【作者】 吴林巧；

【导师】 周春国；

【作者基本信息】 南京林业大学 ， 森林经理学， 2009， 硕士

【摘要】 遥感影像数据不仅能为森林资源调查提供丰富的地物光谱特性、空间位置等重要信息,而且具有宏观、便捷、周期重复、动态检测和低成本等优点,已经成为森林资源调查新的信息源。利用遥感技术结合GIS、GPS技术进行森林资源调查已经成为一种必然的趋势。本文以中山陵风景区2006年QuickBird遥感影像、2002年中山陵二类调查小班数据、1999年中山陵1:10000地形图为研究数据源,以RS、GIS为主要技术手段,对中山陵森林资源植被分类进行应用研究。通过研究,得出以下结论:(1)对研究区QuickBird影像进行一级分类,发现最大似然法能较好的区分林业用地、水体、农田及其它等一级地类,并将分类结果用于中山陵及各景区地类统计。(2)对研究区QuickBird影像进行二级分类,研究发现支持向量机的分类能取得较好的效果,并将其分类结果用于阔叶林、针叶林、针阔混交林、竹林、苗圃、水域、农田及其它等二级地类的面积统计,但其中的阔叶林和针阔混交林混分现象严重,分类得到的阔叶林面积偏小,针阔混交林面积偏大。(3)利用2002年中山陵二类调查小班数据计算主要树种栎类、马尾松、刺槐的生物量和生产力。结果表明:刺槐的平均生产力最高。更多还原

【Abstract】 The Remote Sensing image has become a new source of information for forest resources survey not only because it can provide plenty of information of spectrum and spatial location of features but also because it has many advantages, such as macro, convenient, the cycle repeated, dynamic survey and low-cost and so on.Using the technology of RS along with GIS and GPS has become an inevitable trend.In this paper, the writer chose the QuickBird image of Dr. Sun yat-sen’s Mausoleum 2006, the data of forest management inventory of Dr. Sun yat-sen’s Mausoleum 2002 and the relief map of 1:10000 of Dr. Sun yat-sen’s Mausoleum 1999 as the information source, chose the technology of RS and GIS as the main technical methods, studied the forest resources classification of Dr. Sun yat-sen’s Mausoleum.The main contents and conclusions were summarized as follows through the research:(1) Classifying the QuickBird image of the research region in level 1 by Maximum Likelihood Classifier, there was a better result of the distinction between the forest and water, farmland ,other. The result was used for the statistics of the area of the four types of land in Dr. Sun yat-sen’s Mausoleum and its six scenic spots.(2) Classifying the QuickBird image of the research region in level 2. The research showed that Support Vector Machine could has a better classification results and the results were used for the statistics of the area of broadleaf, conifer, con_broad, bamboo, nursery, water, farmland and other. But broadleaf and con_broad were mixed in some degree, so the area of the broadleaf was smaller than its real area and the area of the con_broad was bigger than its real area.(3) Calculated the biomass and productivity of Quercus、Pinus massoniana、Robinia pseudoacacia with the data of forest management inventory,2002, Dr. Sun yat-sen’s Mausoleum. The research showed that the average productivity of Robinia pseudoacacia was the biggest among the three trees species.更多还原