,and the detection of mesoscale vortex centers is achieved through threshold screening.This algorithm has strong flexibility and rich functionality,which can help researchers obtain sufficient samples for studying mesoscale eddies and has high application value.GF-3 and Sentinel-1 satellite images were selected for the experiment.The results indicate that the proposed detection algorithm can accurately detect mesoscale eddies and visualize them.

关 键 词:SAR海洋遥感 中尺度涡 自动检测 

分 类 号:TP274.5[自动化与计算机技术—检测技术与自动化装置]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象