Design Of Performance Detection Device Based On Multi-sensor Fusion Burrowing Device

Cotton is the main cash crop in Xinjiang.The planting quality of cotton is very important to its yield.At present,precision burrowing device is often used to plant cotton on film.In order to ensure the reliability and stability of the precision burrowing device,it is necessary to test its performance.Because the current manual detection method is time-consuming and laborious,it is very important to design the performance detection device of precision burrowing machine.In this paper,three kinds of detection technology of seed metering device are analyzed comprehensively: photoelectric sensing detection technology,piezoelectric pulse detection technology and CCD camera detection technology,and the advantages of these three detection technologies are combined.A multi-sensor fusion performance detection device for cave-seeding device is designed.With photoelectric sensing detection technology and image technology as the core,the device carries on continuous collection and display of seeding data to realize the accurate detection of the number of seeds in each hole during seeding,and analyzes the pass rate,replay rate,miss rate and total seeding amount of the precision burrowing device,so as to judge the seeding performance of the precision burrowing device during working.The main research contents are as follows:(1)Design the overall structure of the precision burrowing device performance detection device.The system is mainly composed of precision burrowing device,CMOS camera,light source,proximity switch,conveyor belt,grating sensor,alarm device,controller and laptop,etc.Select the key components,and build the hardware system most suitable for the performance detection of precision burrowing device.(2)STM32F103C8T6 single chip microcomputer is used as the controller of the detection device to complete the design of power input circuit,jack circuit,alarm circuit,LCD circuit and hardware anti-jamming.KEIL software is used to complete the writing of the main program and subroutine of the control system,and the program debugging.(3)CMOS camera is used to collect seeding images of precision burrowing device,and MATLAB software is used to complete extraction of sequence images and image processing of seeds.The image processing process includes: Image pruning,gray image,gray image enhancement,image binarization,image median filtering,image corrosion and image expansion,to obtain clear seed images,and to extract the seed number of frame image,and to communicate with the single chip microcomputer,complete the detection of the total seed amount of precision burrowing device.(4)Cotton seeds and corn seeds were selected to perform precision burrow seeding performance test with qualified rate,replay rate,leakage rate and total seed discharge as indicators.The qualified rate,replay rate,leakage rate and total seed discharge after image processing displayed on the LCD screen were compared and analyzed with manual detection,and the following results were obtained: The errors of qualified rate of Yuanmian 11 and Tahe 2 seeds were 0.2%-1.5% and 0.5%-2.0%,respectively,and the errors of total seed discharge were 1.83%-3.35% and 1.44%-3.34%,respectively.For the seeds of Xianyu1619 and Xinyu 108,the error of qualified rate was 0.5%-2.0% and 0.5%-1.8%,and the error of total seed number was 1.47%-3.24% and 1.45%-3.74%,respectively.The seed placement performance detection system based on multi-sensor fusion burrow planter is feasible within the error range of test data analysis and meets the actual detection requirements.