Taras Kovalenko

Full text PDF

Abstract. In this article, we developed and described a method for segmentation of informative textured regions of an image that are close in color and structure, consisting of the stages of primary and secondary segmentation that provide the solution to the problem of localization of image areas. Step-by-step processing of the image by the proposed method ensures maximum elimination of localization errors in false regions. In addition, the transition from one processing step to the next decreases the analyzed amount of information, namely the area of segmented image areas. In order to reduce the time during the practical implementation of the method, it becomes possible to parallelize the processing and solving problems in a time scale close to the real one. The input parameters are an aerial photograph, a priori information about the areas to be segmented, and meteorological and navigation-technical conditions for aerial photography. Output parameters are images with localized informative areas.

Key words: image, texture, pixel, aerial image, digital matrix, texture models, information system, method of segmentation, digital image correction


an international quarterly journal on economics of technology and modelling processes