%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2012/12.14.13.28
%2 sid.inpe.br/sibgrapi/2012/12.14.13.28.15
%@isbn 978-85-7669-271-3
%T Non-adaptive robust filters for speckle noise reduction
%D 1993
%A Frery, Alejandro César,
%A Sant’Anna, Sidnei J. S.,
%@affiliation Divisão de Processamento de Imagens do Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Divisão de Processamento de Imagens do Instituto Nacional de Pesquisas Espaciais (INPE)
%E Figueiredo, Luiz Henrique de,
%E Gomes, Jonas de Miranda,
%B Simpósio Brasileiro de Computação Gráfica e Processamento de Imagens, 6 (SIBGRAPI)
%C Recife, PE, Brazil
%8 19-22 Oct. 1993
%I Sociedade Brasileira de Computação
%J Porto Alegre
%V 1
%P 165-174
%S Anais
%K estimation, filtering, maximum likelihood, multiplicative models, order statistics, robustness, speckle, statistics, synthetic aperture radar.
%X After briefly reviewing some classical filters for speckle removal, five filters with characteristics of robustness, suitable for speckle noise reduction, are derived and implemented. These filters are the ones based on the trimmed maximum likelihood and moments estimators, the ones based on the median, on the inter-quartile range and on the median absolute deviation. Assuming that observations within a synthetic aperture radar image are outcomes of independent Rayleigh random variables, these filters exhibit a good performance from both the signal-to-noise reduction and from the edge preserving criteria. The problem of filtering in an image is posed as an estimation problem.
%9 Processamento de Imagens II
%@language en
%3 20 Non adaptive robust filters for speckle noise reduction.pdf