Page 8 - КЛАСИФІКАЦІЯ ЛОКАЛІЗОВАНИХ ДЕФЕКТІВ ЗВАРНИХ ШВІВ НА РЕНТГЕНІВСЬКИХ ЗОБРАЖЕННЯХ ТРУБ

8

ABSTRACT

Ivasenko I.B. Estimation of inhomogeneity and damage of construction

materials by methods of structure-adaptive image processing. – Manuscript.

Thesis for the Doctor of sciences degree in technical science on speciality

05.12.10 – Diagnostics of materials and constructions. – H.V. Karpenko Physico-

Mechanical Institute of National Academy of Sciences of Ukraine, Lviv, 2020.

The dissertation solves an important scientific and technical problem, which is

to develop methodological foundations for building methods of structure-adaptive

processing, modeling and analysis of images obtained by various physical methods

and creating computer-based tools for assessing the condition of materials and

structural elements.

The development of methods for noise-resistant image processing that provide
stable recognition, classification and quantitative analysis of parameters and

characteristics of the studied materials and surfaces is an important task of non-
destructive testing. Solving of these scientific and technical problems of diagnostics

requires the development of new methods of digital image processing. One of such
research areas is the use of structural elements. Structural element operations identify

or act on groups of image elements that belong to the same object or image fragment.

Fragments can be identified based on features such as gray level, color, or features

such as edges, corners, texture, and so on. Structural element operations can result in

both images and descriptions of objects, locations in the image, and so on. This

became the basis for the construction of new methods of technical diagnostics for the

analysis of digital images.

The developed methods include image pre-processing, localization of objects

of interest, segmentation, quantitative parameter estimation. It is necessary to

highlight the localization of welds and determining the size of welding defects on a

complex-structured background, analysis of corrosion damage on images of painted

surfaces of arbitrary color, modeling and analysis of laser reinforcement of aluminum