Metallographic Geometric Parameters Based On Digital Image Processing, Quantitative Analysis And Research

This many a day, the metallography is always assessed by means of traditionally determine the nature or half quantitative analysis technology. With the development of quantitative metallography theory and digital image process technology, the quantitative metallography is much widely used in and become a branch of metallographic. In fact, the quantitative metallography is a synthetically application science with relating to material science, computer science, digital image processing technology and pattern recognition techniques.The quantitative analyze of metallography is about how to determinately express the organization of metallography by measurable and calculable parameters, and find the relationship and ruler between material performance and these parameters.Because of the unsymmetrical distributing of particles in metallography no parameter can be ascertained by several data measured in a visual area. It is absolutely necessarily to have a great lot data to assure the reliability of results.In this paper the new methods of deblurring, thinning and edge reconstructing are put forward by means of experiments. The new methods decrease the difference of results tested by Digital Micrographic Software. Also with these data, a conclusion which grain size can be described by fractal dimension is presented.The dissertation includes seven chapters and the contents are as follow:In chapter 1, the paper includes the importance of the computer-aided assessment and the development of relating research, and the significance of the application of fractal dimension in quantitative metallographic area.In chapter 2, the dissertation introduce the conception of quantitative metallographic, some of denotations, formulations, and the method of manual testing.In chapter 3, a digital metallographic image pre-processing with a new algorithm for shadow remedy and edge detection is presented to make images clearly.In chapter 4, the algorithms based on significantly different methods for grain edge detection and reconstruction are used to demonstrate the relationship more clearly and to give a general view of computer-aided image processing.In chapter 5, the paper introduce the design, algorithms, flow chat and some functions of The Digital Micrographic Tester developed by VC++ and show its interface in the last section.In chapter 6, the application of fractal on grain size analysis is presented by means of results tested by Digital Micrographic.In chapter 7, the work of this dissertation is summarized and the further research work is looked ahead.