Fast non-local image denoising using a hardware implementation

Lucas Lucena GambarraJosé Antonio Gomes de LimaHamilton Soares da SilvaLeonardo Vidal BatistaYuri Gonzaga G. da Costa

The problem of noise removal is a critical issue in image restoration because it must maintain the integrity of relevant image information. Considering all the necessary tasks for a quantitative analysis of images, the noise removal is an essential step to increase their quality and improve theperformance of such analysis. For non-local means (NLM) denoising approach presented by Buades et al., notable denoising results are obtained at the expense of high computational cost. This paper proposes a hardware implementation of the NLM algorithm with less computation time. Besides being faster than the software approach, this implementation produces comparable results in terms of mean-squared error (MSE) and perceptual image quality.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: