Design And Experimental Research Of Leave Droplet Deposition Detection System

In agricultural production,the main method of preventing and controlling pests and diseases is spraying chemical pesticides.The droplet size,uniformity and coverage density sprayed on crops are closely related to the control effect of pests and diseases.At present,the main detection methods of droplet deposition parameters include water-sensitive paper detection method,sample elution method and sensor method.The water-sensitive paper detection method can obtain the coverage and distribution of the droplet deposition;the sample elution method can detect the deposition amount of the droplet;the sensor method mainly obtains the droplet deposition parameters through chemical or physical means.In addition to the elution method to obtain the amount of droplet deposition on the leaf surface,the water-sensitive paper method and the sensor method both use the sampling medium to replace the leaf to collect the droplet deposition.Because the material of the droplet sampling medium is different from the actual plant,it cannot be truly reflected.The deposition effect of droplets on the leaf surface.In order to quickly measure the deposition distribution parameters of droplets on field leaves,this paper designs a detection system that can directly measure the spray deposition parameters on the leaf surface.The main research content includes the following four aspects:(1)The design of leaf droplet deposition detection system: The leaf droplet deposition detection system realizes the detection of leaf droplet deposition based on the fluorescence phenomenon of fluorescent solution droplets under ultraviolet light,combined with image processing technology.This paper designs a detection system including(1)the shell of the detection device,(2)the image acquisition and display module,(3)the ultraviolet light source module and(4)the image processing module.According to the requirements of image processing of the leaf droplet deposition,the software part of the image processing module was developed based on Android and Open CV.(2)Image processing algorithm optimization: In view of the non-fixed value of binarization threshold,the central hole of the droplet and the adhesion of the droplet found in the process of the fluorescence image processing of the leaf droplet deposition,the detection algorithm is optimized,combined with the Open CV algorithm library,achieved the adaptive threshold segmentation,flood filling,and pit segmentation respectively solve the above three problems and improve the detection accuracy of droplet deposition parameters.(3)Detection effect test of leaf droplet deposition detection system: The light stability of the fluorescent reagent Acid Brilliant Flavine 7G was investigated,The test results showed that the fluorescence intensity of the fluorescent reagent was 97.03% of that before exposure to sunlight for 0.5 hours,and it was recovered rate is 98.24%before exposure to sunlight for 0.5 hours,Proved the reagent can be used for leaf spray deposition detection in the field;the spray concentration of the fluorescent reagent is optimized,and the deposition parameters of the leaf droplet deposition are detected by the detection system,and the coverage rate of the detection system is compared with the results obtained by Image J’s Analyze particles function,and the number of deposited particles is compared with manual counting methods.The test results show that the detection system has an average error of 5.41% ±3.01% compared to the Analyze particles function of Image J when calculating the coverage rate.When calculating the number of deposited particles,the average error compared to the manual counting result is 4.01± 0.63%.(4)Charged droplet and non-charged droplet deposition detection test: The spray test under different electrostatic voltages was carried out,and the droplet deposition data on the leaf surface and the fog droplet deposition data on the droplet sampling medium were compared.The test results show that when non-electrostatic spray is used,the deposition amount per unit area and the number of deposited particles of the droplet sampling medium differs from the droplet deposition data on the leaf surface by 3.4%and 3.7%,respectively.This indicates that when the non-electrostatic spray is tested,the sampling The droplet deposition data of the medium is basically the same as the real droplet deposition data of the foliage.When static electricity is applied to the spray,the difference between the amount of droplets per unit area and the number of deposits on the leaf surface and the droplet sampling medium increases.Based on the data of the sampling medium,the amount of droplet deposition per unit area differs by 9.8%,13.4%,21.9% and-19.7% at 2.5k V,5k V,7.5k V and 10 k V,respectively,and the number of droplet deposition differs at 2.5k V,5k V,7.5k V and 10 k V,respectively.9.8%,21.8%,30.4% and-16.2%.