Share Email Print

Proceedings Paper

Texture segmentation based on combination of second-order features and spatial information
Author(s): Qingqing Zheng; Nong Sang; Yuehuan Wang
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Texture segmentation is the process of partitioning an image into regions with different textures containing similar group of pixels. Most of segmentation algorithms can be regarded as consisting of two successive processes: feature extraction and feature-based segmentation. In this paper, a new texture segmentation method based on the combination of secondorder features and spatial information is proposed. Our method has been compared with the Only Second-order Features(OSF) based algorithm and the Krishnapuram and Freg's (KF) algorithm by segmenting images from the Brodatz album and a real-life scenery image dataset. The results show that the proposed approach reduces the inaccurate small regions and keeps the edge of contiguous target regions more smooth. In addition, the parameters of filter bank are selected elaborately according to human psychophysics research, thus our algorithm is so intuitive and physiological relevant that it reserves opportunities for further approaches.

Paper Details

Date Published: 30 October 2009
PDF: 7 pages
Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 749519 (30 October 2009); doi: 10.1117/12.832506
Show Author Affiliations
Qingqing Zheng, Huazhong Univ. of Science and Technology (China)
Wuhan Univ. of Science and Technology (China)
Nong Sang, Huazhong Univ. of Science and Technology (China)
Yuehuan Wang, Huazhong Univ. of Science and Technology (China)

Published in SPIE Proceedings Vol. 7495:
MIPPR 2009: Automatic Target Recognition and Image Analysis
Tianxu Zhang; Bruce Hirsch; Zhiguo Cao; Hanqing Lu, Editor(s)

© SPIE. Terms of Use
Back to Top