测绘学报 ›› 2015, Vol. 44 ›› Issue (1): 67-75.doi: 10.11947/j.AGCS.2015.20130524

• 摄影测量学与遥感 • 上一篇    下一篇

地形开度和差值图像阈值分割原理相结合的黄土高原沟沿线提取法

王轲1,2, 王琤1, 张青峰1, 丁凯隆1   

  1. 1. 西北农林科技大学资源环境学院, 陕西 杨凌 712100;
    2. 南京师范大学地理科学学院, 江苏 南京 210023
  • 收稿日期:2013-12-27 修回日期:2014-09-17 出版日期:2015-01-20 发布日期:2015-01-22
  • 通讯作者: 王琤 E-mail:chengwong@nwsuaf.edu.cn
  • 作者简介:王轲(1991-), 男, 硕士生, 研究方向为地图学与数字地貌.
  • 基金资助:

    西北农林科技大学中央高校基本科研业务费专项(QN2012042);西北农林科技大学博士科研启动基金(201104050395)

Loess Shoulder Line Extraction Based on Openness and Threshold Segmentation

WANG Ke1,2, WANG Cheng1, ZHANG Qingfeng1, DING Kailong1   

  1. 1. College of Resources & Environment, Northwest A&F University, Yangling 712100, China;
    2. School of Geography Science, Nanjing Normal University, Nanjing 210023, China
  • Received:2013-12-27 Revised:2014-09-17 Online:2015-01-20 Published:2015-01-22
  • Supported by:

    The Fundamental Research Funds for the Central Universities(No.QN2012042) Doctoral Scientific Research Foundation of Northwest A&F University(No.201104050395)

摘要:

鉴于地形正负开度对地貌的良好表达且具有分析尺度灵活性这一重要特性,本文提出一种借助地形正负开度及其差值图像阈值分割的黄土地貌沟沿线提取方法.首先,计算DEM的地形正负开度,对正负开度进行差值运算以得到开度差值图;然后,对开度差值图进行阈值处理以得到研究区正负地形空间分布特征;最后,借助数学形态学原理完成对二值化开度差值图正负地形边界——沟沿线的自动提取.试验采用高分辨率DEM数据对陕西洛川塬部分地区进行沟沿线信息提取.结果表明,与其他提取技术方法相比,该方法不仅增强了地貌基本形态特征在沟沿线提取过程中的作用与影响,同时也在一定程度上实现了沟沿线提取的精确化和自动化.

关键词: 黄土高原, 高分辨率DEM, 地形开度, 正负地形, 沟沿线

Abstract:

This paper presents a new method of loess shoulder line extraction with openness and threshold segmentation. Openness is an angular measurement of the relationship between surface relief and horizontal distance which can be derived from DEM. As it has the outstanding ability of expressing the positive and negative terrain characteristics and the attribute of "analysis-scale flexibility", in this study, we developed the idea of loess shoulder line extraction based on the difference of positive and negative openness images and threshould segmentation. Firstly, we calculated the positive and negative openness of our study area based on DEM. Then, a difference image was generated by the subtraction of positive and negative openness images. Afer that, it was processed by a threshold segmentation method, and we analyzed some important spatial distribution characteristics of positive and negative terrain in study area. Finally, we extracted the shoulder lines semi-automatically by an edge-detection method which came from the concept of morphology. Our study area locates in Louchuan, Shaanxi province, a typical hilly gully region in northwstern China. Considering the effect of the local window size on analysis accuracy and the match of the fitness surface and DEM terrain surface, we chose a 5 m×5 m high-resolution DEM of Luochuan which was purchased from Shaanxi Administration of Surveying, Mapping and Geoinformation. The result manifested that our method enhanced and guaranteed the accuracy ofthe basic geomorphological characters of study area.

Key words: loess plateau, high-resolution DEM, topographic openness, positive and negative terrain, loess shoulder line

中图分类号: