地理研究 ›› 2016, Vol. 35 ›› Issue (9): 1637-1646.doi: 10.11821/dlyj201609004

• 研究论文 • 上一篇    下一篇

多分析尺度下综合判别的地形元素分类方法

康鑫1,2(), 王彦文1, 秦承志1,4(), 程维明1, 赵尚民3, 朱阿兴1,4,5, 张文诗2   

  1. 1. 中国科学院地理科学与资源研究所,资源与环境信息系统国家重点实验室,北京 100101
    2. 信息工程大学地理空间信息学院,郑州 450052
    3. 太原理工大学矿业工程学院,太原 030024
    4. 江苏省地理信息资源开发与利用协同创新中心,南京师范大学地理科学学院,南京 210023
    5. Department of Geography, University of Wisconsin-Madison, Madison, WI 53706, USA
  • 收稿日期:2016-03-05 修回日期:2016-06-27 出版日期:2016-09-10 发布日期:2016-09-23
  • 作者简介:

    作者简介:康鑫(1982- ),男,新疆米泉人,博士,讲师,主要从事数字地形分析研究。E-mail: kangx@lreis.ac.cn

  • 基金资助:
    国家自然科学基金项目(41422109,41431177);信息工程大学学术基金项目(20153204)

A new method of landform element classificationbased on multi-scale morphology

Xin KANG1,2(), Yanwen WANG1, Chengzhi QIN1,4(), Weiming CHENG1, Shangmin ZHAO3, Axing ZHU1,4,5, Wenshi ZHANG2   

  1. 1. State Key Lab of Resources and Environmental Information System, Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
    2. College of Geospatial information, The PLA Information Engineering University, Zhengzhou 450052, China
    3. College of Mining Engineering, Taiyuan University of Technology, Taiyuan 030024, China
    4. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, School of Geographic Science, Nanjing Normal University, Nanjing 210023, China
    5. Department of Geography, University of Wisconsin-Madison, Madison, WI 53706, USA
  • Received:2016-03-05 Revised:2016-06-27 Online:2016-09-10 Published:2016-09-23

摘要:

地形元素(如山脊、沟谷等)是地表形态类型基本单元,通过地形元素的不同空间组合可形成更高级别的地貌类型。现有的地形元素提取方法大多依靠地形属性计算,难以克服地形元素的空间相关性表达与局部地形属性计算存在不对应的矛盾,Jasiewicz和Stepinski提出的Geomorphons方法——基于高程相对差异信息进行地形元素分类,可避免这一问题,但Geomorphons方法本质上是在单一分析尺度上选择地形特征点用于判别,易受局部地形起伏的影响而造成误分类。针对这一问题,设计出一种多分析尺度下综合判别的地形元素分类方法。应用结果表明:相比Geomorphons方法,利用该方法得到的地形元素的分类结果更为合理。

关键词: 地形元素, 分析尺度, 多尺度综合, 数字地形分析, DEM

Abstract:

Landform elements (e.g., ridges and valleys) are the basic morphologic features on the surface of the Earth. They can be assembled into various higher-level geomorphic types and aid in many other geographical modeling domains (e.g., digital soil mapping and landslide susceptibility mapping). Using a user-assigned set of topographic attributes derived from a gridded digital elevation model (DEM), most existing methods of landform element classification are likely to encounter a problem with local topographic attributes that cannot adequately capture the terrain context, which in turn will affect the reasonability of results from these landform element mapping methods. Additionally, the topographic attributes used for these methods are assigned by the user, when different topographic attribute sets affect the result obviously. These problems in existing methods of landform element classification can be avoided with the 'Geomorphons' method, recently proposed by Jasiewicz and Stepinski (2013). Instead of using topographic attributes, the Geomorphons method maps landform elements by recognizing the morphology of each cell of interest in a DEM that is defined according to its relative altitudes within the neighboring window of the interest cell. However, our analysis shows that this method recognizes the morphology based on the feature point of a single scale. This single-scale characteristic in the Geomorphons method cannot show the natural multi-scale characteristics of landform elements and thus might result in misclassification, especially in rugged areas. To solve the problem, we improved the Geomorphons method and propose a new method of landform element classification based on multi-scale morphology. The proposed method classifies each cell in a DEM to be a specific landform element by synthesizing morphologies that are judged under a series of analysis window. Experimental results show that the proposed method produces more reasonable results than the original Geomorphons method.

Key words: landform elements, analysis scale, multi-scale synthesis, digital terrain analysis, DEM