An Active Contour-Based Unsupervised Texture Segmentation

doi:10.7523/j.issn.2095-6134.2005.5.012

Abstract

Abstract:

In this paper, an unsupervised texture image segmentation method based on active contour is presented. Gabor filters and the anisotropic diffusion method are combined to obtain the texture feature for segmentation. A small set of Gabor filters are used to extract scale and orientation properties of the texture, and original image is also included as a feature channel for the intensity informat ion of the texture. If the diffusion function based on total variation is chosen, vector anisotropic diffusion can smooth the details away with the region border reserved in every channel. Finally the geometric MDL active contour image segmentation algorithm is expanded to vector-value data to segment the features vectors. Experiments on various kinds of texture images show that the method is effective.

Key words: anisotropic diffusion equation, Gabor filter, active contour, texture segmentation

CLC Number:

TP391.41

CAI Guo-Lei, YANG Hong-Bo, ZOU Mou-Yan. An Active Contour-Based Unsupervised Texture Segmentation[J]. , 2005, 22(5): 624-630.

References

[1] Tsai A,Yezzi A,Willsky A.Curve evolution implementation of the Mumford-Shah funct ional for image segmentation,denoising,interpolat ion,andmagnif ication.IEEE Trans.Image Processing,2001,10(8) : 1169~ 1186

[2] Chan T,Vese L.Act ive contour w ithout edges.IEEE Trans.Image Processing,2001,10(2) : 266~ 277

[3] Sandberg B,Chan T,Vese L.A leve-l set and gabor-based act ive contour algorithm for segment ing t extured images.UCLA CAM Report 02-39,2002

[4] Rousson M,Brox T,Deri che R.Active unsupervised t exture segmentation on a diffusion based feature space.In: IEEE Computer Society Conf.onComputer Vision and Pattern Recognit ion.Madison,2003,2: 699~ 704

[5] Dunn D,Higgins W.Optimal gabor filters for t exture segmentat ion.IEEE Trans.Image Processing,1995,4(7) :964~ 974

[6] Unser M.Texture classif ication and segmentation using wavelet frames.IEEE Trans.Image Processing,1995,4(11) : 1549~ 1560

[7] Gabor D.Theory of communicat ion.Journal of the Institute of Electrical Engineers,1946,93: 429~ 457

[8] Daugman JG.Uncertainty relation for resolut ion in space,spatial frequency,and orientat ion optimized by two dimensional visual cortical f ilt er.J.Opt.Soc.Am.A,1985,2(7) : 1160~ 1169

[9] Bovik,Clark AC,Geisler WS.Multichannel t exture analysis using localized spat ial filters.IEEE Transactions on Pattern Analysis and Machine Intelligence,1990,12(1) : 55~ 73

[10] Mal ik J,Perona P.Preat tent ive texture discriminat ion with early vision mechanisms.J.Opt.Soc.Amer.A.1990,7(5) : 923~ 932

[11] Rudin L,Osher S.Total variat ion based image restoration with free local const raints.In: Proc IEEE ICIP.Texas Austin,1994.I: 31~ 35

[12] Gerig G,K?bler O,Kikinis R,et al.Nonlinear anisotropic f ilt ering of MRI data.IEEE Transactions on Medi cal Imaging,1992,11: 221~ 232

[13] J Weickert,BM t er Haar Romeny,MA Viergever.Efficient and reliable schemes for nonlinear diffusion f ilt ering.IEEE Transactions on ImageProcessing,1998,7(3) : 398~ 410

[14] Zhang LF.Act ive contour technique Based on minimum description length principle:

[Ph.D.thesis].Beijing: Inst itute of Electronics,ChineseAcademy of Sciences,2002 (in Chinese with English abstract)

[14] 张丽飞.基于最小描述长度准则的活动围道方法及相关技术

[博士学位论文].北京: 中国科学院电子学研究所,2002