UDC 004.032.26
DEVELOPMENT OF METHODS AND ALGORITHM FOR FORMALIZING A TEXT REQUEST TO ONLINE SERVICES GENERATING IMAGES THROUGH NEURAL NETWORK TECHNOLOGIE
Yu. L. Leokhin, Dr. Sc. (Tech.), full professor, Deputy Rector for Scientific Work, MTUCI, Moscow, Russia;
orcid.org/0000-0003-3321-4497, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
T. D. Fatkhulin, Ph.D. (Tech.), associate professor, department of MC and IT, MTUCI, Moscow, Russia;
orcid.org/0000-0003-0998-1055, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The problem of studying the influence of a text request compiled by a user on the image quality generated by neural network online services is considered. The aim of the work is to develop methods for compiling a formalized request to neural network services and an algorithm, that allow obtaining correct results of image generation. The relevance of the work is due to the growing interest in the use of neural network technologies in various areas of human life. The domestic online services «Shedevrum» and «Kandinsky 2.1» are considered, which implement diffusion and generative-adversarial generation technologies of image generation, respectively. A technique is given that allows to formalize a request for image generation by neural networks and an algorithm implementing it. The approbation of the developed methods and algorithm is given by entering human language requests and formalized requests and analyzing images generated by neural networks. In conclusion, the results of the analysis are presented and findings are shown
Key words: : neural networks, service, image, generation, recommendations, technique, algorithm, request, methods.