Font Size: a A A

Study On The Optical Electricity Checking System Of The Cottonseeds Color Sorting Equipment

Posted on:2007-12-14Degree:MasterType:Thesis
Country:ChinaCandidate:J B LiFull Text:PDF
GTID:2133360185465080Subject:Agricultural mechanization project
Abstract/Summary:PDF Full Text Request
High-quality cottonseeds are the carrier of advanced technology and the basis that Xinjiang BingTuan extend the precision sowing and develop the exact agriculture. In recent years, due to the special climate and geography causation, the maturity degree of cottonseeds have obviously descended .The proportion that the"red seeds"occupied have increased. Furthermore during the machining process when the cottonseeds pass through the cotton gin, linter, acentric platen, rising machine, polishing machine it will result in a great damage to the cottonseeds. The quality of cottonseeds is seriously affected by the"red seeds"and the"crashed grains". The germinating rate of the cottonseeds descends sharply which prevent the development of precision sowing. At present cottonseeds selection's working procedures mainly adopt the wind filter and gravity selection which mainly depend on mechanism and physical principle. Making use of the specific gravity and air characteristics of seeds to select the cottonseeds and eliminate the impurity but the specific gravity and the dimension of the"red seeds"and the"crashed grains"are similar with the plump seeds, traditional selection can not make a precise decision. At present the method to eliminate the"red seeds"and the"crashed grains"is depended on the manual work. However, this method need great labor strength and high cost but the produce efficiency is low, furthermore the selection effect is influenced to a certain extent by the manmade factors.The optical electricity checking system of cottonseeds is based on the machine sense of vision to check and classify the cottonseeds. Based on the characteristics of the cottonseeds'color and fragmentation characteristics, setting up the optical system and selecting three parameters which according with the cottonseeds appearance characteristic: color, grain model, rotundity degree. By using the nervous network we classify the cottonseeds into two grades eligible product and ineligible product.The optical electricity checking system of cottonseeds mainly consists in five modules: picture collection, picture pretreatment, characteristic parameter, nervous network training and classification.The hardware system of picture collection conclude: host computer system, America costar SI-M350 camera, Matrox Meteor-â…¡/Multi-channel picture collection card, COMPUTAR FA TEC-55 industrial lens, LED lamp house, experimental platform and so on .The gathered pictures are carried through the lightness amendment the linearity transformation of lightness and median sieved wave. Distilling the characteristic parameter and fringe character of the picture and taking a research to the crashed cotton seeds.BP network ranking system adopt three layer of network construction, three input nodes, eight concealed layer node, two output nodes. The deliver function adopts S model function, between the input layer and concealed layer we adopt double bent tangent function and between the concealed layer we adopt linearity function. During the experiment, we use ten eligible stylebooks and ten ineligible stylebooks to carry out the network training, retaking one hundred stylebooks for checking. According to the experiment result we can draw the conclusion that the nicety rate can obtained over eighty seven percent.The software system adopt the VC++6.0 and Adept HexSight3.3 video software package exploitation. After the experiment it testifies that the interface of the software is friendly and the operation is simple.
Keywords/Search Tags:cottonseeds, optical electricity checking, machine vision, neural network
PDF Full Text Request
Related items