应用数字地形模型提高山地遥感数据自动分类精度
文献类型：期刊论文
| 作者 | 黄雪樵 |
| 刊名 | 山地研究
 |
| 出版日期 | 1986 |
| 卷号 | 4期号:1页码:96-103 |
| 关键词 | 自动分类 遥感数据 地形模型 应用数字 象元 土地覆盖类型 山地 分层分类法 最大似然率 地形数据 |
| 其他题名 | IMPROVING MOUNTAINOUS LAND COVER CLASSIFICATION ACCURACY BY COMBINING A DIGITAL TOPOGRAPHIC MODEL WITH LANDSAT DATA |
| 通讯作者 | 黄雪樵 |
| 中文摘要 | 山地地区自然环境复杂多变,山地遥感有着与平原遥感不同的特点。一般地说,山地遥感数据自动分类精度低于平原地区。对于地面分辨率较低的数据如陆地卫星的多光谱扫描(MSS)数据尤其如此。提高土地覆盖自动分类精度,是发展有效的山地遥感方法的重要内容之一。它对于调查和评价山地自然环境、资源和土地利用状况,规划和开发山区,发展山区经济是十分有益的。 本文根据地形是山地地理环境空间分异的主导因素这一观点，以位于横断山区的四川省米易试验区为例，将数字地形信息引入遥感数据计算机分类过程，以提高山地覆盖分类精度。具体方法是将数字地形模型（DTM）与MSS数据配准，采用一种多阶段的分层分类法进行分类。 米易试验区土地覆盖自动分类试验结果表明，DTM是山地遥感自动分类的一种有效辅助数据。分层分类法是一种适合于复合数据的分类方法，其精度显著高于单独用MSS数据的最大似然率分类的精度。由于分类结果已和DTM配准，利用计算机可以快速、定量地分析土地覆盖分布规律，并且可以很方便地与地理信息系统的其他数据相结合，进行多种地学综合分析和评价。 |
| 英文摘要 | The accuracy of Landsat digital classification in mountainous region is generally low for complixity of natural environment. But we can get some improvements by integrating topographic data.Author performed a computer classification of land cover with combined Landsat MSS data and a digital topographic model (DTM) of Miyi study area, located in mountainous region of Southwest Sichuan, First,the DTM was registered to the MSS data which had been corrected geometrically, with the average registration accuracy of one pixel. The registered data were then classified using a two-stage classification technique, "Layered classification", which involves two steps: 1.the MSS data were clustered to 27 spectrally homogenerous classes; 2.A linear discrimination, "multi-classes regression discrimination", was used with the DTM as verables to partition the spetral classes into level I land cover categories, i.e. paddy field, dry land, forest, sparse forest, shrub, grass, settlement, water and sand bank.A accuracy assessment was conducted.The overall accuracy of classification is 81.5 per cent for layered technique, comparing with 55.4 per cent for maxium-likehood classification with the MSS data alone. The significance test showed that the difference between two methods was significant.On the basis of the classification, author has conducted a computer analysis of relationship between land cover distribution and topography in the study site. |
| 语种 | 中文 |
| 公开日期 | 2014-08-20 |
| 源URL | [http://ir.imde.ac.cn/handle/131551/7116]  |
| 专题 | 成都山地灾害与环境研究所_成都山地所知识仓储（2009年以前） |
| 推荐引用方式 GB/T 7714 | 黄雪樵. 应用数字地形模型提高山地遥感数据自动分类精度[J]. 山地研究,1986,4(1):96-103. |
| APA | 黄雪樵.(1986).应用数字地形模型提高山地遥感数据自动分类精度.山地研究,4(1),96-103. |
| MLA | 黄雪樵."应用数字地形模型提高山地遥感数据自动分类精度".山地研究 4.1(1986):96-103. |
入库方式： OAI收割
来源：成都山地灾害与环境研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。