节点文献

基于SAR图像的城市形态时空变化的研究

Study on Temporal and Spatial Evolution of Urban Pattern Based on SAR Images

【作者】 黄玉琴

【导师】 郭华东;

【作者基本信息】 中国科学院研究生院(遥感应用研究所) , 地图学与地理信息系统, 2006, 博士

【副题名】以北京市为例

【摘要】 城市形态是城市这一复杂巨系统中的重要体现,定量研究城市形态对城市科学管理和规划具有重要的实践意义。 本文首先对城市形态的研究作了回顾,了解国内外在这一领域的研究成果和进展状况。其次讨论了研究城市形态的手段和工具随着信息时代的到来发生很大的变化,遥感技术为城市研究提供了重要的数据源,GIS为城市研究提供了强有力的工具。这两者的结合为城市化研究带来新的课题。在此基础上,本文利用遥感技术和地理信息系统,从SAR图像中提取了城市建筑覆盖区专题信息,为进一步研究城市形态提供了数据支持。 早期由于受到分辨率和成像系统的影响,雷达在城市探测应用方面的研究比较少,成像雷达技术的出现为城市遥感带来新的应用潜力。大量机载SAR的高质量图像改变了人们对雷达遥感空间分辨率不高的旧观念,而雷达的全天候工作的特点更使人刮目相看。最近几年,雷达在探测城市方面的应用逐渐增加。 微波信号对城市建筑区比较敏感,可以识别出城市和郊区的边缘,SAR图像对城市的研究是对光学图像的一个重要的补充,而用SAR图像对城市形态的研究目前在国内外还尚未多见,因此本文的研究具有一定的前沿性。 本文基于以上的研究状况,针对SAR图像中提取的城市形态,以北京市为例对自1995年到2002年间的城市形态随时间在空间上的变化情况进行了分析。从对微波信号对城市典型地物的成像机理进行分析后,利用纹理信息和ART-Ⅱ神经网络开展了对SAR图像的分类的研究,并提取出城市建筑覆盖区专题信息。利用提取的城市建筑覆盖区的专题信息,借助GIS工具和分形学理论对北京市的城市形态进行了深入的研究和分析,从两个方面研究了城市形态的变化特征,一是研究城市形态的自身的演化特征,将城市区域分为四个方向进行研究,分析了城市扩展方向和在每个方向的演化特点和规律。通过对城市形态的度量,研究了城市形态的分形特征,说明城市形态随着时间变化,其复杂性和稳定性的变化情况;二是分析城市形态变化规律,包括城市扩展方向和扩展规模的差异程度分析。城市的扩展方向分四个方向进行研究,对每个方向扩展的差异程度进行了分析。

【Abstract】 Beijing, the capital of our great country, is one of metropolis in the world. The velocity of urbanization is very rapid since 1990’. The urbanization has brought a series of problems such as environment deterioration, aerial land reduction and so on. And the urban pattern is changed with the urbanization.A means for a better understanding of urban areas is remote sensing techniques. Satellite imagery can significantly improve the monitoring of cities in a wide range of applications, e.g., the detection of urban changing, mapping road and streets, mapping urban demarcation, and mapping the urban pattern.Nevertheless the number of studies using SAR data is much less than that using optical ones notably concerning investigation on urban areas, though the studies for detecting built-up area as man-made features. In addition, radar imagery may provide the third dimensional information. Hence, SAR imaging technique is relevant tools to analyze urban land use patterns.The automated production of maps of human settlement from recent satellite images is essential to detailed studies of urbanization, population movement, and the like. Optical satellite remote sensing depends on sunlight illuminating the Earth in order to obtain useful images. Its performance is therefore restricted by the presence of clouds, fog, smoke or darkness. On the contrary, a SAR system can operate at day and night regardless of cloud cover and weather conditions. Furthermore, the SAR imaging system will show the microwave backscattering properties of ground surface covers and manmade structures. Settlements are especially sensitive to radar microwaves due to the presence and effects of horizontal and vertical structures (dihedral and trihedral reflectors). The geometric arrangement of settlements, the shapes and sizes of structures, and the morphology of settlements create high contrasts between human settlements and natural terrain and surface features because of an abundance of unique dihedral and trihedral reflectors and patterns ideally susceptible to radar sensors and radar analysis in an urban environment.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络