Design And Experiment Of Real-time Detection System For Fruiter Canopy Information Based On Ultrasonic Echo Signal

The use of chemical pesticides is the main way of plant protection operations in orchards,and the conventional orchard application equipment has a large spraying volume,which also brings a series of problems,such as environmental pollution,increased pest resistance,and more pesticide residues in fruits and vegetables.Accurate detection of canopy information is a key technology for precision application.Nowadays,precision application technologies such as mimic spraying,target spraying and variable spraying can effectively reduce pesticide use.In this paper,we built a canopy information detection test system for fruit trees,established and analyzed a mathematical model of canopy leaf area density based on this test system,developed a real-time fruit tree canopy information detection system using Lab VIEW software,and conducted indoor and outdoor tests and performance analysis of the fruit tree canopy real-time detection system.The work carried out in this paper can provide technical reference for ultrasonic wave for real-time detection in orchards.（1）The design and construction of the fruit tree canopy information real-time detection test system introduces the hardware selection of the test system,builds a simulated canopy test rig,analyzes the propagation characteristics of the ultrasonic sensor,and carries out the ultrasonic beam detection range test.（2）Based on the principle of ultrasonic echo signal for canopy leaf area density detection,and three different methods of ultrasonic echo signal analysis were proposed.A quadratic regression orthogonal test design method was used to establish a mathematical model of leaf area density based on the ultrasonic echo signal.The experimental results showed that the mathematical model obtained by the V_b-based echo analysis method was the best among the three different ultrasonic echo signal analysis methods,and the specific model was:y=1.9-0.688x₁+0.169x₂+0.281x₁²-0.021x₂².The adjusted R²of the model was 0.9180,and the p-value of the misfit term was 0.0735>0.05,indicating that the out-of-fit term is not significant and the mathematical model has a high degree of fit.（3）A real-time canopy information detection system written based on Lab VIEW software is introduced,which completes the design of algorithms such as ultrasonic echo signal acquisition and peak extraction,canopy thickness,canopy leaf area density.The information of the measured canopy can be observed in real time in the display interface,and the user can view the canopy information data collected by this system on the desktop after the system is finished running.The system can provide real-time data support for canopy parameter detection in precision planting operation.（4）Tested the detection error and adaptability of the fruit tree canopy information real-time detection system,indoor and field trials of the fruit tree canopy information real-time detection system were conducted.The results of the indoor thickness detection test showed that the average absolute error value of 70%did not exceed 5.0 cm,which indicated that the canopy information real-time detection had high accuracy in the indoor thickness detection test;the field thickness detection test selected two kinds of grape canopies,Shine Muscat and Summer Black,and the results of the canopy thickness detection of Summer Black grapes showed that the average absolute error value between the actual value and the measured value was 5.3 cm;The average absolute error between the actual and measured values was 7.9 cm,and the results of the field thickness detection tests showed that the system had good applicability for both types of grapevine canopy thickness measurements.The indoor canopy leaf area density detection results showed that the overall average relative error of the system for the simulated canopy leaf area density detection was 5.36%;the outdoor laurel leaf area density detection results showed that the average relative error of the system for the laurel leaf area detection was 7.79%,and the system had small relative errors for both indoor and outdoor canopy leaf area density detection results.