Restauração de imagens coloridas usando uma rede neural convolucional residual
Guilherme A. Lizarzaburu, Renato Candido, Magno T. M. Silva. Restauração de imagens coloridas usando uma rede neural convolucional residual. In: Simpósio Brasileiro de Telecomunicações – SBrT’2020, 2020, Online Conference. Anais do SBrT 2020, 2020, pp.1–5 (available only in portuguese).
Image restoration aims to attenuate distortions caused in the acquisition process. In this paper, we use a residual convolutional neural network to restore colored images which were degraded by a Gaussian point spread function. As cost function, we consider the mean square error, and a function based on the structural similarity (SSIM) index. By means of simulations, we verify that the best restoration results were achieved with the function based on SSIM.
Convolutional neural network with residual skip connections, image restoration, gaussian point spread function, structural similarity index.