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 > 基于高光谱“玉米田养鹅”模式下作物冠层氮素反演 农机化研究 2020 (3) 5-11
Position: Home > Articles > Crop Canopy Nitrogen Inversion Based on Hyperspectral Corn Field Geese Model Journal of Agricultural Mechanization Research 2020 (3) 5-11

基于高光谱“玉米田养鹅”模式下作物冠层氮素反演

作  者：
冯江;陆治冶;马昕宇;陈双龙;赵庆贺;王树文
单  位：
东北农业大学电气与信息学院;岭南师范学院信息工程学院
关键词：
高光谱;农牧一体化;氮素;玉米冠层
摘  要：
为实现"玉米田养鹅"模式下作物氮素养分快速无损监测,开展了基于高光谱"农牧一体化"实验.以放牧前后的玉米冠层作为研究对象,利用高光谱技术分析不同时期玉米冠层叶片光谱,采用波段自相关分析(Bands Inter-correlation Analysis, BICA)与主成分分析(Principal Components Analysis, PCA)相结合的方法提取特征波段,并构建多种光谱参数,进而建立了基于BICA-PCA方法的多元回归模型(Multivariable Regression Model,MRA),并筛选和验证了所建模型.结果表明:随着不同生育期的进行,鹅的粪便会作为肥料给作物补充氮素,近红外光谱反射率增高,红边位置向左移动.建立模型在放牧前期校正集决定系数R_C~2为0.796,校正集均方根误差(Root-mean-square Error Correction, RMSEC)为0.133;在预测集决定系数R_p~2为0.840,预测集均方根误差(Root-mean-square Error Prediction, RMSEP)为0.147.放牧后期R_C~2为0.761,RMSEC为0.094;R_p~2为0.789,RMSEP为0.141.研究结果可为"农牧一体化"优化生产管理和建立氮素养分特征平衡模型提供支持和帮助.
译  名：
Crop Canopy Nitrogen Inversion Based on Hyperspectral Corn Field Geese Model
作  者：
Feng Jiang;Lu zhiye;Ma Xinyu;Chen Shuanglong;Zhao Qinghe;Wang Shuwen;College of Electric and Information, Northeast Agricultural University;College of Imformation Engineering, Lingnan Normal College;
关键词：
hyperspectral;;agro-pastoral integration;;nitrogen;;corn canopy
摘  要：
In order to establish a rapid and non-destructivenitrogen nutrient monitoring of agro-pastoral integration, experiments based on hyperspectral agro-pastoral integration were carried out.Through taking different periods of corn canopy as the research object, different periods of corn canopy leaf spectrum were analyzed by using hyperspectral technology. The method was to combine Bands inter-correlation analysis(Bands inter-correlation analysis, BICA) and Principal component analysis(Principal component analysis, PCA) to extract the feature band. A simple regression analysis(SRA) and multivariable regression model(MRA) were established after constructing a variety of spectral parameters, and the established model was screened and validated. The results showed that with the excrement of goose at different growth stages, nitrogen was supplemented to crops as fertilizer, the reflectivity of NIR spectrum increased, and the position of red edge moved to the left. In terms of model building, the multiple linear regression model has better accuracy. The coefficient of determination of the model was 0.796 and the root mean square error of calibration set was 0.133. The coefficient of determination in the prediction set was 0.840 and the root mean square error of prediction set was 0.147. The coefficient of determination of the late-grazing calibration set was 0.761, and the root-mean-square error of the calibration set was 0.094. The coefficient of determination in the prediction set was 0.789 and the root mean square error of the prediction set was 0.141. The results will help to optimize the production management and establish a balanced model of nitrogen nutrition of "agro-pastoral integration" in the future.

相似文章

导出引用
计量
文章访问数: 10
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