Industrial Automatic Control Systems and Controllers Annotation << Back Blur Measure as an Informative Parameter in Solving Pattern Recognition Problems Loktev D.A., Izotov K.A., Loktev A.A. In this paper, we consider the problem of image quality assessment depending on the tasks under study, including computer vision tasks such as object detection and recognition. Attention is focused on the fact that the image quality depends on both objective characteristics and subjective perception, which necessitates the selection of optimal parameters depending on the task being solved. The existing basic informative parameters such as contrast, sharpness, signal-to–noise ratio, and brightness dispersion are described, which serve as the basis for further signal processing, including two-dimensional ones such as images. From the point of view of completeness of the information contained, it is proposed to use a measure of image blurring to determine image quality. Special attention is paid to the effect of blurring on the quality of image perception and on the results of automated object recognition, for example, using the YOLO algorithm. It examines how the blur parameters can vary for static and moving objects and how this affects the recognition accuracy. The results obtained make it possible to develop more effective image processing methods for specific computer vision tasks. Keywords: informative parameter, blur measure, pattern recognition, image quality, signal processing. DOI: 10.25791/asu.8.2025.1602 Pp. 11-19.