首页 | 官方网站   微博 | 高级检索  
     

基于人工神经网络的CHRIS数据内陆水体叶绿素浓度反演研究
引用本文:柳彩霞,傅南翔,郭子祺,付薇,蒋金雄,李琳.基于人工神经网络的CHRIS数据内陆水体叶绿素浓度反演研究[J].安徽农业科学,2009,37(26).
作者姓名:柳彩霞  傅南翔  郭子祺  付薇  蒋金雄  李琳
作者单位:1. 中国科学院遥感应用研究所,北京,100101
2. 北京交通大学土木建筑工程学院,北京,100044
3. 中国地质大学土地科学技术学院,北京,100083
摘    要:利用BP神经网络和CHRIS高光谱数据反演了富营养化非常严重的太湖梅梁湾地区叶绿素A浓度。首先计算了CHRIS模式2的18个波段与叶绿素A浓度的皮尔森相关系数,选择CHRIS的前5个波段和第13波段的反射率值作为神经网络的输入,以野外测量的叶绿素A浓度为神经网络的输出。实验表明,BP神经网络具有很好的非线性拟合能力,叶绿素A浓度的反演精度相对误差仅为22%,明显优于传统的多项式模型,显示BP神经网络与CHRIS高光谱数据结合的方法在内陆水体水质参数反演领域的应用具有相当的优势。

关 键 词:航天成像光谱仪CHRIS  人工神经网络  内陆水体  叶绿素浓度

Inversion Study on Chlorophyll Concentration of Inland Water from CHRIS Data Based on Artificial Neural Networks
LIU Cai-xia et al.Inversion Study on Chlorophyll Concentration of Inland Water from CHRIS Data Based on Artificial Neural Networks[J].Journal of Anhui Agricultural Sciences,2009,37(26).
Authors:LIU Cai-xia
Affiliation:LIU Cai-xia et al(Institute of Remote Sensing Applications,Chinese Academy of Sciences,Beijing 100101)
Abstract:Based on back-propagation neural networks and CHRIS data,this paper retrieved the chlorophyll-a concentration in Meiliang Bay of Taihu Lake,where water eutrophication was very serious.Pearson correlation coefficients between chlorophyll-a and eighteen bands of CHRIS were calculated and the first five bands and the thirteenth band were selected to be the inputs of the neural networks.The neural networks' output was the chlorophyll-a concentration obtained in field survey.The results showed that BP neural net...
Keywords:Spaceborne imaging spectrometry CHRIS  Artificial neural networks  Inland water  chlorophyll-a concentration  
本文献已被 CNKI 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-23

京公网安备 11010802026262号