中国农业科学院农业信息研究所 中国农学会
Agricultural Information Institute of CAAS China Association of Agricultural Science Societies
En
Ch
中国农业科学院农业信息研究所 中国农学会 主办

AI检索

订阅

返回顶部
当前位置: 首页 > 文章 > 山东滨海盐土盐分含量高光谱特性及其反演研究 土壤通报 2013 (5) 1096-1100
Position: Home > Articles > Hyperspectral Characteristics and Retrieval of Salt Content in the Coastal Saline Soil of Shandong Province Chinese Journal of Soil Science 2013 (5) 1096-1100

山东滨海盐土盐分含量高光谱特性及其反演研究

作  者：
王娜娜;齐伟;宋萍;刘炳良;李哲
单  位：
山东农业大学资源与环境学院
关键词：
滨海盐土;高光谱;预测模型;BP神经网络
摘  要：
为实现快速、准确估测山东滨海盐土中盐分含量水平,推动土壤信息化管理进程,该研究利用ASD FieldSpec 3地物光谱仪在室内条件下测定了77个风干土壤样品的高光谱反射率。在分析土壤原始光谱特征的基础上,分析原始光谱及各种变换形式与土壤盐分含量的相关关系,确定特征点和敏感波段,采用一元曲线回归、多元线性逐步回归和基于BP神经网络回归三种模型进行模拟,对模型进行优选和检验。结果表明,经一阶微分变换能显著提高光谱反射率与土壤盐分含量的相关性,滨海盐土光谱的敏感波段范围是810~830 nm、1490~1520 nm、1900~1950 nm和1990~2075 nm;建立的3种模型中,基于BP人工神经网络模型优于多元线性逐步回归和一元曲线回归模型,建模样本和验证样本的预测值和实测值的相关系数分别达到0.9560和0.8840,斜率分别为0.9193和1.0728,表明模型的自预测能力和预测能力均较高。研究成果为快速预测山东滨海盐土的盐分含量提供了理论依据和技术支撑。
译  名：
Hyperspectral Characteristics and Retrieval of Salt Content in the Coastal Saline Soil of Shandong Province
作  者：
WANG Na-na;QI Wei;SONG Ping;LIU Bing-liang;LI Zhe;College of Resource and Environment, Shandong Agricultural University;
关键词：
Coastal Saline Soil;;Hyperspectrum;;Prediction mode;;BP neural network
摘  要：
In order to evaluate rapidly and accurately the soil salinity content and promote the soil digital management in coastal saline area of Shandong Province, this research measured the hyperspectral reflectance of 77 soil samples under laboratory conditions with ASD FieldSpec 3 spectrometer. The correlations of the original spectrum and spectral transformation forms with the salt content were analyzed based on the original spectral characteristics of soil, then the feature points and sensitive wavebands were determined. The one-variable linear regression model, multi-variable linear regression model and BP neural network model were built, optimized and tested, respectively. The results indicated that the salt content of soil had better relationship with the first deviative reflectance than the original spectrum, and the sensitive wavebands were 810 ~ 830 nm, 1490 ~ 1520 nm, 1900 ~ 1950 nm and 1990 ~ 2075 nm.In the three models, the model accuracy of BP neural network was better than one-variable linear regression and multi-variable regression. By validity, the Calibration Rc and Validation Rv were 0.9560 and 0.8840, the slopes were 0.9193 and 1.0728 respectively. This study will provide theoretical basis and technical support for the rapid prediction of the coastal saline soil salt content in Shandong Province.

相似文章

导出引用
计量
文章访问数: 8
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊

友情链接Friendly link
农业信息研究所Agricultural information institute 农业智能知识服务平台Agricultural intelligent knowledge service platform 农业学术服务平台Agricultural Knowledge Service System 中国知网CNKI 维普网VIP 万方WANFANG DATA
联系我们Contact us

电话Tell：010-82109652-808

邮箱E-mail：agrijournal@caas.cn

服务Service
关于我们About us 意见建议Opinions and suggest
版权说明Copyright
单位：中国农业科学院农业信息研究所 中国农学会 © 2017中国农业科学院农业信息研究所 ICP备案号：京ICP备09089781号-46