首页 | 本学科首页   官方微博 | 高级检索  
     检索      

高光谱遥感数据用于水稻精细分类研究
引用本文:张丰,熊桢,寇宁.高光谱遥感数据用于水稻精细分类研究[J].武汉理工大学学报,2002,24(10):36-39.
作者姓名:张丰  熊桢  寇宁
作者单位:1. 武汉大学
2. 中科院遥感所
基金项目:中科院“九五”重大资助项目 (KZ95 1- A1- 30 2 - 0 4 - 0 1),中国资源环境遥感信息系统及农情速报资助项目 (KZ95 T- 0 3- 0 4 -0 1)
摘    要:根据水稻生长期的高光谱数据的光谱特征,设计了一个混合决策树分类算法。该算法的特征波段根据波段间的可分离性决定,算法的选择根据实际分类效果决定,波段间的可分离性根据各波段均值的归一化距离决定。最后用江苏常州市金坛良种场的高光谱图象数据做了分类实验,取得了测试样本总体分类精度94.9%的好结果。

关 键 词:高光谱遥感数据  混合决策树  水稻  精细分类
文章编号:1671-4431(2002)10-0036-04
修稿时间:2002年7月26日

Airborne Hyperspectral Remote Sensing Image Data is Used for Rice Precise Classification
Zhang Feng Xiong Zhen Kou Ning Experimenter,School of Remote Sensing Information Engineering,Wuhan University,Wuhan ,China..Airborne Hyperspectral Remote Sensing Image Data is Used for Rice Precise Classification[J].Journal of Wuhan University of Technology,2002,24(10):36-39.
Authors:Zhang Feng Xiong Zhen Kou Ning Experimenter  School of Remote Sensing Information Engineering  Wuhan University  Wuhan  China
Institution:Zhang Feng Xiong Zhen Kou Ning Experimenter,School of Remote Sensing Information Engineering,Wuhan University,Wuhan 430079,China.
Abstract:According to the rice spectral features of hyperspectral image data acquired during the rice is growing, a hybrid decision tree classification algorithm dealing with the variety of rice is developed. The feature bands are selected according to the separability among bands, but the classification algorithm is selected according to its classification result. The separability among bands is calculated by the normalized mean value of each band. In the end, a classification experiment is done. In the experiment the hyperspectral image data acquired in Jintan rice breeding farm is used. A good classification result is achieved. The classification accuracy of test samples is reached 94.9 percent.
Keywords:hybrid decision tree  \ rice  \ precise classification
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号