Abstract: Computer image analysis is an automatic image processing, in the process of which the definition and classification of objects located on the image takes place. One of the most important stages of this analysis is image segmentation, by means of which, based on a set of characteristics (color, texture, brightness, etc.), the initial image is divided into many non-intersecting areas. The importance of the stage lies in the significant impact of segmentation on the final result of the analysis.The object of the research is the method of texture segmentation of the image based on the construction and use of luminance dependence matrices. The subject of the research is the effect of textural features on the quality of image segmentation. Special attention is paid to the calculation of the textural attributes and segmentation evaluation criteria.The research methodology is based on the analysis of texture segmentation of images using empirical evaluation criteria and reference segmentation. The main conclusion of the study is the conclusion about the choice of a set of textural features that showed the best segmentation results. This conclusion was made on the basis of the analysis of the values of the selected criteria for assessing the quality of segmentation. The textural segmentation of images and the evaluation criteria were performed on the basis of the developed program in the C ++ programming language. The novelty of the study is in the analysis of textural characteristics regarding the quality of image segmentation, made on their basis.