投稿中心

审稿中心

编辑中心

期刊出版

网站地图

友情链接

引用本文:宋开山,张柏,王宗明,段洪涛,徐京萍,陈铭.吉林查干湖水体叶绿素a含量高光谱模型研究.湖泊科学,2007,19(3):275-282. DOI:10.18307/2007.0308
SONG Kaishan,ZHANG Bai,WANG Zongming,DUAN Hongtao,XU Jingping,CHEN Ming.ANN-BP model for Chl-a concentration estimation with in-situ collected hyperspectral refelctance in Lake Chagan, Jilin Province. J. Lake Sci.2007,19(3):275-282. DOI:10.18307/2007.0308
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 1855次   下载 1023 本文二维码信息
码上扫一扫!
分享到: 微信 更多
吉林查干湖水体叶绿素a含量高光谱模型研究
宋开山,张柏,王宗明,段洪涛,徐京萍,陈铭
中国科学院东北地理与农业生态研究所,长春 130012
摘要:
叶绿素a含量能够在一定程度上反映水质状况, 高光谱遥感可有效反演叶绿素a含量。该研究通过分析水体叶绿素a浓度与其高光谱反射特征之间的相关关系, 采用单波段相关分析、波段比值、微分光谱和神经网络模型等多种算法建立了叶绿素a高光谱定量模型。结果表明:叶绿素a与单波段光谱在蓝、绿波段相关系数较低, 而在红光与近红外波段有明显提高, 微分光谱也表现出同样的趋势;反射率比值算法模拟效果好于线性回归法;神经元网络模型可以大大提高实测光谱数据的反演能力, 确定性系数高达0.95。这为今后利用星载高光谱传感器在查干湖进行叶绿素a浓度大面积遥感反演提供了研究基础。
关键词:  查干湖  叶绿素a  高光谱  ANN-BP
DOI:10.18307/2007.0308
分类号:
基金项目:国家自然科学基金(40401003);中国科学院野外台站基金资助
ANN-BP model for Chl-a concentration estimation with in-situ collected hyperspectral refelctance in Lake Chagan, Jilin Province
SONG Kaishan,ZHANG Bai,WANG Zongming,DUAN Hongtao,XU Jingping,CHEN Ming
Abstract:
This study presented an approach for the determination of chlorophyll-a concentration from field reflectance spectral data in Lake Chagan, Jilin Province.Reflectance spectral data was collected from May 2004 to June 2005 with ASD FieldSpec spectrometer.Correlations between water chlorophyll a concentration and water reflectance, derivative reflectance were analyzed;linear regression models were constructed with every single band reflectance and derivative reflectance against water chlorophyll a data;band ratio model with reflectance at 700 nm and 580 nm was also established.Finally, ANN-BP model was established with diagnostic band derivative reflectance as input vectors.The results showed that chlorophyll a concentration had a close relationship with water reflectance and derivative reflectance from 400 to 900 nm.The regression model established with single band got the similar performance as that with band ratio as regression dependent variable.By comparison, the ANN-BP model performed best with determination coefficient (R2) of 0.95 and with the least RMSE of 1.50(μg/L).
Key words:  Lake Chagan  chlorophyll-a  hyperspectral  ANN-BP
分享按钮