EN中文

Journal of Geo-information Science >

2016 , Vol. 18 >Issue 4: 544 - 552

DOI: https://doi.org/10.3724/SP.J.1047.2016.00544

Orginal Article

A Measure of Urban Green Index in Urban Areas Based on Moving Window Method

  • WU Jun , 1 ,
  • MENG Qingyan , 1, * ,
  • ZHAN Yulin 1 ,
  • GU Xingfa 1 ,
  • ZHANG Jiahui 1, 2
Expand
  • 1. Institute of Remote Sensing and Digital Earth, Chinese Academy of science, Beijing 100101, China
  • 2. College of Resources and Environment, University of Chinese Academy of Science, Beijing 100049, China
*Corresponding author: MENG Qingyan, E-mail:

Received date: 2015-07-15

  Request revised date: 2015-09-14

  Online published: 2016-04-19

Copyright

《地球信息科学学报》编辑部 所有
Fold

Abstract

Urban green space is an important element of ecological networks. Conservation policies urge the protection of urban green space in cities possessed by dense buildings and focus on the improvement of intercommunity relationships among the green space plots within different areas. Urban green space forms an indispensable part of urban ecosystems. Quantifying the urban green space is of substantial importance for urban planning and development. In order to solve the problem of measuring the proximity of residents to green space (PRG), this paper proposed a method of urban green index (UGI) based on multi-spectral remote sensing data and Lidar data (LiDAR), which provide the accurate information of urban buildings and green spaces. In this study, the urban green index of the central pixel is developed based on calculating the area of urban green space and a fixed window area. The results show that the urban green index can be an effective method to measure the proximity of residents to green space and its spatial distribution characteristics. It avoids the edge effects generated by grid method, overcomes the discontinuity problem occurred in the buffer space, and could obtain the urban green index of any point in the region. At the same time, the newly constructed model can be used to identify the key protection regions that will have typical demonstration effects. The input parameters of the model can be obtained from the remote sensing images, which are easier to obtain than from the field measurements. Therefore, it could be considered to be a new method for studying the urban green space and could offer effective technical assistances to urban planning and management. The quantitative evaluation of the urban green space will provide the scientific support to urban greening and urban environmental protection. This study exploits the advantages of multi-source remote sensing data with high ground resolution in extracting building and vegetation information.

Key words： moving window; urban green index; remote sensing; proximity of residents to green space (PRG)

Cite this article

WU Jun , MENG Qingyan , ZHAN Yulin , GU Xingfa , ZHANG Jiahui . A Measure of Urban Green Index in Urban Areas Based on Moving Window Method[J]. Journal of Geo-information Science, 2016 , 18(4) : 544 -552 . DOI: 10.3724/SP.J.1047.2016.00544

1 引言

城市居民接触城市绿地的概率是衡量居民生活环境质量的一个重要指标。目前利用遥感技术度量城市居民接触绿地概率的方法主要有面积法、格网法、缓冲区法等。面积法是通过计算整个城市内部绿地所占面积比例或人均绿地面积来度量居民接触绿地的概率[1];格网法是将研究区划分为固定大小的均匀格网,计算网格单元中城市绿地参量或进行简单数值组合[2];缓冲区法是不同缓冲区半径构建面向对象度量指数模型[3],如基于单体建筑物尺度的绿度空间指数,以建筑物为中心通过计算每座建筑物周围的绿地参量与建筑物参量的关系来度量。面积法简单但不能描述城市绿度的空间分布;网格法虽考虑了绿地分布,但以网格为单元,未考虑相邻网格的相互影响,边缘效应严重地影响了度量准确性,不能准确地指示居民接触城市绿地的概率;基于建筑物对象的缓冲区法以建筑物为单元,需精确提取建筑物边缘,操作过程复杂,实施效率较低,空间不连续性使其不能度量区域任意点城市绿度。
针对传统方法的不足,本文基于多光谱遥感数据与激光雷达数据(LiDAR),充分利用遥感数据的高空间分辨率优势来描述城市植被的空间分布,在高精度提取城市绿地与建筑物平面信息的基础上,提出一种基于移动窗口的城市绿度遥感度量方法,度量城市居民对周围城市绿地的接触概率。其通过遍历研究区像元,建立移动窗口内的城市绿地面积与固定窗口面积关系,构建一种面向中心像元的绿度指数模型,该绿度指数反映城市居民对周围城市绿地的接触概率及其空间分布特征。研究表明：该绿度指数解决了常规度量方法对城市植被空间分布描述不足的问题,有效地分析了城市任意位置点与其周围绿地的接触概率,能作为城市绿地管理规划工作的有效技术手段,同时该方法避免了格网法产生的边缘效应,且在度量过程中无需提取对象边界,简化度量流程,提升度量过程中的自动化效率。

2 研究区与数据

2.1 研究区与数据源

本文研究区为匈牙利布达佩斯西南65 km处的Transdanubian高地中部的塞克什白堡市(Székesfehérvár)中心城区,经纬度坐标为47°11′23″~47°11′57″N,18°24′23″~18°25′11″E,总面积为1 km2（图1）。研究区内建筑物具有典型的西方建筑特征,即开放的单体空间格局向高空发展。过去10年因经济发展影响,城市规划过程中绿地面积不断减少且分布趋于集中化;而近年来居民生活水平大幅度提高,当地政府为改善城市居民生活环境,在不断优化城市绿地空间布局,主要表现在增加公共绿地面积、注重城市建筑物与绿地景观格局配置等方面。
Fig. 1 Sketch map of the study area

图1 研究区示意图

研究采用德国TopoSys公司于2008年5月30日获取的高空间分辨率多光谱和LiDAR点云数据,机载传感器能够同时获取这2种数据。航拍相机采用RC20模拟框架相机,相机焦距为153 mm。多光谱和LiDAR点云数据的空间分辨率分别为0.5 m和1 m,具体参数如表1所示。
Tab. 1 Parameters of TopoSys

表1 TopoSys系统参数

数据 参数信息 采集日期
TopoSys Inc.
多光谱数据和LiDAR点云
数据		 空间分辨率 RGBI：0.5 m
空间分辨率 LiDAR：1 m
水平精度<±0.5 m
垂直精度<±0.15 m
投影：匈牙利EOV投影
模式：First and Last Echo,RGBI
影像块：4块2000 m×2000 m		 2008-5-30

2.2 数据预处理

利用ERDAS的LPS模块分别对每个多光谱影像块进行正射校正,然后通过镶嵌得到整个研究区的正射校正影像。除了投影定义与图像配准,在预处理阶段还对影像进行了影像裁切、辐射定标、大气校正等预处理,以保证最终测算结果的准确及量纲的规范。
对LiDAR点云数据进行栅格化,得到规则格网数据;进行去噪处理,去除和抑制噪声;采用地形虑波方法生成研究区数字地形模型DTM（Digital Terrain Model）[4]（图2）。利用三维可视化工具观察首次脉冲数据,填补空值区数据,剔除区域异常值,并赋予领域值的方法得到数字地面模型DSM（Digital Surface Model）[5-6]（图3）。
Fig. 2 Digital Terrain Model

图2 数字地形模型(DTM)

Fig. 3 Digital Surface Model

图3 数字地面模型(DSM)

2.3 城市绿地与建筑物信息提取

在高分辨率影像中,利用多光谱遥感影像中的绿色波段与近红外波段构建归一化绿色指数NDGI（Normalized Difference Green Index）,对阴影区非绿地信息具有减弱作用,利用NDGI可有效地区分阴影区的绿地与非绿地信息,计算公式如式（1）所示。
NDVI = ( NIR - RED ) ( NIR + RED ) ≥ α NDGI = ( NIR - GREEN ) ( NIR + GREEN ) ≥ β （1）
式中：α为NDVI的阈值;β为NDGI的阈值;当地物的NDVI大于或等于α,且NDGI大于或等于β时,该地物类型可确认为植被。研究中经过反复试验确定NDVI最终阈值α为0.1,NDGI最终阈值β为0.1。基于NDGI与NDVI指数实现研究区内城市绿地高精度信息提取,提取结果如图4所示。
利用LiDAR数据获取数字表面模型(DSM),即城市地物的高度分布图。基于绿地分布图,利用非绿地分布区对DSM进行掩膜,使用高度阈值法（高度阈值设置为高于3 m的像元为建筑物）从DSM提取得到建筑物分布图。图5为城市绿地与建筑物平面分布图。
Fig. 4 Urban green spaces

图4 城市绿地提取结果

Fig. 5 Urban green spaces and buildings

图5 城市绿地与建筑物分布图

3 城市绿度遥感度量方法

3.1 网格法

网格法一般有简单格网法和复杂格网法2种。简单格网法是对网格单元中城市绿地参量进行简单计算或数值组合,复杂格网法加入建筑物密度、高度及其与绿地的邻接距离等参数作为整个格网单元的城市绿度度量结果,即整个格网单元内城市居民接触城市绿地的概率[7-8]。例如,Schöpfer将研究区划分为100 m×100 m的格网单元,计算格网单元内的城市绿地面积百分比作为该网格单元所覆盖区域的城市绿度指标[9];Gupta通过计算格网单元内的绿地覆盖面积百分比、建筑物覆盖面积百分比、建筑物高度等参数建立不同的指标体系,并计算综合度量结果作为格网单元的城市绿度指标[2]。

3.2 缓冲区法

缓冲区法以城市建筑物为研究对象,以一定缓冲半径向建筑物对象外侧缓冲,计算缓冲区内的城市绿地或建筑物参数来度量建筑物的绿度,即城市建筑物接触城市绿地的概率[10-12]。例如,李小江以建筑物为对象建立向外缓冲区,分别计算缓冲区内绿地面积、周长、叶面积指数及侧面积、体积等参量构建建筑物的绿度度量指标[13];刘玉琴以建筑物为研究对象建立单体建筑物缓冲区,计算缓冲区中绿地面积、不同绿地类型缓冲区面积、建筑物面积百分比、高建筑物面积百分比构建建筑物绿度度量指数模型[14]。

3.3 移动窗口法

移动窗口法最早被用于分析沿水份梯度植被的变化和城市化过程对植被分布和生态特征的影响研究过程中[15]。目前移动窗口法仍然被广泛应用于城市景观格局和人居环境适应性研究过程中[16-17]。本文提出的移动窗口法是基于城市绿地分布图,以每个像元为中心,构建N像元×N像元（N为奇数）的移动窗口,计算窗口内绿地总面积与窗口总面积的比值,将该比值赋给窗口中心像元,作为窗口中心像元的城市绿度指数（Urban Green Index,UGI）,即城市居民在窗口中心像元覆盖区域所接触周围城市绿地的概率（图6）。当中心像元周围城市绿地覆盖度较大时,中心像元的城市绿度指数（UGI）也相应较大,该区域城市居民接触城市绿地的概率相应较大,反之,则较小,计算公式如式（2）所示。遍历每个像元,将研究区遥感影像数据的每个像元依次作为中心像元,同时构建相同大小窗口得到研究区像元绿度指数分布图;对建筑物分布图进行掩膜,与像元绿度指数分布图叠加,得到城市居民长期停留的建筑物覆盖区域的城市绿度指数。
UGI i ， j = Ve g area i ， j / Windo w area （2）
式中： UGI i , j 表示位置为 i , j 的像元的绿度指数,表示研究区内该像元周围接触城市绿地的概率;UGI值越大,表明中心像元周围城市绿地的覆盖度越高,该区域城市居民接触城市绿地的概率也越大,反之,则越小。 Ve g area i , j 表示以像元 i , j 为中心的N×N（N为像元数,取奇数）窗口内城市绿地的总面积; Windo w area 表示以像元 i , j 为中心的N×N窗口的总面积。
Fig. 6 Moving window diagram

图6 移动窗口示意图

4 城市绿度的移动窗口度量结果与分析

4.1 度量结果

采用移动窗口法对研究区城市绿度进行度量时,充分考虑研究区地物特征及空间配置紧密程度,结合当地居民对接触城市绿地距离的期望值,并参考城市绿地对周围环境的积极作用在15~25 m范围的研究成果[18],最终确定以中心像元周围25 m范围内的城市绿地作为其潜在接触对象,即移动窗口的N=101,边长L=50.5 m。图7(a)为得到的研究区绿度指数UGI的平面分布图;图7(b)为利用建筑物分布图掩膜后得到的城市建筑物绿度指数,表示研究区内居民长期所处地建筑物的每一处接触城市绿地的概率。
Fig. 7 Urban green index

图7 城市绿度遥感度量指数分布图

采用格网法进行绿度指数计算时具体使用Schöpfer的面积比值法[9]。为了使度量结果与移动窗口法度量结果具有可比性,保持输入尺度一致,设定网格边长D/2=25 m,即将研究区划分为50 m× 50 m大小网格单元。根据研究区植被分布图计算每个网格单元内的城市绿地的总面积与格网单元面积的比值,将该比值作为格网单元的属性值,即格网单元的城市绿度指标,表示格网单元区域内居民接触城市绿地的概率。图7(c)为该方法得到的研究区绿度指数分布图。
采用缓冲区法度量研究区城市绿度指数时选择李小江的面向建筑物尺度的城市绿度指数构建模型[13],具体过程为利用多光谱遥感数据和LiDAR点云数据实现城市绿地与建筑物的粗提取,然后对利用sobel算子得到的建筑物边缘图像进行重建,利用分水岭法进行图像分割,最后基于OTSU算法和投票法则得到高精度建筑物边界,获取研究区建筑物对象分布图,对研究区内所有建筑物分别以每个单体建筑物为分析对象,构建缓冲半径R=25 m的缓冲区,计算缓冲区内部城市绿地面积占缓冲区面积的比值作为该建筑物对象的绿度指数。图7(d)为该方法得到的研究区建筑物绿度指数分布图。

4.2 度量结果分析

4.2.1 方法验证
为验证移动窗口法在度量城市居民在活动区域范围内接触周围绿地概率的真实性与适用性,根据区域类型和功能划分出研究区内部的商业区(A)、居民区(B)、文化区(C)3个子功能区（图8）,分别统计3种方法在不同功能区的度量结果,统计结果如表2所示。
Fig. 8 Three functional domains in the study area

图8 研究区域功能区分布图

Tab. 2 Statistical table of the three methods

表2 3种方法功能区度量结果统计表

特征值 研究区 A区 B区 C区
格网法 中值 0.35 0.15 0.37 0.32
平均值 0.32 0.19 0.35 0.31
标准差 0.17 0.18 0.16 0.17
缓冲区法 中值 0.44 0.21 0.36 0.35
平均值 0.29 0.17 0.37 0.36
标准差 0.21 0.19 0.15 0.17
移动窗口法 中值 0.36 0.15 0.36 0.37
平均值 0.31 0.12 0.35 0.33
标准差 0.23 0.19 0.14 0.15
由表2可见,3种方法得到的度量结果比较接近,整体研究区绿度平均值分别为0.32、0.29、0.31。且居民区(B)的平均绿度（0.35、0.37、0.35）略高于文化区(C)平均绿度（0.31、0.36、0.33）,远高于商业区(A)平均绿度（0.19、0.17、0.12）,即居民区(B)内城市居民接触绿地的概率稍高于文化区(C)内城市居民接触绿地的概率,而远高于商业区(A)城市居民接触绿地的概率。这是由于当地城市规划使得商业区内建筑物面积远大于居民区与文化区,居民区绿化面积远大于商业区而略优于文化区。
移动窗口法的标准差（0.23）大于格网法标准差（0.17）和缓冲区法标准差（0.21）,说明移动窗口法度量结果内部差异性更大,对城市绿地空间分布特征区分性更好。在3个子功能区内部,在商业区3种方法度量结果的标准差相近（分别为0.18、0.19、0.19）,而在居民区和文化区3种方法度量结果的标准差表明,移动窗口法度量结果标准差（B区0.14、C区0.15）均小于格网法度量结果标准差（B区0.16、C区0.17）和缓冲区法度量结果标准差（B区0.15、C区0.17）,说明移动窗口法对功能区内部城市居民接触周围绿地概率的区分性更小。所以,相对于格网法指数模型和缓冲区法指数模型,基于移动窗口法的城市绿度指数模型的敏感性更好,该指数能更好地反映研究区整体绿地分布差异,同时削弱功能区内部差异,更适用于区分城市内部功能区之间绿度景观格局合理性及比较分析。
为比较3种方法得到的城市绿度指数UGI的分布特点,对3种方法的最终度量结果进行分级统计分析：设定对于格网法绿度指数（GUGI）、缓冲区绿度指数（BUGI）和移动窗口法绿度指数（MUGI）在[0,0.25）范围内时,表示研究单元内居民接触城市绿地的概率很低;在[0.25,0.50）范围内时,表示研究单元内居民接触城市绿地的概率一般;在[0.50,0.75）范围内时,表示研究单元内居民接触城市绿地的概率适中;在[0.75,1.00）范围内时,表示研究单元内居民接触城市绿地的概率很高。具体统计结果如表3所示。
Tab. 3 Spatial distribution characteristics of urban green index (%)

表3 3种绿度指数空间分布特征(%)

研究区 A区 B区 C区
范围 GUGI BUGI MUGI GUGI BUGI MUGI GUGI BUGI MUGI GUGI BUGI MUGI
[0,0.25) 31.2 33.1 31.4 44.2 39.2 47.3 14.6 19.2 26.6 17.1 20.3 28.9
[0.25,0.50) 36.3 35.3 31.6 34.1 37.3 26.5 40.7 32.3 34.2 41.3 34.4 37.4
[0.50,0.75) 23.8 23.7 25.4 17.1 19.6 17.2 29.8 39.4 29.4 31. 33.9 20.6
[0.75,1.00) 8.7 7.9 11.6 4.6 3.9 9.0 14.9 9.1 9.8 10.2 12.4 13.1
总计 100 100 100 100 100 100 100 100 100 100 100 100
由表3可见,在整个研究区范围内3种方法的度量结果在空间分布上(低、一般、中、高)呈现出一致性。3种绿度指数均显示研究区范围内城市居民接触周围绿地的概率在一般标准的区域面积最大（36.3%、35.3%、31.6%）,概率在低标准的次之（31.2%、33.1%、31.4%）,概率在适中标准（23.8%、23.7%、25.4%）的再次之,而概率在高标准的面积最少（8.7%、7.9%、11.6%）。而在3个子功能区范围内分布特征不一致:在商业区(A),绿度指数在低标准的区域面积最大,其中移动窗口法绿度指数为47.3%,说明该区域内城市居民在接近1/2的活动范围接触绿地的概率低于0.25,其次面积占优的是一般标准,这是由于商业区建筑物密集、绿地面积少而且过度集中;而在居民区(B)和文化区(C),因为绿地总量较多且分布较为均匀,整体规划效果较好,绿度指数在一般标准的区域面积最大,均在30%~40%之间,其次面积占优的是适中标准。此外3个子功能区内不存在像城市公园类型的较大绿地斑块,故居民接触城市绿地概率高于0.75的区域面积普遍低于10%。
由表2、表3可见,移动窗口法与其他2种方法均在一定程度上很好地实现了度量城市居民在活动区域范围内接触城市绿地的概率,反映城市绿地的空间分布特征。
4.2.2 样区分析
为进一步分析3种方法在度量城市居民在活动区域范围内接触城市绿地概率的适用性与真实性,综合考虑研究区内城市地物类型及空间配置关系,选择典型样区,并在典型样区内选择A、B、C、D 4个样点,基于典型样区与样点对3种方法度量结果的准确性进行系统分析。
图9中,A点为绿地覆盖点位于公园内部;B点为非绿地和建筑物覆盖点,位于交叉路口处;C点为建筑物覆盖点,位于靠近公园和路口的建筑物拐角处;D点为建筑物覆盖点,位于远离公园的建筑物拐角处且C、D点位于同一建筑物覆盖区。
Tab. 4 Parse list of the sample region

表4 样区分析表

位置点 格网法 移动窗口法 缓冲区法 备注
A 0.67 0.73 - A、B点在格网法中处于同一格网中;C、D属于同一建筑物2个点;dAB> dBC
B 0.67 0.52 -
C 0.46 0.43 0.39
D 0.05 0.17 0.39
Fig. 9 Sample region

图9 样区示例图

从样区示例图（图9）与分析表（表4）可得出：
(1) A、B点为非建筑物点,面向建筑物对象的基于缓冲法的城市绿度度量方法无法有效地度量非建筑物点（如A、B点）周围接触城市绿地的概率。而在实际应用中,需充分考虑该种情况,移动窗口法可实现对研究区所有位置点的城市绿度度量。
(2) C、D属于同一建筑物的2个点,但当这2个点距离较远时,由于周边的绿地环境不一样,C、D点周围接触城市绿地的概率理论上应该是不同的,但在采用面向对象的缓冲区法时,2个点绿度度量结果相同,与实际情况存在差距,而移动窗口法可以有效地区分2点周围接触城市绿地概率的差距。
(3) A、B点为非建筑物点且具有较大距离,从实际遥感影像中也可得出2个点周围接触城市绿地的概率是不一样的,但在格网法中A、B点被作为同一研究对象,度量结果相同,与实际情况不符,移动窗口法以像元为研究对象,可以有效避免该种现象。
(4) A、B点之间的距离远大于B、C点之间的距离,即dAB> dBC,实际上A、B点接触城市绿度的概率的差距应远大于B、C点之间接触城市绿度的概率的差距,即∣UGIA-UGIB∣>∣UGIC-UGIB∣,但在格网法中A、B点接触城市绿地的概率相同,B、C接触城市绿地的概率差距却很大,这与实际情况存在较大差距,而采用移动窗口法则不会出现该问题,从度量结果可看出移动窗口法度量结果更符合实际情况。
4.2.3 应用分析
现阶段城市绿度空间建设的主观性强,缺乏科学系统的分析,对城市绿地进行定量化分析能及时准确地辨识城市绿地服务水平不足的区域,为城市绿地的合理布局和结构优化提供依据[19]。传统的格网法和缓冲区法以及本文提出的移动窗口法,能在一定程度上衡量城市居民接触周围城市绿地的概率,其构建起的绿度指数能有效地分析研究区内城市绿地分布格局的合理程度,准确地判断城市绿地资源配置效率的高低。在整个研究过程中,移动窗口法运算耗时约1.02 h,而格网法和缓冲区法分别耗时1.2 h和3.7 h,移动窗口法的数据处理效率明显高于格网法和缓冲区法,将3种方法应用到面积更为广泛的本文研究区所在的城市塞克什白堡市（Székesfehérvár）的主要城区,应用示范区总面积49 km2,根据统计分析结果：移动窗口法的运算速度比格网法快12%,比缓冲区法快73%。
基于移动窗口的城市绿度遥感度量方法为城市绿地空间统计分析提供了一种新的视角,对城市绿地景观格局的合理性评价和城市绿地生态系统服务研究具有指导意义,开拓了城市景观生态学研究思路。同时,作为衡量城市居民绿地环境的重要参量,城市绿度指数可作为开展城市绿地规划管理工作的重要科学依据。但移动窗口法也存在一些局限,如靠近研究边缘区域的像元受限于移动窗口边长的大小,不能构成完整的移动窗口无法评价,在被大量建筑物等不透水面覆盖区域其中心像元的城市绿度指数会出现0值等,从而使移动窗口法度量结果存在一定误差。

5 结论

本文利用多光谱遥感数据和LiDAR点云数据构建基于格网、缓冲区和移动窗口的城市绿度指数模型。基于格网和缓冲区的城市绿度遥感度量方法能实现度量研究区城市居民接触城市绿地概率。但网格法以网格为单元,未充分考虑相邻网格的相互影响,在相邻格网单元分界线处,两格网单元边界处像元差异性被放大,边缘效应严重影响度量准确性,不能准确地指示居民接触城市绿地的概率;基于缓冲区法建立的面向建筑物尺度的绿度指数以建筑物为单元,实施过程中需精确提取建筑物边缘,数据源要求高,操作过程复杂,仅面向建筑物对象的空间不连续性使其不能度量区域任意点城市绿度,不能满足城市居民实际评估自身生活环境质量的要求。以二者为主的传统方法开拓了遥感技术在度量城市绿度研究过程中运用,为城市规划、城市绿化的科学依据指导拓展了思路,但相应存在的局限也限制了其在城市管理规划中的应用。
基于移动窗口的城市绿度遥感度量方法能很好地度量城市居民在活动区域范围内接触城市绿地的概率,反映城市绿地的空间分布特征。在度量过程中能客观度量城市每个角落实际接触城市绿地的概率,以及这种概率的空间分布特征,有效地避免了面向对象法只能针对建筑物或绿地对象等开展度量的局限,是一种连续性较强的度量方法,充分考虑城市居民活动范围,分析城市的任何一处与其周围绿地接触的概率,能更客观地度量城市的哪个区域适宜于居民生活;能更为精确地区分不同位置实际接触城市绿地的概率,有效解决了格网法边缘效应产生度量结果的准确性,提高了利用遥感技术度量研究区内接触周围城市绿地概率的精度;同时,基于移动窗口的城市绿度遥感度量方法操作简单,数据源要求低,降低了现有度量方法的复杂性,运算效率高,有利于计算机自动化操作,便于推广应用。

The authors have declared that no competing interests exist.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
荣冰凌,陈春娣,邓红兵.城市绿色空间综合评价指标体系构建及应用[J].城市环境与城市生态,2009,22(1):33-37.为科学合理地评价城市绿色空间的生态景观效益、社会效益以及经济效益,建立了一套较为科学、全面的评价指标体系,其包括基本数量特征、景观格局、社会管理因素和生态功能等4个Ⅱ级指标以及16个Ⅲ级指标,并依次对北京城市绿色空间现状进行了综合评价。结果表明:北京市绿色空间综合评价指标值为0.67,处于第Ⅱ级下游,表明北京城市绿色空间面积适当、结构较合理,生态功能发挥良好,总体建设较好;但同时也存在景观连接度较差、可达性不好、居民满意度不高等问题,从而影响了其生态功能和社会功能的有效发挥。

[ Rong B L, Chen C D, Deng H B.Establishment and application of integrated evaluation index system of urban green space[J]. Urban Environment &Urban Ecology, 2009,22(1):33-37. ]

[2]
Gupta K, Kumar P, Pathan S K, et al.Urban neighborhood green index-A measure of green spaces in urban areas[J]. Landscape & Urban Planning, 2012,105(3):325-335.Urban green spaces (UGS) form an integral part of any urban area and quantity and quality of UGS is of prime concern for planners and city administrators. Objective measure of greenness using remote sensing images is percentage area of green, i.e., Green Index (GI), which is insensitive to spatial arrangement within the areal units. Measuring UGS at neighborhood level is important as neighborhood is the working level for application of greening strategies. Neighborhood (NH) is synonymous of nearness and can be defined as an area of homogeneous characteristics. The Urban Neighborhood Green Index (UNGI) aims to assess the greenness and can help in identifying the critical areas, which in turn can be used to identify action areas for improving the quality of green. For the development of UNGI, four parameters, i.e., GI, proximity to green, built up density and height of structures were used and weighted using Saaty's pair wise comparison method. Four different types of NH were compared and it was found that mean GI (0.44) is equal for high-rise low density and low-rise low density NH, i.e., both areas have same quality of urban green based on GI. But mean UNGI is higher for low-rise low-density NH (0.62), as compared to high-rise low-density NH (0.54), hence, area of highrise NH requires more amounts of good quality properly distributed green as compared to low-rise NH. The input for UNGI is easily derivable from RS images, besides the developed method is simple, and easily comprehendible by city administrators and planners.

DOI

[3]
Pirnat H J.Urban forestry-linking naturalness and amenity: the case of Ljubljana, Slovenia[J]. Urban Forestry & Urban Greening, 2011,10(2):105-112.ABSTRACT Urban forestry tries to connect the concept of forest ecosystem with the idea of nature preservation as well as with social needs of urban society, often living close to such a natural environment, thus being a part of it. The ability to connect the human need for the natural environment with life support system of persistent forest ecosystem should become an important contributing factor of urban forestry for well-being of urban society.In the range of 1km walking distance around the city of Ljubljana and within urban areas of the city, forests comprise more than one quarter of the area. The most prevalent in the area are patches of forest fragments, estimated to have a shallow depth of the inner environment. The concept of urban forests in the Ljubljana Green System as a backbone of a green network has been confirmed especially by the two largest forest patches of Golovec and Ro啪nik, indented as a part of forest matrix up to the city centre. Based on the analysis of the continuum of these forests, it was possible to assume that in the past, more than three quarters of forest grounds were meant for no other land use. In the other quarter of the city and its Green System, forests were more significantly reduced in the past, given the fact that we can estimate its continuum only for one half of today's area.On a detailed level, it is recommended for the urban forests to contain a significant percentage of trees and other species that would naturally appear on the sites of the area. Whereas on a landscape level, it is not enough to address only such species, but also temporal stability of forest patches, corridors or even forest matrix in urban areas and its urban-rural gradient into suburban areas.

DOI

[4]
Vosselman G.Slope based filtering of laser altimetry data[J]. International Archives of Photogrammetry and Remote Sensing, 2000,33(4):203-210.ABSTRACT Laser altimetry is becoming the prime method for large scale acquisition of height data. Although laser altimetry is full integrated into processes for the production of digital elevation models in different countries, the derivation of DEM's from the raw laser altimetry measurements still causes problems. In particular the laser pulses reflected on the ground surface need to be distinguished from those reflecting on buildings and vegetation. In this paper a new method is proposed for filtering laser data. This method is closely related to the erosion operator used for mathematical grey scale morphology. Based on height differences in a representative training dataset, filter functions are derived that either preserve important terrain characteristics or minimise the number of classification errors. In experiments it is shown that the latter filter causes smaller errors in the resulting digital elevation models. In general the performance of the filters deteriorates with a decreasing point density.

[5]
汪承义. 基于航空激光雷达数据的建筑物重建技术研究[D].北京:中国科学院遥感应用研究所,2007.

[ Wang C Y.Airborne LiDAR point cloud data processing and 3D building reconstruction[D]. Beijing: Institute of Remote Sensing Application Chinese Academy of Sciences, 2007. ]

[6]
潘慧波,胡友健,王大莹.从LiDAR数据中获取DSM生成真正射影像[J].测绘工程,2009,18(3):47-50.传统的正射影像在进行数字纠正时通常是将地物纠正在DTM上,致 使城区造成屋顶和树木的坐标投影差.若利用LiDAR生成的密集DSM则可以制作真正射影像.阐述LiDAR技术的工作原理和系统组成以及真正射影像的制 作原理,详细介绍从LiDAR数据中获取DSM,再结合同步采集的数码影像进行正射纠正生成真正射影像的可行性方法.

DOI

[ Pan H B, Hu Y J, Wang D Y.DSM extraction from LiDAR data and its application to processing true-ortho images[J]. Engineering of Surveying and Mapping, 2009,18(3):47-50. ]

[7]
Hofmann P, Strobl J, Nazarkulova. A mapping green spaces in Bishkek—how reliable can spatial analysis be?[J]. Remote Sensing, 2011,3(6):1088-1103.Within urban areas, green spaces play a critically important role in the quality of life. They have remarkable impact on the local microclimate and the regional climate of the city. Quantifying the 'greenness' of urban areas allows comparing urban areas at several levels, as well as monitoring the evolution of green spaces in urban areas, thus serving as a tool for urban and developmental planning. Different categories of vegetation have different impacts on recreation potential and microclimate, as well as on the individual perception of green spaces. However, when quantifying the 'greenness' of urban areas the reliability of the underlying information is important in order to qualify analysis results. The reliability of geo-information derived from remote sensing data is usually assessed by ground truth validation or by comparison with other reference data. When applying methods of object based image analysis (OBIA) and fuzzy classification, the degrees of fuzzy membership per object in general describe to what degree an object fits (prototypical) class descriptions. Thus, analyzing the fuzzy membership degrees can contribute to the estimation of reliability and stability of classification results, even when no reference data are available. This paper presents an object based method using fuzzy class assignments to outline and classify three different classes of vegetation from GeoEye imagery. The classification result, its reliability and stability are evaluated using the reference-free parameters Best Classification Result and Classification Stability as introduced by Benz et al. in 2004 and implemented in the software package eCognition (www.ecognition.com). To demonstrate the application potentials of results a scenario for quantifying urban 'greenness' is presented.

DOI

[8]
Moeller M S, Blaschke T.A new index for the differentiation of vegetation fractions in urban neighborhoods based on satellite imagery[J]. The imaging and geospatial information society, 2006,39(11):1-5.ABSTRACT Export Date: 15 July 2014, References: Baatz, M., Schaepe, A., Multiresolution segmentation: An optimization approach for high quality multiscale image segmentation (2000) Angewandte Geographische Informationsverarbeitung XII. Beitr&auml;ge Zum AGIT-Symposium Salzburg 2000, pp. 12-23. , Strobl, J. et al. (Edts.), Karlsruhe, Wichmann;

[9]
Schöpfer E, Lang S, Blaschke T.A "Green Index" incorporating remote sensing and citizen's perception of green space[J]. International Archives of Photogramm, Remote Sensing and Spatial Information Sciences, 2005,27(6):1-6.

[10]
CuttsB, Darby K J, Boone C G, et al. City structure, obesity, and environmental justice: an integrated analysis of physical and social barriers to walkable streets and park access[J]. Social Science & Medicine, 2009,69(9):1314-1322.Local parks and walkable neighborhoods are commonly cited as elements of the urban environment that promote physical activity and reduce obesity risk. When those vulnerable to obesity-related diseases live in neighborhoods without these qualities, it works against environmental justice goals that aim for a fair distribution of amenities. We use geographic information systems (GIS) to evaluate the relationship between the distribution of populations vulnerable to obesity and proximity to parks and walkable street networks in Phoenix, Arizona, USA. Though previous studies have used GIS to assess the distribution of access to opportunities for physical activity, none have analyzed access to both parks and walkable resources at once. Neither have they included data that reflects findings on a smaller scale indicating that perceptions of resource quality, safety, and cultural relevance also affect physical activity levels. We include these safety and quality factors in our study through statistical data on traffic fatalities, crime rates and park size. We find that, counter to predictions, subpopulations generally considered vulnerable to obesity (and environmental injustices more generally) are more likely to live in walkable neighborhoods and have better walking access to neighborhood parks than other groups in Phoenix. However, crime is highest in walkable neighborhoods with large Latino/a and African-American populations and parks are smaller in areas populated by Latino/as. Given the higher prevalence of obesity and related diseases in lower income and minority populations in Phoenix, the results suggest that benefits of built environments may be offset by social characteristics. Our most consistent finding indicates a strong negative relationship between the percentage of the population under 18 years of age living in an area and the likelihood that the structure of the built environment supports physical activity. Children under 18 are significantly underrepresented in regions deemed highly walkable and those with access to parks.

DOI PMID

[11]
Niemela J, Saarela S R, Soderman T, et al.Using the ecosystem services approach for better planning and conservation of urban green spaces: a Finland case study[J]. Biodiversity & Conservation, 2010,19(11):3225-3243.Ecosystem services are vital for humans in urban regions. However, urban development poses a great risk for the ability of ecosystems to provide these services. In this paper we first address the most important ecosystem services in functional urban regions in Finland. Well accessible and good quality recreational ecosystem services, for example, provided by urban nature, are an important part of a high-quality living environment and important for public health. Vegetation of urban regions can have a role in carbon dioxide sequestration and thus in climate change mitigation. For instance, estimates of carbon sinks can be compared to total COemissions of an urban region, and the municipality can aim at both increasing carbon sinks and decreasing COemissions with proper land-use planning. Large and contiguous core nature areas, smaller green areas and ecological connections between them are the essence of regional ecological networks and are essential for maintaining interconnected habitats for species and thus biological diversity. Thus, both local and regional level ecological networks are vital for maintaining ecosystem services in urban regions. The impacts of climate change coupled with land-use and land cover change will bring serious challenges for maintaining ecosystem services in urban areas. Although not yet widely used in planning practices, the ecosystem services approach can provide an opportunity for land-use planning to develop ecologically sustainable urban regions. Currently, information on ecosystem services of urban regions is lacking and there is a need to improve the knowledge base for land-use planning.

DOI

[12]
Nazarkulova A, Strobl J.Green spaces in Bishkek-A satellite perspective[C]. Proceedings of the Fourth Central Asia GIS Conference, 2010,10(5):32-41.

[13]
李小江. 基于多源遥感数据的城市绿度空间指数研究[D].北京:中国科学院大学,2013.

[ Li X J.Research on urban green index base on multi-source remote sensing data. Beijing: University of Chinese Academy of science, 2013.

[14]
刘玉琴. 基于建筑物尺度的城市绿度空间遥感建模研究[D].福州:福州师范大学,2014.

[ Liu Y Q.Study on urban green space based on built up scale using high resolution satellite data[D]. Fuzhou: Fujian Normal University. 2014. ]

[15]
Whittaker R H, Niering W A.Vegetation of the Santa Catalina mountains, Arizona. V. Biomass, production, and diversity along the elevation gradient[J]. Ecological Society of America, 1975,56(4):771-790.This item is part of the Arizona Land and People (formerly Progressive Agriculture in Arizona) archive. It was digitized from a physical copy provided by the College of Agriculture and Life Sciences at The University of Arizona. For more information about this periodical, please email CALS Publications at pubs@ag.arizona.edu.

[16]
谢晓议,曾咺,李军.基于移动窗口法和栅格数据的重庆市人居环境自然适宜性评价[J].长江流域资源与环境,2014,23(10):1351.以GIS为主要技术平台,采用 移动窗口法,以250m×250m栅格为基本单元,通过选择地形、气候、水文、植被等自然因子,基于人居环境指数的自然适宜性评价模型,对重庆市人居环境 自然适宜性进行相关研究,并揭示其地理特征和空间格局。结果表明:(1)重庆市人居环境自然适宜度介于0.255~0.821,西部高于东部,北部高于南 部;(2)临界适宜区面积最大,占全市面积的34.53%;低度适宜区次之,占22.52%;高度适宜区居第三,占21.14%;中度适宜区占 17.84%;不适宜区面积最小,仅占3.96%;(3)重庆市85.19%的人口分布在低度适宜及其以上的区域,相应面积占研究区面积的61.5%。实 证表明,该研究结果可为重庆市人口空间布局的合理性、社会经济的可持续发展和方针政策的构建提供科学依据和重要参考。

DOI

[ Xie X Y, Zeng X, Li J.Evalution of nature suitability for human settlement in ChongQing[J]. Resources and Environment in the Yangtze Basin,2014,23(10):1351. ]

[17]
刘昕,国庆喜.基于移动窗口法的中国东北地区景观格局[J].应用生态学报,2009(6):1415-1422.基于GIS技术,利用移动窗口法,对2006年中国东北地区景观格局特征及其与水分梯度、热 量梯度、海拔因子和人为干扰因子的相关关系进行了研究.结果表明：2006年,中国东北地区林地景观所占比例最大,为61.69%,耕地次之,占 25.11%;中国东北地区的景观多样性具有圈层结构的特点,为破碎化程度较高的敏感区域提供了缓冲区间,使不良的生态后果可以控制在一定区域内;研究区 景观格局指数与水分梯度、热量梯度的相关系数均在0.4以下,与海拔因子的相关系数在0.07以下,相关性不高说明研究区景观格局的异质性不是由单一的自 然因素所决定.

[ Liu X, Guo Q X.Landscape pattern in Northeast China based on moving window method[J]. Chinese Journal of Applied Ecology, 2009,6:1415-1422. ]

[18]
蔺银鼎,韩学孟,武小刚,等.城市绿地空间结构对绿地生态场的影响[J].生态学报,2006,26(10):3339-3346.国内外对城市绿地生态效应的研究主要有以下3个特点:一是基于 GIS技术,在中尺度上(一般以一个城市作为研究对象)开展城市绿地与城市气候的相关分析;二是以绿地斑块为单位,观测比较不同结构绿地内部的小气候效 应;三是基于植物蒸腾理论,通过计算绿色生长量估测不同结构绿地内的小气候效应.对自然生态系统的研究结果表明,城市绿地作为一个开放的生物系统,必将通 过系统交换对绿地周围的环境产生影响.绿地与非绿地空间的系统交换过程不仅仅与植物的叶面积指数有关,还要受到绿地斑块的大小、几何形状、植物类型、生长 密度与高度及周边环境和气候条件等因素的影响.尽管植物生态场理论的研究侧重对植物群落中个体的空间作用,尤其是相邻植物竞争过程的分析,但其理论构架和 计算方法有可能为城市绿地生态效应的研究开辟一条新的路径. 试验者在太原市区选择了6个不同空间结构的绿地样地.使用温湿度记录仪观测了绿地周边的温湿度变化,并利用植物生态场理论作了分析.提出用生态场强、场梯 度和场幅作为城市绿地生态效应的主要评价指标.其中,场强用绿地内侧5m处的测试数据与对照的差值来表示.其含义为绿地对周边环境温度或湿度的干扰强度. 场梯度是指相邻两个数据的差值.场幅即场影响范围. 结果表明,绿地面积、林分和生长量等绿地空间结构因子对绿地的生态场特征都不同程度地产生影响.在其他结构因子相近或相同的条件下,当绿地面积达到一定 时,随着面积的进一步增加,绿地降温和增湿的幅度(场幅)有降低的趋势.与片林相比,草坪的增湿效果好于降温效果.分析结果显示,利用生态场理论能够更好 地描述城市不同空间结构绿地的生态效应及其差异.

DOI

[ Lin Y D, Han X M, Wu X Y.Ecological field characteristic of green land based on urban green space structure[J]. Acta Ecologica Sinica, 2006,26(10):3339-3346. ]

[19]
陶宇,李锋,王如松,等.城市绿色空间格局的定量化方法研究进展[J].生态学报,2013,33(8):2330-2342.城市绿色空间格局的变化是影响 城市生态系统社会、经济与生态功能的重要因素。在分析城市绿地数量和结构时空动态变化的基础上,重点综述了城市绿地斑块和廊道连接的景观格局指数法和网络 分析方法,探讨了城市绿地与居住用地的空间交互作用以及可达性分析方法,比较了城市绿地沿城乡分布的梯度分析方法。并总结了城市绿色空间格局研究的热点领 域,包括城市绿色空间格局的空间显式表征和多尺度分析,以及格局的定量研究与规划的结合,并应用于生态系统服务的评价。

DOI

[ Tao Y, Li F, Wang R S.Research progress in the quantitative methods of urban green space patterns. Acta Ecologica Sinica, 2013,33(8):2330-2342. ]

Options
Outlines

/

〈 〉