Perceptual grouping model for color natural images

Author

Zhong, Jing-jing ; Luo, Si-Wei ; Zou, Qi

Author_Institution

Beijing Jiaotong Univ., Beijing

Volume

2

fYear

2007

fDate

2-4 Nov. 2007

Firstpage

885

Lastpage

890

Abstract

Perceptual grouping is an important mechanism in human visual system, which happened in the middle stage of information processing. In this paper, we mainly care about edge perceptual grouping, and define new measurement of gestalt rules. Proximity, good continuation and similarity of edges are measured by local features, distance, orientation changes and HSV histogram difference. Closure of a contour is assured by global topological property considering edges affinity probability. Experiment results show good performance of our model, especially when gap or occlusion happens.

Keywords

edge detection; image colour analysis; measurement; probability; topology; HSV histogram difference; color natural images; edge perceptual grouping; edge similarity measurement; edges affinity probability; gestalt rules measurement; global topological property; good continuation measurement; human visual system; information processing; perceptual grouping model; proximity measurement; Bayesian methods; Computer vision; Humans; Image analysis; Image color analysis; Image edge detection; Information analysis; Pattern analysis; Psychology; Wavelet analysis; Perceptual grouping; closed contour; gestalt rules; selective attention;

fLanguage

English

Publisher

ieee

Conference_Titel

Wavelet Analysis and Pattern Recognition, 2007. ICWAPR '07. International Conference on

Conference_Location

Beijing

Print_ISBN

978-1-4244-1065-1

Electronic_ISBN

978-1-4244-1066-8

Type

conf

DOI

10.1109/ICWAPR.2007.4420794

Filename

4420794