Design And Development Of Self-walking System Of Intelligent Bionic Picking Tobacco

This topic comes from the research and development of the key project of Henan Provincial Department of science and technology research of the Ministry of education.The purpose of this paper is to develop a can realize intelligent bionic picking set automatic leaf picking basket in one of the self walking type tobacco harvester,hand picked leaves completely the process of mechanization and automation,completely change the traditional manual picking leaf farming methods.After the promotion,there is a huge market potential,and produce better technical and economic benefits.In this paper,under the background of automatic bionic picking of tobacco leaf and based on the basis of binocular vision technology,image processing technology,automatic control and so on,researching and designing the intelligent tobacco leaf bionic picking system.Firstly,discusses the difference of agricultural robot in the characteristics of the industrial robot,and in the domestic and foreign development history,research status and future development trend,followed by a discussion of the development history and application of machine vision technology,and then separately from the hardware and software design and the mechanical structure of the design are discussed in this paper.Finally in test based on the innovation and development trend of the summary and expectation.The digital image processing part includes the process of image acquisition,image preprocessing,image feature extraction and picking point of tobacco plant.In the image acquisition stage of tobacco plant,the gradient function and spectrum function of combining the clarity evaluation function;image pretreatment process of tobacco plant analysis the principle of spatial filtering,frequency domain filtering and wavelet denoising and compare the three methods in the pros and cons of image denoising effect;in smoke image enhancement aspects of the traditional method and based on the principle of mathematical morphology method and analysis to compare the two methods in image enhancement processing in the process of enhancement effect is good or bad;in the feature extraction and summary based on the former research of feature extraction algorithm to optimize,the binocular stereo vision technology to determine leaf bionic picking point.Tobacco image feature extraction,summary of previous studies based on feature extraction algorithm to optimize,finally using binocular vision technology lock stalk position of tobacco plant,using characteristics of mature tobacco hierarchical prior adjustment picking height,picking the maturity of tobacco leaves.This paper designs a self walking tobacco leaf intelligent bionic picking system,the whole system includes bionic picking system,intelligent control system and self walking mechanism.Bionic picking system fully simulated artificial picking tobacco process,reduce the harm of tobacco and tobacco;intelligent control system comprises a tobacco image processing system and the vehicle running system.According to the planting characteristics of tobacco plant design mechanical ontology,can be achieved at the same time to carry out the picking process of two lines of tobacco,the paper used Solidworks drawing software to draw the three view of the mechanical Ontology.Finally,the tobacco stem recognition experiment is conducted in the national tobacco physiological and biochemical bases(Henan Agricultural University).The experimental’s results show that under the different conditions of tobacco stem average recognition accuracy was 72.67%.