EN中文

Journal of Geo-information Science >

2016 , Vol. 18 >Issue 2: 220 - 226

DOI: https://doi.org/10.3724/SP.J.1047.2016. 00220

Orginal Article

Shoulder Line Extraction in the Loess Plateau Based on Region Growing Algorithm

  • LIU Wei ,
  • LI Fayuan , * ,
  • XIONG Liyang ,
  • LIU Shuanglin ,
  • WANG Ke
Expand
  • 1. Key Laboratory of Virtual Geographic Environment of Ministry of Education, Nanjing Normal University, Nanjing 210023, China
  • 2. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, China
*Corresponding author: LI fayuan, E-mail:

Received date: 2013-11-15

  Request revised date: 2014-03-26

  Online published: 2016-02-04

Copyright

《地球信息科学学报》编辑部 所有
Fold

Abstract

Shoulder line is one of the most effective terrain structure line used to describe the loess landform. It plays an important role in the study of the spatial distribution and landform evolution of the loess landform. Shoulder line lies on the boundary of the positive and negative terrain, where the elevation and slope of the loess surface reveal obvious changes. Many efforts have been done for the extraction of shoulder line, but these methods have disadvantages, of which the experimental process requires human intervention or the shoulder line is discontinuous. P-N terrain method can effectively extract the shoulder line, but while using this method to segment the positive and negative terrains, it tends to produce large amounts of broken polygons and classification errors, which affect the accuracy of shoulder line. This paper investigates a region growing algorithm to improve the P-N terrain method. Using the highest elevation of the local area as a growing point for positive terrain and the outlet as a growing point for negative terrain, four-neighborhood growth were carried out until they reached the boundary of the positive and negative terrains or a slope threshold. Then the edge detection method was used to extract the critical boundary. Finally, the morphological image processing method was used to eliminate burrs to get the final shoulder line. In order to verify the result, this paper used the 0.6 m resolution remote sensing image to get the relatively accurate shoulder line by visual interpretation. And then different results were compared using overlay analysis. It is revealed that the shoulder line extracted using the improved method is closer to the visual interpretation results. This method is an automatic way to extract shoulder line, which solves the inaccurate location problem of P-N terrain method. Meanwhile, this method keeps the integrity and continuity of shoulder line and avoids the emergence of broken shoulder line and closure shoulder line. The use of morphological image processing method in burrs removal also ensures the accuracy.

Key words： shoulder line; region growing; morphological image processing; P-N terrain method; DEM

Cite this article

LIU Wei , LI Fayuan , XIONG Liyang , LIU Shuanglin , WANG Ke . Shoulder Line Extraction in the Loess Plateau Based on Region Growing Algorithm[J]. Journal of Geo-information Science, 2016 , 18(2) : 220 -226 . DOI: 10.3724/SP.J.1047.2016. 00220

1 引言

黄土地貌及其空间分异研究是黄土高原研究的重点之一[1-8]。沟沿线是一条最能体现黄土地貌特征的地形结构线,对其正确提取与有效分析是探究黄土高原地形、土壤侵蚀方式和土地利用格局的基础[9],对研究黄土高原空间分异规律和地貌演化机制具有重要意义。
DEM提取沟沿线的研究已有许多学者进行了诸多探索。闾国年[10]提出基于地貌形态和递归思想的沟沿线提取算法;朱红春[11]利用坡度变率和剖面曲率来定位沟沿线;刘鹏举等[12]提出利用水流路径进行分布式水流计算来提取沟沿线的方法;肖晨超[13]根据黄土地貌坡面坡度转折特征提出了基于坡面朝向的沟沿线提取方法;李小曼[14]以沟沿线和坡脚线形态为基础,使用窗口分析方法来定义规则得到沟沿线和坡脚线;周毅[15]通过上下游栅格点坡度对比,利用汇水模型提取沟沿线点约束的上游汇水区域来分割正负地形,提取沟沿线;晏实江[16]使用图像二值化和多种边缘检测算子提取沟沿线;陈永刚[17]提出了基于多方位DEM地形晕渲的黄土地貌正负地形提取;晏实江[18]提出了一种整合D8算法和Snake模型来提取沟沿线的方法;王珂[19]借助地形正负开度及其差值图像阈值分割原理相结合,并使用数学形态学原理来提取沟沿线。但是,上述方法都存在实验过程需要人工干预或沟沿线不连续的缺陷。周毅[20-21]提出的坡面畸变邻域法,充分考虑了黄土地貌起伏特征,适用性较强,但由于黄土地貌的复杂性,在正负地形分类过程中存在错误分类的情况,这些错误会造成沟沿线定位的不准确。现有纠错方法主要通过人工编辑或根据坡度、斑块面积等进行修正,这些方法操作繁琐,需要人工干预,降低了提取效率。区域生长方法是一种依据像元之间的相似性进行区域合并的方法,区域内像元的相似性度量可包括平均灰度值、纹理、颜色等信息,该方法已广泛地应用于遥感图像处理、图形分割、目标检测等方面[22-26]。区域生长法在栅格图像分割和边界提取方面的优势,为提高沟沿线提取精度提供了新的思路。本文依据区域生长的方法,实现了对传统坡面畸变邻域法提取沟沿线的改进,提高了沟沿线的识别精度,保证了沟沿线的完整性。

2 沟沿线提取的方法

2.1 区域生长算法

区域生长是根据定义的准则将像元或者子区域聚合成更大区域的过程。其基本思想是从一组生长点开始,将与该生长点性质相似或相同的相邻像元进行合并,形成新的生长点,重复此过程直至没有可合并的点为止。区域生长一般有3个步骤：（1）选择合适的生长点;（2）确定相似性准则即生长准则;（3）确定生长停止条件[27]。
图1是区域生长的示意图,图中数字表示像元灰度,假设以灰度为8的像元为初始生长点,在4邻域内,生长准则是待测点灰度与生长点灰度值相差为1或0。如果相邻2次生长结果不发生变化,即已经不存在满足生长准则的像元点,则停止生长。
Fig. 1 Illusions of region growing

图1 区域生长示意图

2.2 正负地形的识别

沟沿线是正负地形的分界线,正负地形边界的确定是准确提取沟沿线的前提。正地形是相对高于邻区或构造上升的地形,坡面畸变邻域法提取沟沿线时,正地形上分类错误的区域分为2种情况：(1)洼地,由于人工改造或小范围地形起伏造成;(2)平地,当滤波窗口较小时,分析窗口在近似平坦的DEM栅格上滑动,原始高程值与平滑后的高程均值相差很小,稍有高程变化都会影响运算结果,这种现象在平坦黄土塬区表现最为明显。换言之,局部最高点一定是分类正确且位于正地形上的点,因此,将局部最高点作为正地形的生长点,生长准则是使用坡面畸变邻域法提取出的正地形,生长至到没有满足生长条件的点为止。由于本文并不需要十分精确地确定每个正地形生长点的位置和级别,只是提取局部凸出的点,所以,采用流域边界线的邻域分析方法提取山顶点[28]。邻域分析窗口的大小取决于样区地形的破碎程度,地形越破碎,使用的窗口越小,以此保证正地形提取的完整性。
负地形是相对低于邻区或新构造下沉地区的地形,由于实验样区是完整的流域,所以整个负地形区域都是连通的,只需选取一个点作为负地形的生长点即可。流域出水口处高程值最低且位于负地形上,所以选取出水口作为生长点,生长准则是使用坡面畸变邻域法提取出的负地形,生长至没有满足生长条件的点为止。
区域生长的方法可避开大部分分类错误的正负地形,但是如果分类错误的区域与正确的正负地形区域连通（图2）,会被识别为与邻接斑块相同的地貌类型,继而造成沟沿线定位错误,这种现象在黄土塬区的正地形上最为明显。为了避免这种现象,本文加入了坡度变量,定义规则为：与正地形邻接的负地形区域如果坡度小于给定的坡度值,则认为该区域为正地形区域。坡度阈值的确定与研究区黄土地貌类型有关,通过对不同坡度值进行试验并将实验结果与高分辨率遥感影像进行对比,本文选择7°作为坡度阈值,实验流程如图3所示。
Fig. 2 Influence of slope to the positive and negative terrain classification result

图2 坡度对正负地形分类结果的影响

Fig. 3 Flow chart of the study

图3 流程图

2.3 沟沿线的提取

经过上述处理,错误分类的区域会形成空值区,而正确分类的区域形成正地形区（图4中蓝色区域）或负地形区（图4中绿色区域）,沟沿线位于正确分类的正负地形交界处（图4）。将绿色区域赋值为1,蓝色区域赋值为0,空值区赋值为-9999。根据图像处理中的边缘检测方法提取沟沿线,使用简单的边缘检测算子（图5）对全图进行扫描,如果窗口内同时存在0和1,即可认为此处存在沟沿线并对此处值为1的正地形点进行标记,所有标记点导出后可得到整个流域的沟沿线。
Fig. 4 The contrast between shoulder line and other features

图4 沟沿线和非沟沿线边界对比图

Fig. 5 Edge detection operators

图5 边缘检测算子

2.4 沟沿线上寄生分量的去除

按照上述方法提取的沟沿线,已基本符合使用要求,但是通过观察沟沿线细部特征,发现沟沿线在转角处仍然存在少量的寄生分量,即通常所说的“毛刺”现象。这种现象会造成沟沿线边界的不确定性,尤其在栅格转矢量时,该缺陷变得更为突出。
本文运用形态学图像处理方法去除“毛刺”,其基本原理是通过使用固定结构元检测端点并进行细化（式(1））,即用模板对图像进行匹配检测和处理,从而达到去除“毛刺”的效果。算法如下：（1）确定端点细化阈值。通过观察图像,发现“毛刺”长度一般不超过3个像元,为保证实验结果的可靠性,将阈值设置为5个像元。（2）确定结构元。由于上述使用的是4邻域生长,去“毛刺”时也无需考虑对角线像元,所以使用如图6的结构元。“×”表示一个不必考虑的条件,“1”表示沟沿线像元,“0”表示背景像元,将符合此结构元的中心像元设为背景像元（式(2））。结构元全图扫描一次只能去除每个端点处的一个“毛刺”像元,为了保证连续像元在5个以下“毛刺”全部去除,需对全图进行5次扫描,处理前后如图7所示。（3）复原正确端点。由于处理过程是对所有端点进行处理,所以正确的端点处也进行了“去毛刺”处理。对原本正确的端点进行复原需再次对全图进行“去毛刺”处理,由于“毛刺”已经在前5次细化后全部去除,所以,本次将要去的是正确端点处的像元,对这些像元进行4邻域膨胀并与原始沟沿线取交集（式(3））,膨胀次数与细化次数相同,这样就实现了端点的复原。
X 1 = A ⊚ { B } （1）
式中： X 1 为细化结果;A表示原始沟沿线;B为结构元; ⊚ 表示细化操作。
X 2 = ⋃ k = 1 8 ( X 1 ⊙ B k ) （2）
式中： X 2 为端点集合; B k (k=1,2,3,4)为结构元; ⊙ 为击中击不中变换。
X 3 = ( X 2 ⊕ H ) ∩ A （3）
式中： X 3 为沟沿线结果;H为4邻域结构元;A为原始沟沿线; ⊕ 为膨胀操作。
Fig. 6 Structure element

图6 结构元

Fig. 7 The contrast of data before and after burr removal

图7 去“毛刺”前后对比图

3 黄土地貌沟沿线提取实例分析

3.1 实验数据

本文实验样区位于陕西省宜君县城东北部洛河中下游地区（图8）,地貌类型属于黄土长墚残塬沟壑区,该地区沟谷溯源侵蚀强烈,重力侵蚀活跃,相对切割深度70~200 m,海拔777~1112 m。实验数据为国家测绘局生产的1:1万比例尺5 m分辨率DEM和与DEM配准的0.6 m分辨率全色波段遥感影像。DEM数据及其派生的坡度数据用于沟沿线的提取。影像数据用于沟沿线的目视解译,对比分析本文提取方法的精度。在进行黄土地貌分析时,通常以流域为研究单元,通过计算沟沿线相关指标来探究黄土高原的沟蚀程度,故此,本文以流域为单位对沟沿线提取方法进行探讨。
Fig. 8 Test area and data

图8 实验样区及数据

3.2 参数设置

首先,采用坡面畸变邻域法实现沟沿线的初步提取,通过对不同分析窗口的多次试验,采用17×17[29]窗口大小作为沟沿线提取的分析单元。其次,以区域生长方法改进由于正负地形分类错误造成的沟沿线定位不准确的问题;正负地形生长点分别设置为山顶点和出水口点,其中这些地形特征点采用流域边界线的方法进行提取[28],生长准则是合并灰度值相同的像元,生长至没有满足条件的点为止。

3.3 结果分析

由图9(a)可知,使用坡面畸变邻域法提取的正负地形,存在许多分类错误的情况,正地形内部存在许多独立的负地形斑块,负地形内也有正地形斑块出现。如果使用该处理结果直接提取沟沿线,将严重影响沟沿线的提取精度和正确率（图9(b)）。改进的沟沿线提取方法,其本质就是提取正负地形的交界线,绕过分类错误的小斑块。通过观察提取结果,发现塬面和沟谷内部细碎的线已经全部清除,沟沿线连续且完整,沟沿线上的“毛刺”也被全部去除（图9(c)）。
Fig. 9 Positive and negative terrains and shoulder line

图9 正负地形及沟沿线

3.4 精度评价

为了验证实验结果,本文使用0.6 m分辨率遥感影像进行人工解译,获得相对精确的沟沿线（图9(d)）。将区域生长法提取的沟沿线与人工解译结果相比较,二者形状和位置都非常相近。由于人工解译结果是在0.6 m分辨率的影像上进行的,而自动提取沟沿线的DEM分辨率为5 m,所以与DEM自动提取的沟沿线相比,人工解译结果更加精细。为了进一步对比沟沿线提取精度,本文统计了在人工解译沟沿线的不同偏移栅格内,坡面畸变邻域法和区域生长法提取的沟沿线栅格落在其中的数目和比例,得到如下结果（表1）：使用坡面畸变邻域法提取的沟沿线栅格总数为51 655,沟沿线栅格落在3个偏移栅格以内的概率为64.02%,5个偏移栅格以内的概率为74.61%,7个偏移栅格以内的概率为78.17%;使用区域生长方法得到的沟沿线栅格总数为30 093,沟沿线栅格落在3个偏移栅格以内的概率为78.07%,5个偏移栅格以内的概率为90.49%,7个偏移栅格以内的概率为94.60%。结果表明,本文改进的沟沿线提取方法大大地提高了沟沿线定位的准确率和识别的正确性。
Tab. 1 The percentage of grid numbers of shoulder line to the total grid numbers within a certain displacement

表1 位移栅格内沟沿线栅格数目占总栅格数目的百分比

人工解译沟沿线的位移栅格数目/个 位移栅格内,坡面畸变法提取沟沿线栅格数目/个 占坡面畸变法提取沟沿线总栅格数的百分比/(%) 位移栅格内,区域生长法提取沟沿线栅格数目/个 占区域生长法提取沟沿线总栅格数的百分比/(%)
3 33 072 64.02 23 494 78.07
5 38 539 74.61 27 231 90.49
7 40 381 78.17 28 469 94.60

4 结论

以区域生长法和形态学图像处理理论,对坡面畸变邻域法提取沟沿线进行改进,在很大程度上提高沟沿线的识别精度。区域生长法避免了坡面畸变邻域法进行正负地形分类时,小斑块分类错误的现象;“去毛刺”处理减少了数据从栅格转到矢量过程中发生的错误。本文改进的沟沿线提取方法,能保证沟沿线提取的连续性和完整性,避免了破碎沟沿线和小范围闭合沟沿线的出现。该方法不需要人工干预,操作简单,具有普遍的适应性,在黄土塬、黄土墚、黄土峁地貌地区均能使用,其中在黄土塬区表现最为明显;并且该方法具有明确的地理学含义,山顶点和出水口点位置明确且一定分别位于正地形和负地形上。

The authors have declared that no competing interests exist.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
曹银真. 黄土地区墚峁坡地特征与土壤研究[J].地理研究,1983,2(3):19-29.本文指出黄土地区梁峁坡的坡角主要分布在14°—28°之间,特征坡角的上限相当于黄土的内摩擦角,下限相当于内摩擦角的0.45倍。黄土地区的坡地形态可分为直形坡、凸形坡、凹形坡、复合形坡和阶梯形坡等五种类型。土壤侵蚀主要受坡地形态和坡度的控制,侵蚀量随着坡度的增大而增加,坡度大于15°时,侵蚀量突增,坡面冲刷加剧,26°达极大值,此后冲刷作用减弱,重力侵蚀逐渐显著,至45°侵蚀达到最大值,以后又趋减弱。

[ Cao Y Z.Slope feature and soil erosion in the loess region[J]. Geographical Research, 1983,2(3):19-29. ]

[2]
王春,李虎,杨军生,等.新型格网DEM等高线生成技术与方法[J].地球信息科学学报,2015,17(2):160-165.<p>以格网DEM数据为基础,生成具有良好保凸性、光滑性、平顺性特征的高质量等高线,已逐渐成为制作等高线图的重点。鉴于现有技术方法在沟道、山脊等地形变化部位生成的等高线中存在生硬、相交、异常闭合等问题,本文以黄土丘陵1:5 万不同格网分辨率DEM为例,采用对比分析方法,建立了基于DEM格网加密技术的高质量等高线生成方法。实验表明:(1)采用以4&times;4 DEM格网单元为搜索圆的完全规则样条函数插值法,以可视化时图面0.1 mm的实际距离为加密DEM的格网分辨率值,对DEM格网进行加密处理后生成的等高线具有更为理想的保凸性、光滑性和平顺性特征;(2)对于已知综合尺度的地形,存在最优DEM格网分辨率阈值,若要高质量地再现原始等高线,则实际DEM格网分辨率不得低于最优DEM格网分辨率阈值。</p>

DOI

[ Wang C, Li H, Yang J S, et al.Study on generation technique of high contour lines based on grid DEM[J]. Journal of Geo-information Science, 2015,17(2):160-165. ]

[3]
励强,陆忠臣,袁宝印.地貌发育阶段的定量研究[J].地理学报,1990,45(1):110-120.文章提出用临界侵蚀积分值作为 划分流域发育阶段的定量指标,是对经典理论的发展,主要表现在:1)将地貌发育阶段纳入开放系统的概念之中;2)避免高程积分法在划分发育阶段上的不确定 性;3)建立地貌发育阶段与流域物质迁移之间的定量关系;4)区分自然侵蚀和加速侵蚀。用文中的临界侵蚀方程对黄土高原丘陵沟壑区分析计算,得到临界侵蚀 积分值为0.30和0.70,即当侵蚀积分值小于0.30寸,流域处于侵蚀早期;当侵蚀积分值大于0.30,小于0.70时,流域处于侵蚀中期;当侵蚀积 分值大于0.70时,流域处于侵蚀晚期。对于纯自然侵蚀过程,与临界侵蚀积分值相对应的年代及侵蚀模数为7万年前,7008t/km~2·a和距今7

[ Li Q, Lu Z C, Yuan B Y.Quantitative study of the stage of geomorphological evolution[J]. ACTA Geogracial Sinica, 1990,45(1):110-120. ]

[4]
吴伯甫,陈明荣,陈宗兴,等.中国的黄土高原[M].西安:陕西人民出版社,1991.

[ Wu B F, Chen M R, Chen Z X, et al.China Loess Plateau[M]. Xi’an: Shaanxi People's Publishing House, 1991. ]

[5]
罗兴来. 划分晋西、陕北、陇东黄土区域沟间地与沟谷的地貌类型[J].地理学报,1956,22(3):201-222.

[ Luo X L.A tentative classification of landforms in the Loess Plateau[J]. ACTA Geogracial Sinica, 1956,22(3):201-222. ]

[6]
朱红春,李永胜,汤国安.面向沟谷特征点簇的空间结构模型与应用[J].地球信息科学学报,2014,16(5):707-711.点簇(即点的集合)具有特定的组织结构；沟谷作为一类自成体系的重要空间实体，表达其形态、结构、发育等关键属性特征的沟谷特征点簇，是有严密的组织结 构、密切的空间关系和完整的属性描述的关键特征点的集合。建立沟谷特征点簇的空间结构模型，进行沟谷特征的空间分析与应用，具有重要的科学意义。本文在分 析沟谷特征点簇的组成要素、空间与结构特征的基础上，确定了点簇的层次结构模型和非结构化文件存储方式，编程实现了沟谷特征点簇数据模型；然后，构建了点 簇文件，并运用C#+AE的开发方式，研发建立了基于点簇的沟谷分析原型系统，实现了沟谷特征点簇数据的可视化、结构空间分析、专题属性特征分析等基本功 能；最后，以特征点的沟谷路径追溯查找效率作了对比分析，验证了基于特征点簇模型的沟谷分析具有较高的处理效率。本研究构建的面向沟谷特征点簇的空间结构 模型，研发的具有沟谷结构、形态与属性分析功能的沟谷分析原型系统，有一定的应用价值。

DOI

[ Zhu H C, Li Y S, Tang G A.The spatial structural model established for gully feature points cluster and its application[J]. Journal of Geo-information Science, 2014,16(5):707-711. ]

[7]
Gong J G, Jia Y W.An experimental study on dynamic processes of ephemeral gully erosion in loess landscapes[J]. Geomorphology, 2011,125(1):203-213.Ephemeral gully erosion is widespread on the Loess Plateau of China. To provide theoretical support for constructing numerical models and soil-water conservation planning, the hydrodynamic characteristics of ephemeral gully flows were studied by laboratory experiments with a physical model. We firstly concluded that, with the influence of “step-slope-pool” (or “step-pool”) terrain on the bottom of gully channels, no obvious or consistent pattern changes were evident in the hydrodynamic parameters along the flow direction. Secondly the change of hydrodynamic parameters during erosion could be classified into an initially fast changing interval followed by a stable period, and the initial adjustment period would become shorter with increasing slope and rainfall intensity. Thirdly ephemeral gully erosion flows were turbulent and changed frequently in space and time between supercritical and subcritical flows. Throughout the experiment, the Reynolds number fluctuated and increased with time. With increasing rainfall intensity and gully slope, the frequency of larger Reynolds numbers increased. In addition, the Froude number also fluctuated, but decreased with time, and the mean Froude number eventually stabilized at ~0.5 at different slope angles and rainfall intensities. Fourthly the distribution of erosional energy was influenced mainly by topography in loess slopes. In the case of 20° slopes, the inverse ratios of dissipation as soil erosion, sediment transport, and flow kinetic energy were relatively stable in time and space during ephemeral gully erosion. Moreover, flow velocity, shear stress, the Darcy–Weisbach friction factor, and the Manning coefficient were influenced by rainfall intensity to a lesser degree during erosion. In the stable interval, for 15° and 25° slopes, shear stress, the Darcy–Weisbach friction factor and Manning coefficient of ephemeral gully flow both increased with increasing rainfall intensity. Flow velocity on the 15° slope decreased initially then increased with increasing rainfall intensity, but flow velocity on the 25° slope decreased with increasing rainfall intensity. Finally the regularity seen in hydrodynamic parameters of ephemeral gully flows occurred despite disturbance from lateral confluences, gravity erosion of the channel bank, and changes in terrain. The sediment transport capacity of ephemeral gully flows increased with USP by a linear function (r=0.64, n=99, P<0.01).

DOI

[8]
Hessel R, Asch T.Modeling gully erosion for a small catchment on the Chinese loess plateau[J]. Catena, 2003,54(1-2):131-146.

[9]
梁广林,陈浩,蔡国强,等.黄土高原现代地貌侵蚀演化研究进展[J].水土保持研究,2004,11(4):131-137.地貌学是研究整个地表形态或某一个地区地表起伏形态特征、成因、发育规律、分布和改造利用的科学.根据前人研究成果,概述了国内外地貌演化,尤其是黄土高原现代地貌侵蚀演化的主要研究进展,并评述了研究中存在的主要问题和目前需要开展的研究内容.

DOI

[ Liang G L, Chen H, Cai G Q, et al.Research progress of modern topographic evolvement and landform erosion in loess Plateau[J]. Research of Soil and Water Conservation,2004,11(4):131-137. ]

[10]
闾国年,钱亚东,陈钟明.基于栅格数字高程模型自动提取黄土地貌沟沿线技术研究[J].地理科学,1998,18(6):567-573.从数字高程模型(DEM)中提 取具有精确定位特征的连续的沟沿线是构建黄土沟壑丘陵地区土壤侵蚀、泥沙搬运和径流过程的空间分布式机理一过程模型的关键技术.本文提出了基于地貌形态学 特征的地貌提取技术,在提取具有代表性的黄土丘陵沟壑区山西离石王家沟流域汇流网络的基础上,自动提取了该流域完整的沟沿线,得到了满意的结果.

[ Lv G N, Qian Y D, Chen Z M.Study of automated extraction of shoulder line of valley from Grid Digital Elevation data[J]. Scientia Geographica Sinica, 1998, 18(6):567-573. ]

[11]
朱红春,汤国安,张友顺,等.基于DEM提取黄土丘陵区沟沿线[J].水土保持通报,2003,23(5):43-45,61.黄土丘陵区的沟沿线,是反映该 区地形地貌特征的重要的地形结构线。以陕北绥德县韭园沟流域为实验样区,探讨利用数字高程模型(DEM)实现自动提取沟沿线的原理方法,并以航片及地形图 等高线为依据,对所提取沟沿线的精度进行了对比分析。实验结果证明,利用DEM自动提取沟沿线是一种快速、有效的技术方法。

DOI

[ Zhu H C, Tang G A, Zhang Y S, et al.Thalweg in Loess Hill area based on DEM[J]. Bulletin of Soil and Water Conservation, 2003,23(5):43-45,61. ]

[12]
刘鹏举,朱清科,吴东亮,等.基于栅格DEM与水流路径的黄土区沟缘线自动提取技术研究[J].北京林业大学学报,2006,28(4):72-76.基于数字高程模型(DEM)进 行沟缘线的自动提取是黄土丘陵沟壑区进行分布式水文计算与土壤侵蚀建模的关键技术.该文以基于栅格数字高程模型进行河网提取与流域划分技术为基础,基于栅 格单元间八流向算法构建水流路径,提出一种基于汇流路径坡度变化特征确定沟坡段,进一步形成封闭沟缘线的新方法,并编程实现了该方法.通过对蔡家川流域 1∶10000等高线生成2m数字高程模型,依据35°作为判断沟坡单元的标准,自动生成沟缘线分布图,并与原始坡度分布图比较,取得满意的结果.

DOI

[ Liu P J, Zhu Q K, Wu D L, et al.Automated extraction of shoulder line of valleys based on flow path from grid digital elevation model (DEM) Data[J]. Journal of Beijing Forestry University, 2006,28(4):72-76. ]

[13]
Tang G, Xiao C, Jia D, et al.DEM based investigation of loess shoulder-line[C]. Geoinformatics 2007. International Society for Optics and Photonics, 2007: 67532E(1-12).

[14]
李小曼,王刚,李锐.基于DEM的沟缘线和坡脚线提取方法研究[J].水土保持通报,2008,28(1):69-72.提出了一种基于DEM的沟缘线和坡脚线的提取方法，该方法以沟缘线和坡脚线的形态特征为基础。沟缘线作为梁峁地和沟坡地的分界线位于凸坡上，沟缘线象素以下的坡度应大于某一坡度值；坡脚线作为沟坡地和沟底地的交界线位于凹坡上，坡脚线象素以上的坡度应大于某一坡度值；沟缘线和坡脚线上的象素都具有局部最大的坡度变化。利用这种特性进行沟缘线和坡脚线的提取，易于实现。处理的速度快、效率高，提取结果较准确。

[ Li X M, Wang G Li R. A DEM based method for extraction of valley shoulder line and slope heel line[J]. Bulletin of Soil and Water Conservation, 2008,28(1):69-72. ]

[15]
Zhou Y, Tang G A, Yang X, et al.Positive and negative terrains on northern Shaanxi Loess Plateau[J]. Journal of Geographical Sciences, 2010,20(1):64-76.lt;a name="Abs1"></a>The Loess positive and negative terrains (P-N terrains), which are widely distributed on the Loess Plateau, are discussed for the first time by introducing its characteristic, demarcation as well as extraction method from high-resolution Digital Elevation Models. Using 5 m-resolution DEMs as original test data, P-N terrains of 48 geomorphological units in different parts of Shaanxi Loess Plateau are extracted accurately. Then six indicators for depicting the geomorphologic landscape and spatial configuration characteristic of P-N terrains are proposed. The spatial distribution rules of these indicators and the relationship between the P-N terrains and Loess relief are discussed for further understanding of Loess landforms. Finally, with the integration of P-N terrains and traditional terrain indices, a series of un-supervised classification methods are applied to make a proper landform classification in northern Shaanxi. Results show that P-N terrains are an effect clue to reveal energy and substance distribution rules on the Loess Plateau. A continuous change of P-N terrains from south to north in Shaanxi Loess Plateau shows an obvious spatial difference of Loess landforms and the positive terrain area only accounted for 60.5% in this region. The P-N terrains participant landform classification method increases validity of the result, especially in the Loess tableland, Loess tableland-ridge and the Loess low-hill area. This research is significant on the study of Loess landforms with the Digital Terrains Analysis methods.

DOI

[16]
晏实江,汤国安,李发源,等.利用DEM边缘检测进行黄土地貌沟沿线自动提取[J].武汉大学学报(信息科学版),2011,36(3):363-366.在分析黄土地貌沟沿线形态特征 的基础上,提出了基于规则格网DEM,通过引入边缘检测算子提取并连接沟沿线候选点,并借助形态学方法滤除细碎线段,最终生成沟沿线的方法。黄土丘陵沟壑 区的实验结果显示,Sobel、Roberts、Prewitt梯度算子检测得到的沟沿线封闭性较差,线段较零碎,与实际沟沿线不符,不适用于沟沿线的自 动提取;而LOG算子提取的线段连续、精度高,是理想的沟沿线检测算子,能够实现黄土地貌沟沿线较准确、有效的自动提取。

[ Yan S J, Tang G A, Li F Y, et al.An edge detection based method for extraction of loess shoulder-line from grid DEM[J]. Geomatics and Information Science of Wuhan University, 2011,36(3):363-366. ]

[17]
陈永刚,汤国安,周毅,等.基于多方位DEM地形晕渲的黄土地貌正负地形提取[J].地理科学,2012,32(1):105-109.以陕北绥德县韭园沟流域5 m分辨率DEM数据为基础,在数字地形分析、多元统计和数据挖掘方法的基础上,提出利用多方位DEM地形晕渲、坡度等多元指标,以主成分分析消除多重共线 性和约减维数,并以Logis-tic回归模型提取黄土高原正、负地形的方法。研究结果表明:模型提取精度为82.1%,Kappa统计量为0.629; 模型在6个不同流域测试样本上正、负地形的平均精度分别为77.6%,84.9%,加权平均精度为81.3%,模型具有较好的一致性和泛化能力,正、负地 形提取效果良好。

[ Chen Y G, Tang G A, Zhou Y, et al.The positive and negative terrain of Loess Plateau extraction based on the Multi-azimuth DEM shade relief[J]. Scientia Geographica Sinica, 2012,32(1):105-109. ]

[18]
Yan S, Tang G, Li F, et al.Snake model for the extraction of loess shoulder-line from DEMs[J]. Journal of Mountain Science, 2014,11(6):1552-1559.Shoulder lines are the most important landform demarcations for geographical analysis, soil erosion modeling and land use planning in the Loess Plateau area of China. This paper proposes an automatic, effective and accurate method of determining loess shoulder line from DEMs by integrating a hydrological D8 algorithm and a snake model. The watershed boundary line is adopted as the initial contour which evolves to identify the exact position of loess shoulder-line by the guidance of an external force of snake model from DEMs . Experiments show that the method overcomes the difficulties in both threshold selection for edge detection and the disconnecting issues in former extraction approaches. The accuracy evaluation of shoulder-line maps from the two test sites of the loess plateau area show obvious improvements in the extraction. The average contour matching distance of the new method is 12.0 m on 5 m resolution DEM, and shows improvement in the accuracy and continuity. The comparisons of accuracy evaluations of the two test sites show that the snake model method performs better in the loess plain area than in the area with high gully density.

DOI

[19]
王珂,王铮,张青峰,等.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[J].测绘学报,2015,44(1):67-75.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。

DOI

[ Wang K, Wang Z, Zhang Q F, et al.Loess shoulder line extraction based on openness and threshold segmentation[J]. Acta Geodaetica et Cartgraphica Ainica, 2015,44(1):67-75. ]

[20]
周毅. 基于DEM的黄土正负地形特征研究[D].南京:南京师范大学,2008.

[ Zhou Y.Investigation of loess positive and negative terrain based on DEMs[D]. Nanjing: Nanjing Normal University, 2008. ]

[21]
周毅,汤国安,王春,等.基于高分辨率DEM的黄土地貌正负地形自动分割技术研究[J].地理科学,2010,30(2):261-266.黄土地貌正负地形自动分割是构 建地表空间分布式机理-过程模型的基础。在分析黄土高原地区典型地貌坡面形态及汇流过程特征的基础上,提出了基于5m分辨率栅格DEM自动分割黄土正、负 地形的技术方案。该方案首先利用坡面上下游栅格点的坡度对比识别沟沿线点,然后利用汇水模型提取沟沿线点约束的上游汇水区域,从而实现正、负地形的自动分 割。在黄土塬区及丘陵沟壑区的实验结果表明,该方法的优点是提取精度高,人工干预少,在不同地貌类型区域内有很好的应用适宜性。

[ Zhou Y, Tang G A, Wang C, et al.Automatic segmentation of loess positive and negative terrain based on high resolution grid DEMs[J]. Scientia Geographica Sinica, 2010,30(2):261-266. ]

[22]
周旻曦,刘永学,李满春,等.多目标珊瑚岛礁地貌遥感信息提取方法——以西沙永乐环礁为例[J].地理研究,2015,34(4):677-690.南海珊瑚岛礁资源极为丰富,实时、快速、高效、准确地获取大范围 珊瑚岛礁地貌遥感信息具有现实意义.研究提出了一种双尺度转化下的模型与数据混合驱动的岛礁地貌信息提取框架,并设计了珊瑚岛礁地貌分类体系及相应技术流 程:首先采用自上而下模型驱动的GVF Snake模型进行宏观地理分带的粗分割,然后采用自下而上数据驱动的云影极值抑制下多阈值OTSU分类算法进行微观地貌类型的精细分类,最终利用区域生 长算法提取离散分布的暗沙、暗滩等浅水地貌单元.针对西沙永乐环礁利用CBERS-02B数据进行实验,精度验证表明:珊瑚岛礁地貌遥感信息提取方法总体 精度优于经典数据驱动的监督分类算法,且具有抗噪能力强、顾及空间拓扑关系、自动灵活等特点.

DOI

[ Zhou M X, Liu Y X, Li M C, et al.Geomorphologic information extraction for multi-objective coral islands from remotely sensed imagery: a case study for Yongle Atoll, South China Sea[J]. Geographical Research, 2015,34(4):677-690. ]

[23]
邓富亮,杨崇俊,曹春香,等.高分辨率影像分割的分形网络演化改进方法[J].地球信息科学学报,2014,16(1):95-101.分形网络演化是针对高分辨遥感 影像的高精度分割方法。它是以像元自下而上进行地物域合并,直至满足区域对象间异质性值大于预设阈值,停止区域合并得到最终分割结果。当对大数据量遥感影 像进行分割时,形成初始区域对象的速度较慢,并且数量较多,导致分割时间长,有待在整体分割效率上进一步提高。一种有效的改进措施是采用某种分割方法,快 速生成初始区域对象,然后再以初始分割结果区域对象进行区域合并。本文提出一种自动种子点的并行区域生长分割方法,用于快速生成初始区域对象;提出均匀数 据划分的并行区域生长策略及消除数据划分线两侧的区域对象方法;采用OpenMP并行技术实现并行区域生长过程。分割效果对比和效率分析结果表明,本文提 出的初始分割方法效率较高,并且分割结果可重现,从可信度、通用性角度来看,具有较高的实用价值。

DOI

[ Deng F L, Yang C J, Cao C X, et al.An improved method of high resolution remote sensing image segmentation[J]. Journal of Geo-information Science, 2014,16(1):95-101. ]

[24]
牟涛,陈文斌,沈一帆.一种融合区域生长与图论的图像分割方法[J].计算机工程与应用, 2005,19:32-34,104.该文提出一种融合区域生长与图论的图像分割方法,一般的基于区域 的分割方法在区域生长完成之后需要进行区域的合并,以消除过分割现象.该文的方法在区域生长完成之后,用Normalized Cut方法在区域之间进行分割,产生最终所分割的图像.在方法上区域生长方法考虑的是图像的局部信息,Normalized Cut方法考虑的是图像的全局信息,该文的方法融合了两者的优点.该文的算法主要以灰度图像为研究对象,实验结果表明可以取得很好的分割效果.

DOI

[ Mou T, Chen W B, Shen Y F. An image segmentation approach by combining region growing and graph theory[J]. Computer Engineering and Applications, 2005,19:32-34,104. ]

[25]
Mirghasemi S, Rayudu R, Mengjie Zhang.A new image segmentation algorithm based on modified seeded region growing and particle swarm[J]. Image and Vision Computing New Zealand, 2013,28:382-387.

[26]
Tuduki Y, Murase K, Izumida M, et al.Automated seeded region growing algorithm for extraction of cerebral blood vessels from magnetic resonance angiographic data[C]. Engineering in Medicine & Biology Society, Proceedings of Annual International Conference of the IEEE, 2000,3:1756-1759.

[27]
张铮,王艳萍,薛桂香.数字图像处理与计算机视觉——Visual C++与Matlab实现[M].北京:人民邮电出版社,2010:368.

[ Zhang Z, Wang Y P, Xue G X, et al.Digital image processing and computer vision—Visual C++ and Matlab implementation[M]. Beijing:People's Posts and Telecommunications Press, 2010:368. ]

[28]
熊礼阳,汤国安,晏实江.基于DEM的山地鞍部点分级提取方法[J].测绘科学,2013,38(2):181-183.本文以地形起伏变化较大的陕北黄土丘陵沟壑区为例,通过流域边界 线上高程相对低点的自动识别,实现基于DEM数据的地面鞍部点自动获取.实验结果显示,采用该方法所提取的地面鞍部点具有层次结构显著、提取精度高、操作 简便、便于确定重要性等级及与山顶点空间关系等优点,对于DEM地形特征点簇的构建与分析具有重要意义.

[ Xiong L Y, Tang G A, Yan S J, et al.Application of regression analysis in monitoring structure settlement[J]. Science of Surveying and Mapping, 2013,38(2):181-183. ]

[29]
Xiong L Y, Tang G A, Yan S J, et al.Landform-oriented flow-routing algorithm for the dual-structure loess terrain based on digital elevation models[J]. Hydrological Processes, 2014,28:1756-1766.ABSTRACT The loess landform in the Loess Plateau of China is with typical dual structure, namely, the upper smooth positive terrain and the lower cliffy negative terrain (P&ndash;N terrain for short). Obvious differences in their morphological feature, geomorphological mechanism, and hydrological process could be found in the both areas. Based on the differences, a flow-routing algorithm that separately addresses the dual-structure terrain would be necessary to encompass this spatial variation in their hydrological behaviour. This paper proposes a mixed flow-routing algorithm to address aforementioned problems. First, the loess landform surface is divided into P&ndash;N terrains based on digital elevation models. Then, specific catchment area is calculated with the new algorithm to simulate the water flows in both positive and negative terrain areas. The mixed algorithm consists of the multiple flow-routing algorithm (multiple-flow direction) for positive areas and the D8 algorithm for negative areas, respectively. The approach is validated in two typical geomorphologic areas with low hills and dense gullies in the northern Shaanxi Loess Plateau. Four indices are used to examine the results, which show that the new algorithm is more suitable for loess terrain in simulating the spatial distribution of water accumulation, as well as in modeling the flow characteristics of the true surface by considering the morphological structures of the terrain. Copyright 漏 2013 John Wiley & Sons, Ltd.

DOI

Options
Outlines

/

〈 〉