| Drawing the grain yield map during the operation of the combine harvester is a requirement to achieve precision agriculture.The yield map must be measured in real time during the combined harvesting of the grain flow into the grain tank.Metering solutions currently applied to combine harvesters:(1)Based on the research of the impulse grain flow sensor technology,the relationship between the sensor voltage and the grain flow model is established to obtain grain flow.(2)Based on the theory of y ray attenuation,the relationship between γ ray radiation intensity and grain thickness model is established to obtain grain flow.(3)Based on the research of the weighing grain flow sensor,the relationship between the voltage of the weighing sensor and the grain quality model is established to obtain the grain flow.(4)Based on the research of the method of photoelectric diffuse reflection,the relationship of photoelectric model of grain flow is established to obtain grain flow.With the rapid development of science and technology,non-contact measurement solutions have gradually matured,and structured light vision technology is widely used in non-contact weighing solutions.Aiming at the problems in the above-mentioned combine harvester scheme,a measurement system for the volume flow of grains conveyed by the chute based on structured light 3D vision was designed.Combined with the developed measurement system software,the enveloping profile of the grain flow cross-section can be obtained by threshold value determination method for structured light image preprocessing.Using trapezoidal micro-element method,the cross-sectional area calculation model and volume calculation of the grain flow in the chute are established model.Calculate the volume flow of the grain material conveyed by the chute.The main research work of this article is as follows:1.A grain volume flow measurement system based on structured light vision technology was constructed.The chute was selected as the grain transportation platform,and Daheng industrial camera,line laser transmitter,photoelectric encoder and computer were used to complete the measurement system.The structure of the whole system is simple,economical and practical.and can effectively collect the structured light contour image of the grain material conveyed by the chute.2.The Zhang Zhengyou calibration method and cross laser calibration method are studied.Using Zhang Zhengyou’s checkerboard calibration method to complete the camera system calibration;based on the principle of laser triangulation,a cross laser calibration method was designed to calibrate the measurement system;the least square method was used to fit the calibration test data to obtain a calibration curve,which improved the interception Area model calculation speed.Through experimental verification,the average relative error between the fitted calibration curve and the obtained test data is 0.93%,which meets the calibration requirements.3.The processing technology of grain flow structured light contour image was studied.Image preprocessing techniques such as spatial domain filtering,frequency domain filtering,and edge detection are used.The median filtering with a template of 5*5 is selected as the image preprocessing method by analyzing the processing effect.The gray center of gravity method,geometric center method,skeleton refinement method and improved extreme value method were used to extract the centerline of the grain flow structured light contour image.The improved extreme value method was used as the centerline extraction method of the structured light contour.Judgment method solves the problem of the interference of the chute wall contour on the extraction of the contour centerline,and completes the extraction of the grain flow contour centerline.4.The method of measuring grain flow velocity was studied,and the calculation model of grain volume flow was established.A photoelectric encoder method for measuring the rotational speed of the driving wheel was designed to measure the grain velocity.Based on the trapezoidal micro-element integral sum method,the calculation models of the cross-sectional area and volume of the grain flow applied to the chute were established.5.The validity,accuracy and reliability of the grain volume flow measurement method and system studied in this paper are verified.For rice and sorghum grains,in four different volumes V1=17.6L,V2=24.8L,V3=32.0L,V4=39.2L,four chute inclination angles al=15°,α2=20°,α3=25°,α4=30°,four different structured light collection frame rates f1=40fps,f1=60fps,f1=80fps,f1=100fps,the test was performed under the conditions,and each group of conditions was repeated four times.With the same grain measurement volume,As the groove inclination increases,the average relative error shows an upward trend.The minimum error of rice and sorghum occurs at a chute inclination of 30°,a measurement volume of 17.6L,and a frame light collection frame rate of 100fps;when the measurement volume is 3.9.2L,the chute inclination angle is 20°,and the structured light image acquisition frame rate At 40fps,the maximum measurement error is obtained.The system measurement error is less than 5.2%,the repeated test variation coefficient is less than 0.021,and the root mean square error is less than 1.268.The validity,accuracy and reliability of the grain volume flow measurement method and system studied in this paper are verified.The test results show that the measurement system and method studied in this paper can meet the online measurement requirements of the chute. |
