Abstract: A computer system is provided for training a neural network that converts images. Input images are applied to the neural network and a difference in image values is determined between predicted image data and target image data. A Fast Fourier Transform is taken of the difference. The neural network is trained on based the L1 Norm of resulting frequency data.

Abstract: In the held of the broadcasting of services, in particular video, on mobile terminals and more particularly that of the method of authenticating the user of a service from a mobile terminal, the invention proposes a procedure for authenticating the user of a service on a mobile terminal making it possible to refrain from manual copying of an identifier by the user. For this purpose, the user enters his mobile telephone number when he registers. A short text message or an e-mail is then sent to him on his mobile terminal. This short message or e-mail contains an identifier of this user's account. When initially launching the application dedicated to access to the service, said application then runs through the short messages or e-mails stored on the terminal, it identifies the message containing the identifier and extracts it.