中文    English
当前位置: 本站首页 » 科学研究 » 科研成果 » 正文

Estimation of nitrogen content based on fluorescence spectrum and principal component analysis in paddy rice

发布日期:2016-12-01 18:12:57 阅读次数:[2348]次 作者:

核心提示:来源出版物: PLANT SOIL AND ENVIRONMENT

作者: Yang, J (Yang, J.); Gong, W (Gong, W.); Shi, S (Shi, S.); Du, L (Du, L.); Sun, J (Sun, J.); Song, SL (Song, S. -L.) 
来源出版物: PLANT SOIL AND ENVIRONMENT  卷: 62  期: 4  页: 178-183  DOI: 10.17221/802/2015-PSE  出版年: 2016
摘要: Paddy rice is one of the most important cereal crops in China. Nitrogen (N) is closely related to crops production by influencing the photosynthetic efficiency of paddy rice. In this study, laser-induced fluorescence (LIF) technology with the help of principal component analysis (PCA) and back-propagation neural network (BPNN) is proposed to monitor leaf N content (LNC) of paddy rice. The PCA is utilized to extract the characteristic variables of LIF spectra by analysing the major attributes. The results showed that the first three principal components (PCs) can explain 95.76% and 93.53% of the total variance contained in the fluorescence spectra for tillering stage and shooting stage, respectively. Then, BPNN was utilized to inverse the LNC on the basis of the first three PCs as input variables and can obtain the satisfactory inversion results (R-2 of tillering stage and shooting stage are 0.952 and 0.931, respectively; residual main range from -0.2 to 0.2 mg/g). The experimental results demonstrated that LIF technique combined with multivariate analysis will be a useful method for monitoring the LNC of paddy rice, which can provide consultations for the decision-making of peasants in their N fertilization strategies.


版权所有:测绘遥感信息工程国家重点实验室   
联系地址: 中国·武汉市珞瑜路129号   邮编: 430079   E-mail:liesmars@whu.edu.cn
Tel/Fax:027-68778969(办公室) 027-68778229(国际交流办公室)027-68778525(研究生管理办公室)