CAJ | 학술논문

针对农业生产环境中植物群体叶绿素的监测，该文提出了一种基于可见-近红外反射光谱的植物群体相对叶绿素含量实时监测系统。系统以太阳光为光源，设计了特殊光路同时获得植物冠层反射的700和830 nm 2种波长光信号以及该2种波长光的入射信号。采用ZigBee协议搭建无线传感器监测网络，由终端节点将以上光路获取的4路信号进行采集和处理，并将其转发至协调器，进而由协调器发送至用户中心进行计算与存储。系统试验研究中测量的有效样本点共124个，将前36个点的两波长吸光度值与样本叶绿素指数值采用最小二乘法进行多元回归，建立定量分析模型，模型决定系数和标准差分别为0.919和9.26958；用后88个点建立预测集，模型预测值与标准值建立的线性函数决定系数为0.913，标准差为9.372，系统稳定性试验变异系数 CV<1.82%。结果表明，该系统可用于自然光照条件下群体叶绿素的准确测量与实时监测，为精细农业自动无损检测技术提供理论及实践基础。
침대농업생산배경중식물군체협록소적감측，해문제출료일충기우가견-근홍외반사광보적식물군체상대협록소함량실시감측계통。계통이태양광위광원，설계료특수광로동시획득식물관층반사적700화830 nm 2충파장광신호이급해2충파장광적입사신호。채용ZigBee협의탑건무선전감기감측망락，유종단절점장이상광로획취적4로신호진행채집화처리，병장기전발지협조기，진이유협조기발송지용호중심진행계산여존저。계통시험연구중측량적유효양본점공124개，장전36개점적량파장흡광도치여양본협록소지수치채용최소이승법진행다원회귀，건립정량분석모형，모형결정계수화표준차분별위0.919화9.26958；용후88개점건립예측집，모형예측치여표준치건립적선성함수결정계수위0.913，표준차위9.372，계통은정성시험변이계수 CV<1.82%。결과표명，해계통가용우자연광조조건하군체협록소적준학측량여실시감측，위정세농업자동무손검측기술제공이론급실천기출。
To monitor the changes of group chlorophyll for field plants, in this paper, a real-time monitoring system was developed based on Vis/NIR (visible and near-infrared) reflected spectroscopy using a nondestructive method. Reflectance at 700 nm was found to be a very sensitive indicator of chlorophyll concentration, but reflectance of wavelengths at a near infrared band was not. Calculating the reflectance properties of wavelengths at 700 nm and 830 nm can represent chlorophyll values in green leaves. To obtain physiological information of a plant from remote and different fields, a wireless sensor network (WSN) was set up using a ZigBee protocol in the system, in which the terminal device consisted of an optical unit, C8051F020 micro controller unit, WSN node, and a temperature and humidity sensor. The terminal device could obtain the signals of an incident and reflected light from the plant canopy at 700 and 830 nm which were transmitted to the coordinator. Then the coordinator sent the signals to the user center to be calculated and stored. Also, the data of temperature and humidity was sent to the user monitoring center at the same time. The performance of the system was investigated in a tall fescue lawn, and a total of 124 samples were measured in the experiments. They were divided into a training set and a test set. Thirty-six samples were selected as the training set to establish the quantitative analytical model for the experiment. The results showed a good correlation (the coefficient of determinationR2 was 0.919 and the standard deviation was 9.26958) between the group chlorophyll index determined by the standard instrument (FieldScout CM-1000 handheld chlorophyll meter) and the absorbance values at the two wavelengths. Based on the model established using a linear prediction equation, we calculated the group chlorophyll value for the test set samples. Promising results were obtained for the predicted values of group chlorophyll for the test samples, withR2 =0.913 for the test set. The coefficient of variation (CV) of the system for showing the stability was less than 1.82%. We found that the changing solar altitude did not affect system measurement during our experiments. The results suggested that this system could be used for accurate measurement and real-time monitoring of the group chlorophyll in a natural light condition, which would provide a theoretical and practical basis for the automatic nondestructive testing technology of precision agriculture.