Image segmentation using Directionlet-domain hidden Markov tree models

Author

Bai, Jing ; Zhao, Jiaqi ; Jiao, Lc

Author_Institution

Key Lab. of Intell. Perception & Image Understanding of Minist. of Educ. of China, Xidian Univ., Xi´´an, China

Volume

2

fYear

2011

fDate

24-27 Oct. 2011

Firstpage

1615

Lastpage

1618

Abstract

In this paper, we modeled the Directionlet coefficients of an image using hidden Markov tree (HMT) model and obtained the image segmentation results using model parameter training, multi-scale likelihood computation and the background of neighborhood-based maximum posterior probability. We demonstrate the performance of the proposed method with synthetic mosaic images and SAR images. The experiment results showed that our method obtained more exact boundary and uniform regions.

Keywords

hidden Markov models; image segmentation; radar imaging; synthetic aperture radar; trees (mathematics); Directionlet-domain hidden Markov tree models; SAR images; image segmentation; model parameter training; multiscale likelihood computation; neighborhood-based maximum posterior probability; synthetic mosaic images; Computational modeling; Hidden Markov models; Image edge detection; Image segmentation; Synthetic aperture radar; Training; Transforms; Contourlet; Directionlet; Hidden Markov tree(HMT); Image segmentation;

fLanguage

English

Publisher

ieee

Conference_Titel

Radar (Radar), 2011 IEEE CIE International Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4244-8444-7

Type

conf

DOI

10.1109/CIE-Radar.2011.6159874

Filename

6159874