面向地理国情监测的水系提取研究与分析

发布时间：2018-09-10 07:42

【摘要】：地理国情,我国重要的基本国情,对社会经济发展,城镇化建设等起着重大的影响作用;地理国情通过对国情要素进行监测与统计分析,得到了反映要素空间分布及演变结果。水系作为基础测绘和地理国情监测的重要要素之一,研究水系提取及水系统计分析就至关重要。本文主要采用三种方法进行水系提取,基于传统的监督分类方法、基于eCognition软件的面向对象分类方法以及利用DEM提取水系的GIS方法。最后,在水系提取的基础上,结合四期的水系专题数据,对西安市周至县水系进行基本统计分析。基于eCognition的水系提取主要分为两步,第一步,选取最优的分割参数,包括尺度参数,形状参数,光谱参数,对遥感影像进行分割,分割质量的好坏影响着分类精度;第二步,在分割的基础上,利用eCognition软件中的特征知识库对地物特征进行分析,选取合适的对象特征,进行水系提取。本研究以分辨率为2米的ZY3融合影像为实验区,采用监督分类和面向对象的影像分类对水系进行提取,对提取结果进行比较,得到适用于地理国情监测项目的水系提取方法。探索出了利用DEM数据提取水系的GIS方法。本研究的创新之处在于,自动化的水系提取方法,极大的提高了项目日常生产效率,减少了人工工作量,为国情监测项目提供了一定的技术参考。
[Abstract]:Geographical situation, an important basic situation of our country, plays an important role in the development of social economy and the construction of urbanization, and through monitoring and statistical analysis of the elements of national conditions, the spatial distribution and evolution results of the elements are obtained. As one of the important elements of basic mapping and geographical monitoring, it is very important to study the extraction and statistical analysis of water system. This paper mainly adopts three methods for water system extraction, which are based on the traditional supervised classification method, the object-oriented classification method based on eCognition software and the GIS method based on DEM. Finally, on the basis of water system extraction, combined with the special data of four periods of water system, the basic statistical analysis of water system in Zhouzhi County, Xi'an is carried out. The water system extraction based on eCognition is divided into two steps. In the first step, the optimal segmentation parameters, including scale parameter, shape parameter and spectral parameter, are selected to segment the remote sensing image. The quality of segmentation affects the classification accuracy. On the basis of the segmentation, the feature knowledge base in eCognition software is used to analyze the feature of the ground object, and the suitable object feature is selected to extract the water system. In this study, the ZY3 fusion image with a resolution of 2 meters was used as the experimental area, the supervised classification and the object-oriented image classification were used to extract the water system, and the extraction results were compared, and a water system extraction method suitable for the geographical situation monitoring project was obtained. The GIS method for extracting water system from DEM data was explored. The innovation of this study is that the automatic water system extraction method greatly improves the daily production efficiency of the project, reduces the workload of labor, and provides a certain technical reference for the monitoring project of national conditions.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P331

【参考文献】