Concrete CT Image Segmentation method based on Markov Random Field

Author

Zhao Liang ; Li Chang-Hua ; Dang Faning ; Chen Deng-Feng ; Xu Sheng-jun

Author_Institution

Dept. of Info & Autom., Xi´an Univ. of Archit. & Technol., Xi´an, China

fYear

2010

fDate

29-31 July 2010

Firstpage

2677

Lastpage

2680

Abstract

The article based on Markov Random Field (MRF) image segmentation model and beyes theory, the segmentation issue is transformed to the Maximum A Posteriori (MAP). The forecast of parameter arithmetic is provide. The article proposed a modified version of the metropolis dynamics (MMD) simulated annealing arithmetic. Experimental results are compared to those obtained by the Metropolis algorithm, the Gibbs sampler and ICM (Iterated Conditional Mode). The efficiency and precision is more improve. Using MMD on Concrete CT Image Segmentation can reflect the interior spatial distribution of the concrete materials on deformation, and afford an effective method on concrete meso-structure CT image study on Architecture projection.

Keywords

Markov processes; civil engineering computing; computerised tomography; concrete; image segmentation; maximum likelihood estimation; random processes; sampling methods; simulated annealing; Gibbs sampler; Markov random field; architecture projection; concrete CT image segmentation; concrete mesostructure; iterated conditional mode; maximum a posteriori; metropolis algorithm; metropolis dynamics; parameter arithmetic; simulated annealing arithmetic; spatial distribution; Cavity resonators; Computed tomography; Concrete; Heuristic algorithms; Image segmentation; Markov random fields; Pixel; Concrete CT; Image Segmentation; MMD; Markov Random Field (MRF);

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (CCC), 2010 29th Chinese

Conference_Location

Beijing

Print_ISBN

978-1-4244-6263-6

Type

conf

Filename

5573210