Font Size: a A A

Design Of 3D Printing Quality Inspection System Based On Machine Vision

Posted on:2021-03-07Degree:MasterType:Thesis
Country:ChinaCandidate:C Y DuanFull Text:PDF
GTID:2428330632451291Subject:Control engineering
Abstract/Summary:PDF Full Text Request
3D printing is a new manufacturing method which integrates digital modeling technology,intelligent control technology and new material technology.It has the advantages of short processing cycle,low cost,high material utilization and so on.However,there is also the problem that errors affect the forming accuracy in the printing process.due to the limitations of the forming principle and device structure,the traditional machining measurement and testing methods are difficult to apply or have low accuracy.therefore,the research on the new technology and method of real-time accuracy detection in 3D printing process is of great significance to improve the accuracy of 3D printing products.Based on the analysis of 3D printing principle,this paper studies the problem of forming accuracy detection of parallel structure 3D printer based on machine vision.According to the structure and working principle of 3D printer,the mechanism of printing defects is analyzed,and the effects of temperature change,driving structure error and transmission mechanism on forming accuracy and the causes of defects are studied.The real-time detection mechanism and workflow of parallel arm 3D printing process based on machine vision are designed,the machine vision structure model is constructed,and the camera spatial layout is determined.The camera coordinate system and image coordinate system are analyzed.The related algorithms of image processing are studied,the image is grayed out and the edge is detected,the surface path data of the image is obtained,and compared with the path data of slicing software to judge the real-time accuracy error.According to the cascading characteristics of 3D printing,the Gaussian filtering method of Canny algorithm is optimized,and different degrees of one-dimensional Gaussian filtering are carried out in X and Y,respectively,to enhance the linear characteristics of the image.A precision detection method based on optimized Canny algorithm and Vector container is designed.The dynamic storage of image information is realized by Vector container.The printing accuracy of the object is judged by detecting the layer spacing,and compared with the slice data,the fluctuation range is obtained.Based on the principle of Hough transform line detection,the types of printing defects are analyzed,and the contour image is obtained by using Canny edge detection algorithm.According to the characteristics of Hough transform line detection and Set container,a defect detection algorithm is designed,which can effectively identify common defect types,classify and distinguish them.
Keywords/Search Tags:Fused deposition molding, 3D printing, Machine vision, Canny edge detection, Hough Transform
PDF Full Text Request
Related items