UDC 004.85
DETECTION OF TEXTURE ANOMALIES IN LEATHER IMAGES
А. V. Levitin, Ph.D. (Tech.), associate professor, Department of automation and information technologies in control, RSREU, Ryazan, Russia; orcid.org/0000-0002-4829-7398, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
V. S. Muraviev, Ph.D. (Tech.), associate professor, Department of automation and information technologies in control, RSREU, Ryazan, Russia;
orcid.org/0000-0003-4261-859X, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The solution of the problem for texture detection on images containing abnormal areas using one-class support vector machine (one-class SVM) and Gaussian distribution mixture model (GMM) is considered. Similar problems may arise in various applications, e.g., in quality control of product manufacturing. Texture descriptors based on local binary patterns (LBP) that are invariant to image rotation are used as image features in this work. The aim is a statistical study of the above approaches, which shows the advantage of support vector machine to solve the problem in the formulation presented. During the research the algorithm parameters have been adjusted to provide the best results. Segmentation results of skin textures with image processing in a «jumping» window using support vector machine are given.
Key words:: classification, segmentation, texture, descriptor, skin image, support vector machine, Gaussian mixture model, local binary patterns.