Guilherme A. Lizarzaburu, Renato Candido, Magno T. M. Silva. Restaurando imagens com uma rede neural convolucional. In: Congresso Nacional de Matemática Aplciada e Computacional – CNMAC’2021, 2021, Online Conference. Anais do CNMAC 2021, 2021, pp.1–2. (available only in portuguese).

Abstract

Image restoration aims to attenuate distortions caused in the acquisition process. In this paper, we use a residual convolutional neural network to restore images which were degraded by a Gaussian point spread function. As cost function, we consider a function based on the structural similarity (SSIM) index.

Keywords

Convolutional neural network with residual skip connections, image restoration, gaussian point spread function, structural similarity index.

Downloads

• PDF