Font Size: a A A

Research On Measurement Of Workpiece Surface Quality Based On Fractal Theory And Neural Network

Posted on:2007-09-19Degree:MasterType:Thesis
Country:ChinaCandidate:D HuFull Text:PDF
GTID:2121360185484672Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
With the development of technology and society, there is higher and higher requirement in surface quality of mechanical production. As an important merit number, that whether it good or not plays an important role in assemblage quality, fitting character, performance and service life of mechanical production. So it is very necessary to study the computer visual inspecting system applying in the inspect of workpiece surface. The key to study the surface quality inspect is pick-up the character parameter which is used for judging the surface quality. This paper deeply studies the method of pick-up texture space between based on signal analysis algorithm and the method of pick-up Fractal Dimension based on fractal theory. Finally use the neural network to fuse the two parameters and make a judgment whether the quality of workpiece surface is qualified. The main research contents are as follows:(1) This paper talks about the recent development and significance of the research in automatic measurement, which is used to inspect surface quality of workpiece, analyzes the surface and shape character of the study object in this paper. And also introduces the application state of fractal theory in the surface and shape analysis.(2) Studies the angle-autocorrection method for the workpiece's process texture, and emphasizes the measurement to pick-up process texture distance parameter by using FFT.(3) Analyses the influence that process texture plays upon the judgment of surface quality about fractal dimension, puts forth the method to eliminate process texture. And gets the fractal dimension from the processed gray image by using method of box dimension, the calculation result indicates that it effectively represents the quality of the workpiece surface.(4) Studies the workpiece surface quality inspection .method using RBF neural network. It includes network input parameter choice, network design, network learning sample construction.(5) Studies the Mixed Programming technology of VB,MATLAB and SQL...
Keywords/Search Tags:workpiece surface quality, texture space between, fractal dimension, RBF neural network
PDF Full Text Request
Related items