| China is a large rice growing country,and its rice planting area is about 20% of the world's.In paddy field ecosystem,weeds are the main cause of rice yield decline and quality reduction.The application of chemical herbicides is a widely used weed control method,but the slather using of chemical herbicides has brought a lot of negative problems,such as weed resistance,crop damage,ecological and environmental pollution.With the development of modern agriculture,the awareness of environmental protection and the importance of food safety issues,control technology of herbicide reduction has gradually developed,in which the mechanized cultivation of weeding as a way of pollution-free weeding has been rapid development.The mechanical weeding are divided into inter-row weeding and intra-row weeding.Because the machine seedling crop line is ordered and there is no interference between the lines,the inter-row weeding is less difficult,and the mechanical weeding has also developed greatly.However,due to the reasons for the man-made operation at the time of machine planting,the row of seedlings is not strictly arranged according to the straight line,and there are bending in different positions at the same line.In the intra-row mechanical weeding,there is a lateral displacement due to the different positions of rice seedlings,resulting in difficult to man-made turning operations,damaging seedlings and affecting production.In this paper,the automatic avoiding seedlings control system of weeder for paddy field was designed,so that the weeding device to follow the movement of the seedling line to reduce the rolling of the seedlings;simulating of paddy field environment in the soil tank,accuracy of extraction algorithm for seedlings line and control system was test;in field conditions,the effect of automatic avoiding seedling control on the performance of weeding machinet.The research contents and conclusions are as follows:(1)Analyze the characteristics of paddy field,build an image acquisition system,and study the best seedling image sample collection strategy.This paper analyzed the application environment and requirements of the image acquisition system for paddy field,and selected the hardware of the image acquisition system.The factors affecting the image quality in the paddy field and natural environment were studied.The image acquisition system with shading function was constructed.By exploring the characteristics of weeds and crop growth in paddy field,the optimal collection time of image about rice seedlings was found that is between the pre-tillering period and the tillering period.(2)The method of partition edge fitting based on Hough transform that was used to extract the seedling lines and identify seedlings was studied.In this paper,the interference information affecting the recognition of seedlings and extraction of seedling lines in the natural environment was analyzed.Firstly,the gray image of the super-green algorithm was selected to transfer color image to gray image,then the Ostu automatic threshold method was used to binarize the grayscale image,the image noise was filtered by the area.Finally,the hole in the inner of the seedling belt was filled to better extract seedling lines in the natural environment.In the binary image,the Canny operator was used to detect the edge of the seedlings.Dividing the fitting interval in the middle of the image,edges of seedlings in the left and right sides of each subinterval was fitted by the Hough transform.The midpoint coordinates of the edge line in each subinterval were calculated as the midpoint of seedlings in this interval.The seedlings line was extracted by fit midpoints with the least squares method.Simulated the paddy field environment,a test about extraction precision of seedling line was carried out,the results showed that the extraction accuracy of the positive orientation was 0.945°,and the extraction precision of the negative orientation was 0.896°,the image processing speed is 0.4 s/f.(3)According to the working model of tractor advancement and weeding device,automatic avoiding seedlings control system based on PID control was designed.The control system of the automatic control system and the working condition of the system are analyzed,and the control system is designed with PLC as the control core,Hall speed sensor,push button switch and re lay as the peripheral device.This paper studies the camera installation position,constructs the visual system coordinate system,establishes the working model of the seedling belt and the weeding wheel,analyzes the variation law of the seedling belt offset,and sets the output signal of the visual system according to the law of the change of the belt line angle,Mapping Relationship between Line Angle Change and Output Signal C hange.The relationship between the advance speed of the weeding machine and the delay time is studied.The automatic control system based on PID control is designed.The test results show that the standard deviation of the control system control deflection is less than 1.5°and the control precision(root mean square error)is less than 1.5 °.(4)The test in the field was conducted to test the influence on test the effect of the control system on the performance of the weeder.The field experiment about automatic avoiding seedlings control system was carried out to test the effect of a utomatic control system on the performance of weeder.Field test was designed by Unit Linear Regression Whole Analysis Method,the experimental factor was velocity of weeder and the experimental index was injury rate.The results showed that: 1)before and after the installation of the control system,the average seedling rate of the weeding machine was 6.76% and 6.03%,and after installation of the control system,the injury rate is reduced at different speeds.2)Before the installation of the control system,the effect level of the speed on the injury rate was 0.05,and after that the effect level of the speed on the injury rate was 0.10,and the control system reduced the effect of speed on the injury rate.3)Before and after the installation of the control system,the regression equation between the speed and the injury rate is obtained,and fitted well. |
PDF Full Text Request