Page 12 - КЛАСИФІКАЦІЯ ЛОКАЛІЗОВАНИХ ДЕФЕКТІВ ЗВАРНИХ ШВІВ НА РЕНТГЕНІВСЬКИХ ЗОБРАЖЕННЯХ ТРУБ

12

surface objects of the selected area. It was found that oxide ceramic coatings on the

alloy MA 18, which were synthesized in the electrolyte 3 g/k KOH + 2 g/l (RS) at the

2
densities of cathode and anode currents of 20 A/dm for 180 min have the maximum
2
number of surface defects with an area of 2 µm .
A method for image analysis of the surface layer of an aluminum alloy

reinforced with silicon carbide particles has been developed. 3D modeling of silicon

carbide particles of different sizes based on cutting the vertices of the cube is

performed. Simulations for different percentages of particle filling in the surface

layer of aluminum alloy based on uniform distribution are performed. This made it

possible to compare experimental and simulated surface layers and further predict the

mechanical properties of such surfaces.

A number of methods for segmentation of painted coatings of structural

materials are proposed. For most coatings, damage should be assessed by the

intensity of the formation of corrosion products of the base material, based either on

their negative impact on the appearance of the product, or on the inability to use the

product due to its corrosion damage. The diagnostic feature is the ratio of the area of

rust areas to the surface area.

A method of noise-resistant segmentation of corrosion damage on images of

painted surfaces of arbitrary color on the basis of a cylindrical color model and one-
scale retinex has been developed. This increased the reliability of the quantitative

assessment of the area of corroded areas.
Thus, as a result of research conducted in the dissertation, a set of methods for

diagnosing the state of materials and structures, based on the segmentation of their
digital images.

In practice, the results of the dissertation were used in the Ugersk diagnostic

section of the Ivano-Frankivsk territorial center of the Research and Production

Center for Technical Diagnostics "Techdiagaz", c. Uhersko, Lviv region when

analyzing the defect of welded joints by their radiographic images; at the State

Enterprise Engineering Center "Lvivanticor", Lviv in the analysis of images of

corroded surfaces; in LLC DIALAB LTD, Odessa at the analysis of images of the