Obtenção de uma matriz de quantização otimizada para imagens de ultra-sonografia e ressonância magnética

Detalhes bibliográficos
Ano de defesa: 2005
Autor(a) principal: Coelho, Júlio Cézar
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal de Uberlândia
BR
Programa de Pós-graduação em Engenharia Elétrica
Engenharias
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/123456789/14428
Resumo: The amount of medical images is growing, and in hospitals and clinics, where the number of patients is high, the space requested for all those information is very big, and that becomes a problem. One of the ways to reduce this problem is to compress those images. The aim of this work is to obtain one quantization matrix for ultrasonic echo images and magnetic resonance using the arithmetic average method. This matrix should provide a small rate of bits/pixel and a satisfactory quality for the medical diagnosis of those images. In order to obtain that quantization matrix were found four quantization matrix from Q2 to Q5, obtained through the use of five ultrasonic echo images and 454 magnetic resonance images. The results obtained in the tests of this work showed that from the quantization matrix Q2 to Q5 obtained in this dissertation, one that presented the smallest rate of bits/pixel was the Q5 with value equals to 0,38081. This rate is 29 % smaller than the rate of bits/pixel of the matrix Q6 (matrix obtained by Hamamoto for ultrasonic echo images) whose value is equal to 0,53638 and 37,55 % smaller than the rate of bits/pixel of the matrix Q1 (quantization matrix of the pattern JPEG) whose value is equal to 0,61141. By the fact that the percentile differences of PSNRs of the matrix Q2 to Q6 in relation to PSNR of the matrix Q1 are smaller than 10 % and that the six consulted doctors' opinion on the quality of the ultrasonic echo images and resonance magnetic reconstructed using the matrix of Q1 to Q6 are satisfactory for them to accomplish their diagnoses, it can be considered that those images present the same quality. Therefore, the quantization matrix Q5 can be considered as being the quantization matrix optimized for those images, because it presents a small rate of bits/pixel and a satisfactory quality.