首页 | 本学科首页   官方微博 | 高级检索  
     检索      

改进约束D-TIN的等高线陡坡区域识别
引用本文:范亚浩,吴晟,张长胜,李英娜,李川.改进约束D-TIN的等高线陡坡区域识别[J].测绘科学,2017,42(11).
作者姓名:范亚浩  吴晟  张长胜  李英娜  李川
作者单位:昆明理工大学信息工程与自动化学院,昆明,650500
基金项目:国家自然科学基金项目,云南电网科技项目
摘    要:针对高线图中陡坡区域特征的自动识别问题,在应用Delaunay三角网与线性插值法生成等高线图的基础上,改进了基于最大角原则生成约束Delaunay三角网的算法,对等高线图进行了三角剖分,根据等高线之间约束三角形的几何参数,计算了等高线图坡面单元坡度,识别出坡度较高区域。进一步地,给出了一种扩张算法对相连陡坡单元进行划分形成陡坡区域,计算了各个陡坡区域中心坐标、面积与平均坡度。通过对12组有256个坐标高程值的数据构建约束D-TIN并生成三维地形图对识别结果进行了评价,识别正确率达0.903,平均识别时间为33ms。实验表明,对于不同数据生成的不同等级等高线图的识别结果均有较高的效率与准确率。

关 键 词:等高线  陡坡区  约束Delaunay三角网  区域识别

Research on abrupt slope regions recognition of contour map based on improved constrained D-TIN
FAN Yahao,WU Sheng,ZHANG Changsheng,LI Yingna,LI Chuan.Research on abrupt slope regions recognition of contour map based on improved constrained D-TIN[J].Science of Surveying and Mapping,2017,42(11).
Authors:FAN Yahao  WU Sheng  ZHANG Changsheng  LI Yingna  LI Chuan
Abstract:To automatically recognize steep slope regions of contour map,generated contour map using Delaunay Triangulated Irregular Network and linear interpolation method,improved the algorithm of generating constrained Delaunay triangulated irregular network(D-TIN)based on the principle of maximum angle.Subdivided contour map to triangles,according to the geometric parameters of the constrained triangles between contour lines calculated gradients of slope units,recognized regions with high gradient,furthermore,given an expansion algorithm for combining connected abrupt slope units to regions,calculated center coordinates,areas and average gradients of abrupt slope regions.Evaluated the recognition results by using the data of 12 coordinate elevation values to construct the constraint D-TIN and generating three-dimensional topographic map,accuracy is 0.903 and time is 33 milliseconds.Experimental results show that recognition results have high efficiency and accuracy on different level contour maps generated on different data.
Keywords:contour line  abrupt slope regions  constraint Delaunay Triangulated Irregular Network  Region recognition
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号