Close
Metadata

%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi@80/2007/09.21.12.42
%2 sid.inpe.br/sibgrapi@80/2007/09.21.12.42.19
%A Bastos, Lucas,
%A Conci, Aura,
%@affiliation Universidade Federal Fluminense
%@affiliation Universidade Federal Fluminense
%T Automatic Texture Segmentation Based on k-means Clustering and Co-occurrence Features
%B Brazilian Symposium on Computer Graphics and Image Processing, 20 (SIBGRAPI)
%D 2007
%E Gonçalves, Luiz,
%E Wu, Shin Ting,
%S Proceedings
%8 Oct. 7-10, 2007
%J Porto Alegre
%I Sociedade Brasileira de Computação
%C Belo Horizonte
%K Haralick features,automatic texture segmentation, grey level co-occurrence.
%X This work presents a method for automatic texture segmentation based on k-means clustering technique and cooccurrence texture features. A set of up to eight features were extracted from a 256 grey-level co-occurrence information. These features were used to segment image regions regarding the textural homogeneity of its areas. As the process of calculating co-occurrence information demands the majority of computational time required,we propose a new methodology based on a grey-level cooccurrence indexed list (GLCIL) for fast element access and highly optimize this step in the algorithm. Besides that, we compare the efficiency of the proposed method against the GLCM and GLCLL algorithms. The GLCIL shows to be the most efficient method in terms of computational time. Additionally, traditional Brodatz textures and others examples of the literature were tested to evaluate the appropriateness and robustness of the method.
%@language en
%3 lucasfinal.pdf


Close