| China is the main origin of tea,one of the world’s largest tea cultivation,production and export country.In the recent year,tea industry of our country is fast-developing,but the competition becoming increasingly fierce.With the large-scale tea planting,labor costs are getting higher and higher.It is known as the tea workers wages rose more than 10% per year,artificial wage expenditure accounted for more than 40% of sales revenue.Even so,in inland Sichuan and southeast coastal areas appear the phenomenon of could not employing anyone to pick tea.Therefore,it is necessary to popilarize the mechanism of mechanical plucking and machine-made to solve the problem of labor shortage.For ordinary tea,the main picking method is manual picking and portable or backpack picking machine,these harvesting methods not only consumes a lot of manpower and material resources,but also has low picking efficiency.The picking process will be affected by the subjective factors of the pickers,causing mechanical injuries of tea.According to the present situation of bulk tea picking,this thesis designs and implements an intelligentized automatic cutting tea-plucking machine based on machine vision.The main research work and results of this thesis are as follows:1.According to the characteristics of the collected tea image,a series of corresponding algorithms of image preprocessing is researched which can realize the image filtering,gray image,image rotation,distortion correction and so on.2.In order to achieve rapid identification and extraction of leaves,a low complexity and high efficient algorithm is proposed.First of all,the use of OTSU(maximum difference method)to remove the background of image.Then OTSU is used twice on the tea image.The result includes leaves area and a few small holes and burr.Finally,morphological algorithm is applied to eliminate the burr and holes.3.The height of the cutter is needed to be adjusted intelligently according to the current situation of the tea to be cut.So the location of the cutter arc in the image must be detected.The camera is mounted on the fuselage of the machine,the cutter blade is unable to obtain directly because of the block of tea.So cutter is localized indirectly by the beam which rigid connected with the cutter using template match.4.Two picking modes are designed according to the growth characteristics of tea of different seasons and picking characteristics in south.(1)Fixed picking mode,this mode is only for cutting kinfe height rough control and level control.It can also be used for pruning and shaping the tea bushes.(2)Intelligent control mode,adjust cutter intelligently based on the leaf area ratio and average height.5.Level adjustment is designed to help cutter to keep horizontal and make cutter line consistently with tea canopy surface by angle sensor.Make sure that the tea-plucking machine can work in the hilly land.6.The application system is designed and implemented.The system is consists of the hardware device and the software system.The design of hardware device mainly includes the lower computer,the IPC,the tilt sensor and the image sensor.The software mainly includes preprocessing module,leaves detection module,cutter localization module and servo control module.In this thesis,automatic leaves detection and intelligent adjustment control of cutter are designed and implemented.The main modules and corresponding algorithms are introduced in detail,using experiments to illustrate.The experimental results show that intelligentized automatic cutting system proposed in this thesis can realize intelligent cutting of tea-plucking machine.Not only reduces the labor intensity,but also saves the labor costs.The overall quality of tea is improved in the case of no mechanical injuries of tea. |
