Ecological Systems and Devices Annotation << Back ANALYSIS OF IMAGES OF AGRICULTURAL PRODUCTS IN CONDITIONS OF UNCERTAINTY Alekseev V.V., Gromov Yu.Yu., Divin A.G., Lakomov D.V. This paper investigates the possibility of applying the algorithm of the Canny operator augmented with Robinson for recognizing objects in an image and detection of defects on objects. Operator Robinson, with coagulation of the original image in 4 directions, allows for a more accurate look of the contour, compared to other operators (Roberts, Sobel, Prewitt) using 2 directions of convolution. The results of the study of this method in the detection of defects in vegetables. Based on this study, recommendations on the choice of parameters of the algorithm. The fi ndings of this article can be used in vision systems are applicable for detecting defects in fruits and vegetables. Keywords: detection; image fuzzy logic; FCM; metric; the color model; uncertainty. Contacts: E-mail: vvalex1961@mail.ru Pp. 28-40.