水稻穗部氮素含量高光谱估测研究  

Hyperspectral Estimation of Nitrogen Content in Rice Panicle

作　　者：陈瑛瑛[1] 王徐艺凌 朱宇涵 武威[1] 刘涛[1] 孙成明[1] Chen Yingying;Wangxu Yiling;Zhu Yuhan;Wu Wei;Liu Tao;Sun Chengming(College of Agronomy,Yangzhou University/Key Laboratory of Crop Genetics and Physiology of Jiangsu Province/ Collaborative Innovation Center for Modern Production Technology of Grain Crops,Yangzhou 225009,Jiangsu,China)

机构地区：[1]扬州大学农学院/江苏省作物遗传生理重点实验室/粮食作物现代产业技术协同创新中心,江苏扬州225009

出　　处：《作物杂志》2018年第5期116-120,共5页Crops

基　　金：国家自然科学基金(31671615;31701355);江苏省大学生创新训练计划重点项目(201711117026Z);扬州大学大学生学术科技创新基金(20170648)

摘　　要：氮素是影响水稻生长发育的主要营养元素之一,稻穗生长与氮素营养息息相关。本研究利用高光谱技术测定了稻穗的全氮含量并进行了相应的分析,结果表明:稻穗全氮含量与冠层光谱反射率在近红外波段760～1 300nm呈极显著负相关关系,稻穗全氮含量与光谱特征指数λb、SDr、SDr/SDb、DVI的相关性较好,并建立了相应的估算模型。经独立的实测数据检验可知,基于SDr/SDb与DVI指数组合所建线性回归模型y=2.075+0.001x_1-2.952x_2估测稻穗全氮含量效果最好。上述结果为稻穗营养元素的快速诊断提供了新的手段。Nitrogen is one of the main nutrient elements affecting the growth and development of rice. The growth of rice panicle is closely related to nitrogen nutrition. In this study, the nitrogen content in rice panicle was measured and analyzed by using hyperspectral technology, and the results showed that total nitrogen content of rice panicle and canopy spectral reflectance presented highly significant negative correlation in the near infrared band 760 nm to1 300 nm. The correlations between total nitrogen content of rice panicle and spectral characteristic parameters, such as λb, SDr, SDr/SDb and DVI were good, and the estimation models were established based on these relationships. The tested result by independent measured data showed that the quadratic function model y=2.075＋0.001 x1-2.952 x2 based on SDr/SDb and DVI was the best model for estimating total nitrogen content of rice panicle. The above results could provide a new method to diagnose nutrient elements in rice panicle rapidly.

关 键 词：水稻 稻穗 高光谱 氮素含量 估测模型 

分 类 号：S511[农业科学—作物学]