Special Issue on Computational Intelligence in Signal and Image Processing; Guest Editors: Yen-Wei Chen and Vasile PaladeNo Access

TEXTURE SEGMENTATION USING SEMI-SUPERVISED SUPPORT VECTOR MACHINES

SAEID SANEI

SAEID SANEI

Centre for Digital Signal Processing Research, Department of Electronic Engineering, King's College London, London, WC2R 2LS, United Kingdom

Search for more papers by this author

https://doi.org/10.1142/S1469026804001197Cited by:2 (Source: Crossref)

Abstract

Segmentation of natural textures has been investigated by developing a novel semi-supervised support vector machines (S³VM) algorithm with multiple constraints. Unlike conventional segmentation algorithms the proposed method does not classify the textures but classifies the uniform-texture regions and the regions of boundaries. Also the overall algorithm does not use any training set as used by all other learning algorithms such as conventional SVMs. During the process, the images are restored from high spatial frequency noise. Then various-order statistics of the textures within a sliding two-dimensional window are measured. K-mean algorithm is used to initialise the clustering procedure by labelling part of the class members and the classifier parameters. Therefore at this stage we have both the training and the working sets. A non-linear S³VM is then developed to exploit both sets to classify all the regions. The convex algorithm maximises a defined cost function by incorporating a number of constraints. The algorithm has been applied to combinations of a number of natural textures. It is demonstrated that the algorithm is robust, with negligible misclassification error. However, for complex textures there may be a minor misplacement of the edges.

Keywords:

Remember to check out the Most Cited Articles!
Check out these titles in artificial intelligence!

References

M. Tuceryan and A. K. Jain, Handbook Pattern Recognition and Computer Vision, eds. C. H. Chen, L. F. Pau and P. S. P. Wang (World Scientific, Singapore, 1993) pp. 235–276. Link, Google Scholar
R. Haralick, K. Shanmugan and L. Dinstein, IEEE Trans. Syst. Man Cybernet. 3, 610 (1973). Crossref, Google Scholar
T. Randen and J. H. Husoy, IEEE Trans. Pattern Anal. Much. Intell. 21(4), 291 (1999), DOI: 10.1109/34.761261. Crossref, Google Scholar
S. C. Zhu, Y. Wu and D. Mumford, Int. J. Comput. Vision 27(2), (1998), DOI: 10.1023/A:1007925832420. Google Scholar
A. K. Jain and F. Farrokhnia, IEEE Trans. Pattern Recognition 24(12), 1167 (1991), DOI: 10.1016/0031-3203(91)90143-S. Crossref, Google Scholar
M. R. Turner, Biol. Cybernet. 55, 71 (1986). Crossref, Google Scholar
C. S. Lu, P. C. Chung and C. F. Chen, Pattern Recognition 30(5), 729 (1997), DOI: 10.1016/S0031-3203(96)00116-1. Crossref, Google Scholar
A. K. Jain and K. Karu, IEEE Trans. Pattern Anal. Much. Intell. 18(2), 195 (1996), DOI: 10.1109/34.481543. Crossref, Google Scholar
T. Kaspariset al., Pattern Recognition 34, 1963 (2001), DOI: 10.1016/S0031-3203(00)00126-6. Crossref, Google Scholar
R. L. Kashyap, IEEE Trans. Inf. Theory 30, 736 (1984), DOI: 10.1109/TIT.1984.1056955. Crossref, Google Scholar
K. I. Kimet al., IEEE Trans. Pattern Anal. Much. Intell. 24(11), 1542 (2002). Google Scholar
V. Vapnik , The Nature of Statistical Learning Theory ( Springer-Verlag , 1995 ) . Crossref, Google Scholar
B. Scholkopfet al., IEEE Trans. Signal Process. 45(11), 2758 (1997), DOI: 10.1109/78.650102. Crossref, Google Scholar
S. Haykin , Neural Network — A Comprehensive Fundation , 2nd edn. ( Prentice Hall , 1999 ) . Google Scholar
D. DeCoste and B. Scholkopf , Training Invariant Support Vector Machines, Machine Learning ( Kluwer Press , 2001 ) . Google Scholar
C. J. C. Burges, Data Mining and Knowledge Discovery 2(2), 1 (1998), DOI: 10.1023/A:1009715923555. Google Scholar
R. Fletcher , Practical Methods of Optimisation , 2nd edn. ( John Wiley , 1987 ) . Google Scholar
E. Mayoraz and E. Alpaydin, Support vector machines for multi-class classifica tion, Technical Report IDIAP-PR 98–06, Dalle Molle Last. For Perceptual Artificial Intelligence, 1998 . Google Scholar
A. M. Bensaidet al., Pattern Recognition 29(5), 859 (1999), DOI: 10.1016/0031-3203(95)00120-4. Crossref, Google Scholar
M. Vaidyanathan et al. , Tumor volume measurements using supervised and semi-supervised MRI segmentation , Artif. Neural Netw. Eng. Conf ( ANNIE , 1994 ) . Google Scholar
K. P. Benneth and A. Demirez , Semi-supervised support vector machines , Proc. Neural Inf. Process. ( 1998 ) . Google Scholar
J. H. Friedman, F. Baskett and L. J. Shustek, IEEE Trans. Comput. C 24, 1000 (1975). Google Scholar
M. K. Tsatsanis and G. B. Giannakis, A non-parametric approach for detecing changes in the autocorrelation structure, Proc. 6th Eur. Signal Process. Conf. (EUSIPCO-92) (1992) pp. 843–846. Google Scholar
B. M. Sadler, Texture segmentation by change detection in second and higher order statistics, Conf. Record 27th Asilomar Conf. Signals Syst. Corn-put.1 (1993) pp. 436–440. Google Scholar
P. Brodatz , Texture: A Photographic Album for Artists and Designers ( Dover , New York , 1966 ) . Google Scholar