【遥感遥测】【2020.05】利用卫星影像进行城市土地利用类型的遥感分类

【遥感遥测】【2020.05】利用卫星影像进行城市土地利用类型的遥感分类_第1张图片

本文为美国旧金山州立大学(作者:Philip James Lynch)的硕士论文,共71页。

全球城市化的加速趋势和随之而来的环境影响使得频繁更新的土地利用和土地覆盖(LULC)地图至关重要。通过对遥感图像的分类,LULC图已经被广泛地制作出来。城市地区的地图是两分类的(城市或非城市),并包含离散的城市类型。

这项研究旨在增强卫星图像的多光谱建成指数,以制定新的城市分类方案。研究的指标有新建成指数(NBI)、建成区提取指数(BAEI)和归一化差异具体条件指数(NDCCI)。利用覆盖美国佛罗里达州迈阿密市的陆地卫星二级数据、佛罗里达地理空间数据库和佛罗里达州环境保护部的地理数据,开发和验证城市区域监督和非监督分类的新方法。通过面向对象的图像分析,NBI可以有效地对城市特征进行分类。BAEI被用于将城市发展视为一种低-高梯度,并对其进行跟踪。NDCCI与NBI和BAEI相结合,作为稳健的城市强度分类方案基础,优于2016年美国地质调查局国家土地覆盖数据库中的城市强度分类方案。将BAEI作为阴影指数引入到一种新的高层建筑填充地质模拟中。研究结果表明,所提出的分类方案有利于建立更详细的土地利用变化图,以应对全球对土地利用变化图日益增长的需求。

An accelerating trend of globalurbanization and subsequent environmental impacts makes frequently updated landuse and land cover (LULC) maps critical. LULC maps have been widely createdthrough classification of remotely sensed imagery. Maps of urban areas havebeen both dichotomous (urban or non-urban) and entailing of discrete urbantypes. This study incorporated multispectral built-up indices designed toenhance satellite imagery to develop new urban classification schemes. Theindices examined are the New Built-up Index (NBI), the Built-up Area ExtractionIndex (BAEI), and the Normalized Difference Concrete Condition Index (NDCCI).Landsat Level-2 data covering the city of Miami, FL, USA was leveraged withgeographic data from the Florida Geospatial Data Library and Florida Departmentof Environmental Protection to develop and validate new methods of supervisedand unsupervised classification of urban area. NBI was found to be useful forclassifying urban features through object-oriented image analysis. BAEI wasutilized to visualize and track urban development as a lowhigh gradient. NDCCIwas composited with NBI and BAEI as the basis for a robust urban intensityclassification scheme superior to that of the urban intensities featured in the2016 USGS National Land Cover Database. BAEI, implemented as a shadow index,was incorporated in a novel infill geosimulation of high-rise construction. Thefindings suggest that the proposed classification schemes are advantageous tothe process of creating more detailed LULC maps in response to the risingglobal demand for them.

  1. 引言
  2. 研究方法
  3. 研究结果
  4. 讨论
  5. 结论

下载地址:

https://089u.com/f/1850492-504201324-798fce

(访问密码:3660)

更多精彩文章请关注公众号:【遥感遥测】【2020.05】利用卫星影像进行城市土地利用类型的遥感分类_第2张图片

你可能感兴趣的:(【遥感遥测】【2020.05】利用卫星影像进行城市土地利用类型的遥感分类)