基于阈值分割算法的湖泊边缘特征提取研究  

Research on Lake Edge Feature Extraction Based on Threshold Segmentation Algorithm

作　　者：田佳宁 张果荣 宋宝 TIAN Jianing;ZHANG Guorong;SONG Bao(School of Space Information and Surveying Engineering,Anhui University of Science and Technology,Huainan 232000,China;School of Automation,Nanjing University of Information Science and Technology,Nanjing 211800,China)

机构地区：[1]安徽理工大学空间信息与测绘工程学院,安徽淮南232000 [2]南京信息工程大学自动化学院,江苏南京211800

出　　处：《现代信息科技》2021年第4期67-71,76,共6页Modern Information Technology

摘　　要：为解决使用卫星遥感数据进行湖泊边缘特征提取时效率低,算法冗余,无针对性等问题,文章以提取大范围湖泊为基础目标,使用区域生长法、最大类间方差法、最大熵分割法、基本全局阈值分割法对中国五大湖泊边缘特征进行提取,从水体面积、最长处与最宽处距离等三个角度进行验证。结果表明,在大面积湖泊边缘特征提取中,最大类间方差法、基本全局阈值法效果最优,最大熵分割法效果最差。同时,在水体与背景颜色差异性大时区域生长算法提取效果最优。In order to solve the problems of low efficiency,algorithm redundancy and non pertinence when using satellite remote sensing data to extract lack edge features,this paper takes the extraction of a large range of lakes as the basic target,and uses region growing algorithm,maximum inter class variance algorithm,maximum entropy segmentation algorithm and basic global threshold segmentation algorithm to extract the edge features of five major lakes in China,and verifies from three angles of the area of the water body,the distance of the longest area and the distance of the widest area.The results show that in the extraction of large-area lake edge features,the maximum inter class variance algorithm and the basic global threshold algorithm are the best,and the maximum entropy segmentation algorithm is the worst.At the same time,when the color difference between water body and background is obvious,the region growing algorithm has the best extraction effect.

关 键 词：水体提取 区域生长法 最大类间方差法 基本全局阈值分割法 

分 类 号：P237[天文地球—摄影测量与遥感]