%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2015/07.22.14.43
%2 sid.inpe.br/sibgrapi/2015/07.22.14.43.42
%T Avaliação de Desempenho de Atributos de Textura 3D e Nitidez de borda 3D na recuperação de nódulos pulmonares similares
%D 2015
%A Lima, Lucas Lins,
%A Ferreira Junior, José Raniery,
%A Oliveira, Marcelo Costa,
%@affiliation Universidade Federal de Alagoas (UFAL)
%@affiliation Universidade Federal de Alagoas (UFAL)
%@affiliation Universidade Federal de Alagoas (UFAL)
%E Vieira, Thales Miranda de Almeida,
%E Mello, Vinicius Moreira,
%B Conference on Graphics, Patterns and Images, 28 (SIBGRAPI)
%C Salvador, BA, Brazil
%8 26-29 Aug. 2015
%I Sociedade Brasileira de Computação
%J Porto Alegre
%S Proceedings
%K recuperaçãoo de imagens baseada em conteúdo, unidade de processamento gráfico.
%X The lung cancer is the major causer of cancer worldwide and allied to this, medical imaging production has been increasingly growing in the medical field. To combat these two challenges, the Content-Based Image Retrieval (CBIR) tool can provide decision support to experts, enabling them find similar images in a database of images from a reference image. However, a crucial step of CBIR is the stage of Image Similarity Analysis (ISA), and one of the ISAs challenge is the processing time to perform all comparisons. This article presents a performance comparison between two techniques of attributes: 3D Texture Attributes and 3D Margin Sharpness, and also both concatenated, both in CPU and in GPU with the use of the OpenCL framework, and the Manhattan Distance to do the comparisons of similarity.
%@language pt
%3 Artigo - Avaliação de performance de Atributos de Texura 3D e nitidez de Borda 3D na recuperação de nódulos pulmonares.pdf.pdf