中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 地理科学与资源研究所
  3. 中国科学院地理科学与资源研究所
  4. 千烟洲站森林生态系统研究中心
  5. 外文论文
Organ-level nitrogen diagnosis in rice by integrating hyperspectral spectroscopy with advanced feature selection techniques

文献类型：期刊论文

作者Zou, Yaopeng8; Pei, Jie7,8; Fang, Huajun5,6; Liu, Yibo4,8; Geng, Jing7,8; Huang, Huabing7,8; Wang, Tianxing7,8; Huang, Jianxi1,2,3
刊名EUROPEAN JOURNAL OF AGRONOMY
出版日期2026-06-01
卷号177页码:128109
关键词Nitrogen diagnosis Remote sensing Feature selection Nitrogen Nutrition Index (NNI) Rice
ISSN号1161-0301
DOI10.1016/j.eja.2026.128109
产权排序3
文献子类Article
英文摘要Accurate organ-level diagnosis of nitrogen (N) status is critical for precision nutrient management in rice. However, canopy-level hyperspectral monitoring primarily reflects leaf information and tends to overlook N variability in stems and panicles. Moreover, the ability of leaf spectra to predict organ-specific N status across growth stages remains unclear. To address this gap, we conducted hyperspectral measurements on rice plants collected from field plots at multiple growth stages across two experimental sites and using different rice varieties. Spectral similarities among leaves, stems, and panicles were quantified by the Spectral Angle Mapper, revealing consistent spectral patterns despite distinct N concentrations. Organ spectra were then preprocessed with fractional-order derivative (FOD) to sharpen features. Next, a two-step feature selection strategy-combining mRMR with Fuzzy Rough Feature Selection, Boruta, and CARS-was employed to identify N-sensitive bands. Using these selected features, organ-level N concentrations were estimated by pairing leaf spectra with organ-specific N data and applying a Stacking ensemble learning strategy. Moreover, critical nitrogen dilution curves and the Nitrogen Nutrition Index (NNI) were developed to evaluate the capability of leaf spectra for whole-plant N diagnosis. Results demonstrated that leaf spectra served as effective predictors for nitrogen concentrations across all organs, achieving high accuracy with R2 values of 0.924 (leaves), 0.838 (stems), and 0.766 (panicles). NNI estimation based on leaf spectra reached a mean R2 of 0.947 and RRMSE of 7.41%, with cross-site validation confirming robustness (R2 = 0.918, RRMSE = 8.01%). Our findings demonstrate that leaf hyperspectral data provide a robust and effective tool for organ-level N monitoring in rice.
URL标识查看原文
WOS研究方向Agriculture
语种英语
WOS记录号WOS:001735302400001
出版者ELSEVIER
源URL[http://ir.igsnrr.ac.cn/handle/311030/221468]  
专题千烟洲站森林生态系统研究中心_外文论文
通讯作者Pei, Jie
作者单位1.Minist Agr & Rural Affairs, Key Lab Remote Sensing Agrihazards, Beijing 100083, Peoples R China
2.China Agr Univ, Coll Land Sci & Technol, Beijing 100083, Peoples R China;
3.Southwest Jiaotong Univ, Fac Geosci & Engn, Chengdu 611756, Peoples R China;
4.Chinese Acad Sci, Aerosp Informat Res Inst, State Key Lab Remote Sensing & Digital Earth, Beijing 100101, Peoples R China;
5.Zhongke Jian Inst Ecoenvironm Sci, Jian 343000, Peoples R China;
6.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Key Lab Ecosyst Network Observat & Modeling, Beijing 100101, Peoples R China;
7.Sun Yat sen Univ, Key Lab Comprehens Observat Polar Environm, Minist Educ, Zhuhai 519082, Peoples R China;
8.Sun Yat sen Univ, Sch Geospatial Engn & Sci, Zhuhai 519082, Peoples R China;
推荐引用方式
GB/T 7714		 Zou, Yaopeng,Pei, Jie,Fang, Huajun,et al. Organ-level nitrogen diagnosis in rice by integrating hyperspectral spectroscopy with advanced feature selection techniques[J]. EUROPEAN JOURNAL OF AGRONOMY,2026,177:128109.
APA Zou, Yaopeng.,Pei, Jie.,Fang, Huajun.,Liu, Yibo.,Geng, Jing.,...&Huang, Jianxi.(2026).Organ-level nitrogen diagnosis in rice by integrating hyperspectral spectroscopy with advanced feature selection techniques.EUROPEAN JOURNAL OF AGRONOMY,177,128109.
MLA Zou, Yaopeng,et al."Organ-level nitrogen diagnosis in rice by integrating hyperspectral spectroscopy with advanced feature selection techniques".EUROPEAN JOURNAL OF AGRONOMY 177(2026):128109.

入库方式： OAI收割

来源：地理科学与资源研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。