Selection of Morphological Indicators to Identify Anomalous Areas in Photographic Images of Homogeneous Surfaces
O.A. Nazarkin, Vitalii Vedishchev, Marina Zhuravlyova, Vladimir Alexeev, Pavel Domashnev · 2020 2nd International Conference on Control Systems, Mathematical Modeling, Automation and Energy Efficiency (SUMMA) · 2020
Methods for selection of morphological indicators necessary for classification of anomalous areas on photographic images of homogeneous surfaces are considered. The use of binarization and posterization procedures has been studied, morphological indicators based on them have been proposed. An original scheme for analyzing influence of median filtering on morphological indicators has been developed. Possibility of using the DCT-coefficient to improve reliability of results obtained is considered, local indicators are proposed. Results of application of described procedures for analysis of images of homogeneous surfaces with anomalous areas are presented.