期刊
BIOSYSTEMS ENGINEERING
卷 176, 期 -, 页码 1-11出版社
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.biosystemseng.2018.09.019
关键词
Crop nitrogen stress index; Machine vision; Hyperspectral; Reflectance index; Photosynthesis
资金
- European Union
- Greek National Funds through the Operational Program Education and Lifelong Learning of the National Strategic Reference Framework (NSRF) - ARISTEIA GreenSense project
Early detection of nitrogen deficit stress is essential to effectively and precisely manage crop production under greenhouse conditions. This article demonstrates the use of hyperspectral machine vision as a non-contact technique for detecting crop nitrogen deficit in a soilless tomato crop. Three different levels of nitrogen concentration were applied in tomato plants grown in a growth chamber under controlled environment conditions. The results demonstrate that crop reflectance increased due to nitrogen deficiency, mostly in the wavelength bands between 775 nm-850 nm and 910 nm-960 nm. Based on the reflectance measurements several reflectance indices were calculated and correlated with the tomato leaf chlorophyll or nitrogen content and with the leaf photosynthesis rate (As). The results showed that when the As and the chlorophyll content values changed more than 0.5 mu mol m(-2) s(-1) and 2.8 mu g cm(-2), respectively, the photochemical reflectance index (PRI) and the transformed chlorophyll absorption in reflectance index (TCARI) values varied more than 0.05 and 2.8, respectively. In addition, the results showed that for N changes higher than 0.20%, the optimised soil adjusted vegetation index (OSAVI) and the modified soil adjusted vegetation index (MSAVI) values varied more than 0.05 and 0.25, respectively. A new spectral index (background adjustment nitrogen index - BANI) was developed and validated under experimental conditions for the estimation of tomato plant nitrogen concentration. (C) 2018 IAgrE. Published by Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据