Department of Measurements and Control Systems Faculty of Automatic Control, Electronics and Computer Science

His work involved the use of scientific methods to improve the efficiency of feature extraction and detection of objects in various applications. As part of the achievement, he presented a collection of nine thematically related publications, covering various aspects of the main issue, namely, they were devoted to the issues of human face detection, noise reduction in thermal images and human figure detection in thermal images, multimodal processing for effective localization and tracking of the human figure, multimodal processing in applications for mobile platforms and proposed a new approach to the problem of detection and estimation of the grinding degree of loose materials, with particular emphasis on copper ore. The methods used in the mentioned applications focused on the methods of pattern search, extraction of low-level features (HoG, LBP), neural networks and combinations of area and edge methods with modified criteria of area homogeneity.