摘要各种水环境的监测，都离不开对水体信息的及时了解和掌握。由于遥感技术 可以实现大范围地对地宏观监测，并且可以快速且周期性地实现对同一地区的持 续观测，逐渐成为人们地表观测的重要途径。当前遥感正向着越来越高的空间分 辨率方向发展，利用高分辨率影像开展的各种研究受到广泛关注。特别是高空间 分辨率的影像能更为清晰的表达水体的空间特征与结构，为水体信息的高效、准 确识别和提取奠定了坚实的基础。69938

本研究以徐州为研究区，采用国产高分一号（GF-1）高分辨率数据开展水 体信息提取研究，影像经过辐射校正、裁剪研究区、几何校正等预处理。通过单 波段阈值法、谱间关系法、归一化差异水体指数（NDWI）、决策树法分别对徐 州水体进行提取。结果表明，针对徐州的地表特点，谱间关系法对城区和山区为 主的地表提取方法是最为适宜的，而对以植被为主的地表提取 NDWI 最为适宜， 构建的决策树对于大面积的水域提取精度达 76%，比轮廓破碎的水域提取精度 67%要好。

毕业论文关键词：遥感 高分辨率 徐州 水体信息提取 高分一号 分类

Extraction of water information of Xuzhou based on Xuzhou GF-1 remote sense image

Abstract Various monitoring water environment, are inseparable from the water to keep abreast of information and master. Remote sensing technology can achieve a wide range of macro-Earth monitoring, and can be quickly and periodically sustained observations of the same area, it has become an important way for people to surface observations. Current sensing is toward  higher and higher spatial resolution direction, various studies carried out using high-resolution images of widespread concern. Especially the high spatial resolution of the image can be more clearly expressed characteristics and spatial structure of a body of water, the water body information efficiently and accurately identify and extract laid a solid foundation.

In this study of Xuzhou as the study area, the use of domestic high One (GF-1) high-resolution data to carry out water extraction of information, image after radiometric correction, cropping the study area, geometric correction preprocessing. Single-band threshold, spectral Relations Act, Normalized Difference Water Index (NDWI), decision tree method was used to extract water bodies Xuzhou. The results showed that, for Xuzhou surface features, surface extraction method of spectral Relations Act of town and the main mountain is the most appropriate, and based on the vegetation surface extraction NDWI most appropriate,  build decision trees for a large area waters the extraction accuracy of 76%, water extraction accuracy than the broken outline 67% better.

Key  Words:Remote Sensing Highresolution Xuzhou Water  Body Extraction GF-1 Classification

目 录

摘 要 I

Abstract II

目 录 III

1 绪论 1

1.1  研究背景与意义 1

1.2  国内外研究现状 1

1.3  论文研究内容和技术路线 2

2  研究区概况及数据源