Abstract. The paper considers the possibility of using neural network structures of an artificial intelligence system for processing photometric data of remote sensing of winter rye crops grown in the conditions of the Leningrad Region on the field of the educational and experimental garden of Saint Petersburg State Agrarian University in 2014–2015. In the process of cultivating plants, various types of treatments were applied: the application of mineral fertilizers, microelements and a microbial biological product. To process the photometric data, the Rosenblatt perceptron was used, which analyzes the similarities and differences in the photometric NDVI profiles of winter rye crops obtained from different variants of the experiment. According to the numerical indicators of vegetation indices, it was possible to construct phase portraits of the trajectory of their movement on the coordinate plane of the field. Further cluster analysis of the data obtained, converted into a square matrix of paired Euclidean distances, made it possible to identify on the dendrogram a grouping of variants in which the connecting components were the use of a microbiological inoculant. When using a biological product, there is a more complete development of plants in crops and their evenness in the field improves. The minimum coefficient of variation was observed for the variant without the use of a biological product, but with the joint use of a complex of all mineral fertilizers (50 phosphorite flour + 50 KCl + 50 ammonium nitrate) and microelements at a dose of 250 kg/ha. Based on the results of the analysis, it can be concluded that the images of the trajectories of the points of the NDVI profiles provide qualitative information reflecting the dynamics of the ontogeny phases of winter rye plants. Based on the nature of the selected sections of these trajectories, it is possible to create a digital map of the experimental field, with the help of which to conduct a protocol for remote diagnostics of the state of crop productivity and make a forecast of their yield during harvesting.