Eliminação de ruídos em imagens de ultrassonografia via métodos variacionais
Ano de defesa: | 2013 |
---|---|
Autor(a) principal: | |
Orientador(a): | |
Banca de defesa: | |
Tipo de documento: | Dissertação |
Tipo de acesso: | Acesso aberto |
Idioma: | por |
Instituição de defesa: |
Universidade Federal de Uberlândia
BR Programa de Pós-graduação em Ciência da Computação Ciências Exatas e da Terra UFU |
Programa de Pós-Graduação: |
Não Informado pela instituição
|
Departamento: |
Não Informado pela instituição
|
País: |
Não Informado pela instituição
|
Palavras-chave em Português: | |
Link de acesso: | https://repositorio.ufu.br/handle/123456789/12574 |
Resumo: | Medical ultrasound is one of the most used tools for obtaining diagnostic, being a noninvasive technique, with high efficiency, low cost and real-time visualization. But these images are usually contaminated with a type of multiplicative noise known as speckle noise. This noise superimposed on the image granules that distorts and complicates the analysis of the diagnosis. Despite the removal of multiplicative noises has not been studied so extensively as the elimination of additive noise, there are some works that propose solutions to eliminate this type of noise, including works using variational methods. The variational methods have a smoothing term and a fidelity term, which are responsible for smoothing and preserving image characteristics, respectively. Typically, a balance is made between these two terms that will be used for all pixels in the image. This report will propose a variational method in order to reduce speckle noise in ultrasound images. The differential of the proposed method over other methods is the inclusion of a function in order to detect the locations of pixels with high noise level and pixels containing edges, ie, important features of the image that should be preserved. The function will adjust the fidelity and smoothing of the functional, so that this balance causes smoothing to be more severe in very noisy pixels and fidelity is more intense at the image edges. |