Font Size: a A A

The Research On Quality Of Medical Infusion Container Combination Cover Visual Inspection System

Posted on:2015-03-22Degree:MasterType:Thesis
Country:ChinaCandidate:S C HeFull Text:PDF
GTID:2308330461996668Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Polypropylene composite cover is a packaging materials used for plastic medical infusion container and direct contact with liquid, so its effect to the quality of cleanliness of liquids and related to patient safety. The thesis studied the combination cover indices such as appearance,assembly quality and cleanliness of on-line automatic detection method using machine vision inspection technology for developed the high speed automatic testing machine and applied to the combination cover of automatic production line.It alternatives to traditional methods of detection and improves the production efficiency and automation of manufacturer so has a great social and economic benefits.The thesis main research work is as follows:First, this paper discusses the combination cover market background and the present situation of the artificial light inspection, analyzed the advantages of machine vision inspection. This paper expounds the definition of machine vision, machine vision system, the development of machine vision and research status at home and abroad and application of machine vision technology in all aspects of production and life.Second, in the design of combination lid defect detection system requirements and detect objects under the careful analysis, mechanical executive system was design, electrical control system, visual imaging system.Again, in view of the combination of internal and external cover cover inspection, semi-finished products and finished products testing of different characteristics, has designed three sets of complete and practical detection algorithm:internal and external cover defect detection based on the spread of PM defect extraction algorithm, to extract the defect area and the defective area, and then use SVM to the feature of defect area and the defect for classification, and cross validation of choosing the optimal classification parameters; In view of the contour defects, a circle fitting method based on RANSAC. Semi-finished products detection based on edge profile roundness of the stopper is to determine whether the stopper position level; Product testing first fixing the detection area, and then using the adaptive threshold segmentation of automatic threshold selection parameters, solves the diffuse the impact of the light, the last thread search method is used to determine defects. Finally, completed the combination cover defects detection system of software development. The function of the software modules and realization method are introduced; For multithreading technology and modular program design method based on COM components and Modbus protocol was analyzed and implemented.
Keywords/Search Tags:Machine vision, Defect detection, Edge extracting, Circle fitting, Reverse PM diffusion, SVM
PDF Full Text Request
Related items