Design And Implementation Of Online Detection System For Remanufactured Tobacco Leaf And Tobacco

With the development of society and technology,the tobacco industry is moving towards the goal of building an innovative industry,and constantly exploring and applying new technologies.In the process of cigarette production,the different structure of tobacco will directly affect the end of the cigarette silk amount,tar release,absorption resistance and other important indicators related to sensory products.Therefore,the control and measurement of tobacco length is of great significance in cigarette production.In recent years,with the development of industrial technology,compared with off-line manual detection,online tobacco length detection has higher efficiency and more accurate detection effect.However,the existing on-line detection technology still has some problems such as difficulty in sample acquisition and dispersion,clumping or stacking of tobacco,resulting in low measurement accuracy.In order to solve the above problems,the robot arm,conveyor belt and machine vision technology are applied in this topic,and the image processing algorithm is designed to achieve the accurate detection of tobacco length.The main research content of this thesis is as follows:(1)Design the overall scheme of online detection system.According to the testing requirements,the robot arm module,the conveyor belt module and the mechanism of automatic dispersing and thinning tobacco were designed.The machine arm,industrial camera,optical lens,light source and other hardware equipment of the system are selected.Visual Studio 2019 and Halcon as development platforms for the system software.(2)The image processing technology in the online detection system of recycled tobacco is studied.Halcon calibration method was used to calibrate the camera.Firstly,the median filter method is used to denoise the tobacco image,and then the image morphology processing is carried out to make the tobacco detection effect more excellent.Secondly,the region of interest is segmented by the global threshold segmentation method.In order to better solve the problem that the outline of the overlapping or intersection of the recycled tobacco can not be determined,a method of cross tobacco detection is proposed.Finally,the length of the tobacco is calculated accurately.(3)Carry out experimental research and analysis.The on-line testing platform of reconstituted tobacco was summarized,and then the design principle and application of the tobacco length testing system were introduced.Finally,the experimental results of tobacco length detection are analyzed and collated.The experimental results show that the method of online detection system designed in this thesis is feasible,and the accuracy of the detection results is relatively high,which can basically meet the production requirements.In this thesis,a set of online test system for reconstituted tobacco was designed and tested.The experimental results show that the average relative error of tobacco length measured by the system is about 2%,and the standard deviation is less than 0.04 mm,which verifies the high precision and efficiency of the online detection system for recycled tobacco.