| 高光谱林业遥感分类研究进展 |
| |
| 引用本文： | 李瑞平.高光谱林业遥感分类研究进展[J].安徽农业科学,2014(9):2801-2805. |
| |
| 作者姓名： | 李瑞平 |
| |
| 作者单位： | 北京林业大学省部共建森林培育与保护教育部重点实验室; |
| |
| 基金项目： | 国家“十二五”科技支撑项目(2012BAC01B03);教育部新世纪优秀人才支持计划项目(NCET-10-0230);国家自然科学基金项目(41171278) |
| |
| 摘 要： | 为了深入了解高光谱分类领域的研究现状,基于Web of Science数据库和CNKI数据库,检索了关于高光谱遥感分类的相关文献,并对文献的分布情况和研究方法等进行了归纳和分析.结果表明,关于高光谱分类的文献发布数量总体呈上升趋势,其中美国的文献发布量最多,热带森林类型受关注最多.采用最多的分类方法有最大似然法、支持向量机、随机森林、光谱角度制图和判别分析5种,5种方法各有优缺点,分类精度都较高,分类敏感波段大多在可见光、近红外和短波红外等波段.该研究可为高光谱林业遥感分类领域森林类型和分析方法的进一步研究提供参考.
|
| 关 键 词： | 高光谱遥感 文献分布 分类方法 |
Research Progress Analysis of Hyperspectral Remote Sensing Classification in Forestry |
| |
| LI Rui-ping.Research Progress Analysis of Hyperspectral Remote Sensing Classification in Forestry[J].Journal of Anhui Agricultural Sciences,2014(9):2801-2805. |
| |
| Authors: | LI Rui-ping |
| |
| Institution: | LI Rui-ping (The Key Iaboratory for Silviculture and Conservation of Ministry of Education, Beijing Forestry University, Beijing 10083 ) |
| |
| Abstract: | To understand the state of art of hyperspectral remote sensing classification of forestry, both Web of Science and CNKI data - bases were used to search related literature and summarize paper distributions and study methods. Result show that： （a） the number of papers keeps increasing; （b） USA ranks top one with most papers; （c） tropical forest is the most focus. There are five popular classification methods with high accuracies, which are Maximum Likelihood, Support Vector Machine, Random Forest, Spectral Angle Mapping and Discriminant Analy- sis. Each method has its own advantages and limitations. Generally, the sensitive wavelengths for classifications fall in visible, near infrared and shortwave ranges. This study provides reference for further study the method on hyperspectral remote sensing and its application on forest type. |
| |
| Keywords: | Hyperspectral remote sensing Literature distribution Classification methods |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
