Design And Experiment Of Reflective Leaf Chlorophyll Content Detector Based On Active Light Source

The determination of the chlorophyll content of crop leaves is of great significance in the monitoring of agricultural conditions and yield estimation.Therefore,it is particularly important for the fine management of agriculture to obtain the chlorophyll content of crops in real time.At present,the traditional detection methods mainly include spectrophotometry,spectral detection and transmission SPAD detector.The instruments used in the first two methods are relatively large and generally limited to laboratory use.Although the third method can be used in the field,the clamping transmission method used in the instrument does not have the application prospect of real-time monitoring.In order to achieve real-time and effective acquisition of field crop information,some scholars used chlorophyll's sensitive wavelengths to develop special reflective information sensing equipment,but it is still in the research and development stage in China and there are no mature commercial products.In order to promote the commercialization of domestic self-developed instruments,this paper used the field environment as the background to comprehensively analyze the relevant factors that may affect the detection accuracy during the design and use of the instrument,and given detailed solution.Based on this,a reflective leaf chlorophyll content detector based on active light source was designed to improve the practicality of the instrument.The main research contents and conclusions are as follows:(1)The mechanism and method of the reflective leaf chlorophyll content were explored.The independence between the values of normalized vegetation index(NDVI),green normalized vegetation index(GNDVI),ratio vegetation index(RVI),greenness vegetation index(GVI),red characteristic parameter(RCP)and green characteristic parameter(GCP)and detection distance was demonstrated,so that the designed detector could adapt to the change of the detection distance.It also analyzed the related factors that may affect the detection accuracy during the design and use of the detector,and provided clear clues for the subsequent solutions.(2)Spinach,green cabbage and oilseed rape were taken as research objects,the correlation between chlorophyll content and spectral reflectance was analyzed and 550 nm,710nm and850 nm were determined as the best detection wavelengths.The advantages and disadvantages of various optical devices were discussed and the most suitable devices were selected.(3)The structure of each part of the detector was designed.The multi-wavelength detectionoptical path was designed to make the detection area of the instrument at different wavelengths the same,avoiding the deviation of the detection value.The modular design concept was used to design the circuit frame,which not only facilitated the modification of the detection wavelength according to different needs,but also improved the efficiency of secondary development.(4)The software and hardware of the detector were designed.The constant current source drive circuit was designed to ensure the stability of the detection light source.The design scheme of light source modulation and synchronous detection was adopted to suppress the influence of ambient light and dark current,so that the detector could adapt to the complex lighting environment in the field.The influence of PCB design on circuit performance was analyzed,so that the linear correlation between the detected light intensity and voltage was as free as possible from PCB interference.Finally,the soldering board debugging and embedded software development were completed.(5)Relevant tests and experiments of the developed detector were completed.The maximum fluctuation rate of the detector output voltage under different lighting conditions was0.95%,and the maximum fluctuation rate of the detector's output feature parameters at different detection distances was 1.13%,indicating that the detector could adapt to changes in ambient light and detection distance.Taking spinach,green cabbage and oilseed rape as test objects,the least partial square algorithm was used to establish the detection model between the output feature parameters of the detector and the chlorophyll content.The results showed that all feature parameters and chlorophyll content had good linearity,and the correlation coefficients were all greater than 0.8304.However,the correlation between chlorophyll content and feature parameters of different leaves was somewhat different.In order to obtain the best detection model for each leaf,the enumeration method was used to evaluate the effect of the detection models composed of various combined feature parameters,and the best detection model for each leaf was selected.Experiments showed that the root mean square error of the best detection model was 0.1108mg/g,0.0919mg/g and 0.0587mg/g,respectively.The instrument designed in this paper has good practicability,not only can adapt to changes in ambient light and detection distance,but also can output rich characteristic parameters,which can provide a way for differential analysis of field crops.