Zhuravel I. Information technology of automated analysis of metallographic and fractographic images

The dissertation is devoted to the development of information technology, which provides an increase in the level of automation, accuracy and speed of analysis of metallographic and fractographic images to assess the technical state of materials of responsible industrial objects through the construction of new methods of quantitative evaluation of geometry and the distribution of elements of their microstructure. The developed information technology covers all the main stages of research of images in metallography - from their formation, allocation of interest objects, geometric analysis, to visualization of the results. Among the developed methods, one should distinguish: the method of binarization of the halftone images using the optimal, in terms of detail, threshold value, which is calculated using the correlation function; the method of calculating the fractal dimension, which, unlike the known approaches, makes it possible to correctly process halftone images without the need for their binary representation; a method for determining the average grain size of a metal by a metallographic image, which allows automating the calculation process and makes it possible to analyze metallographic images with unclosed grain boundaries.