遥感影像提取图斑中狭长结构的探测与融解方法  

Detection and dissolution methods for narrow structures in patches extracted from remote sensing images

作　　者：孔博 陈文媛 杨敏[1] 艾廷华[1] KONG Bo;CHEN Wenyuan;YANG Min;AI Tinghua(School of Resource and Environmental Sciences, Wuhan University, Wuhan Hubei 430079, China)

机构地区：[1]武汉大学资源与环境科学学院,湖北武汉430079

出　　处：《北京测绘》2022年第3期223-227,共5页Beijing Surveying and Mapping

基　　金：国家自然科学基金(42071450)。

摘　　要：遥感影像提取的地物图斑数据需要进行结构优化,从而满足数据建库与制图表达的质量要求。其中,地物图斑分布上存在的大量狭长结构需要进行一致性融解操作。该项工作目前仍由人机交互方式完成,亟须发展专门的自动探测与一致性融解模型。为解决该问题,本文首先对遥感影像提取图斑中狭长结构产生的缘由和表现形式进行深入分析。在此基础上,分别针对控制图层图斑边界产生的狭长结构和人工地物图斑的局部狭长结构设计专门的探测与一致性融解方法,从而丰富现有的图斑狭长结构处理技术体系。试验表明本文提出的两种算法对狭长结构探测与一致性融解的正确率均超过90%,具有较好的可操作性与实用价值。In order to meet the quality requirements of database construction and cartography,patches extracted from remote sensing images need to be structurally optimized.Particularly,a large number of narrow structures of the patches should be dissolved.This work is still labor-intensive and time-consuming;hence,it is urgent to develop automatic detection and dissolution method.To overcome this drawback,this paper firstly made an in-depth analysis of the causes and manifestations of narrow structures.Then,two methods were proposed to process the narrow structures around the boundaries of patches in the control layer and the local narrow structures of the man-made features,respectively.The experimental results showed that the accuracies of the two methods were higher than 90%for detecting and dissolving the narrow structures,and they had good operability and practical value.

关 键 词：遥感影像 地物图斑 狭长结构 一致性融解 

分 类 号：P231[天文地球—摄影测量与遥感]