Abstract
We present an implementation of the Normalized Cuts method for the solution of the image segmentation problem on polygonal grids. We show that in the presence of rounding errors the eigenvector corresponding to the k-th smallest eigenvalue of the generalized graph Laplacian is likely to contain more than k nodal domains. It follows that the Fiedler vector alone is not always suitable for graph partitioning, while the eigenvector subspace, corresponding to just a few of the lowest eigenvalues, contains sufficient information needed for obtaining meaningful segmentation. At the same time, the eigenvector corresponding to the trivial solution often carries nontrivial information about the nodal domains in the image and can be used as an initial guess for the Krylov subspace eigensolver. We show that proposed algorithm performs favorably when compared to the Multiscale Normalized Cuts and Segmentation by Weighted Aggregation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Fiedler, M.: Algebraic Connectivity of Graphs. Czechoslovak Math. J. 23(98), 298–305 (1973)
Biyikoǧlu, T., Leydold, J., Stadler, P.F.: Laplacian Eigenvectors of Graphs. Perron-Frobenius and Fraber-Krahn Type Theorems, vol. 1915. Springer, Heidelberg (2007)
Dhillon, I.S., Guan, Y., Kulis, B.: Weighted Graph Cuts without Eigenvectors: a Multilevel Approach. IEEE Transactions on Pattern Analysis and Machine Intelligence 29(11), 1944–1957 (2007)
Yu, S.X., Shi, J.: Multiclass Spectral Clustering. In: International Conference on Computer Vision, Nice, France, pp. 11–17 (2003)
Yu, S.X.: Segmentation using multiscale cues. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 70–77 (2004)
Cour, T., Benezit, F., Shi, J.: Spectral Segmentation with Multiscale Graph Decomposition. In: IEEE International Conference on Computer Vision and Pattern Recognition, CVPR (2005), http://www.seas.upenn.edu/~timothee/software/ncut_multiscale/ncut_multiscale.html
Sharon, E., Galun, M., Sharon, D., Basri, R., Brandt, A.: Hierarchy and Adaptivity in Segmenting Visual Scenes. Nature 442, 810–813 (2006), http://www.wisdom.weizmann.ac.il/~swa/index.html
Malik, J., Belongie, S., Leung, T., Shi, J.B.: Contour and Texture Analysis for Image Segmentation. International Journal of Computer Vision 43(1), 7–27 (2001)
Cullum, J.K., Willoughby, R.A.: Lanczos algorithms for large symmetric eigenvalue computations, vol. 1. Society for Industrial and Applied Mathematics, Philadelphia (2002)
Matsekh, A.M., Shurina, E.P.: On Computing Spectral Decomposition of Symmetric Matrices and Singular Value Decomposition of Unsymmetric Matrices with Guaranteed Accuracy. Optoelectronics, Instrumentation and Data Processing 43(2), 159–169 (2007)
Prasad, L., Skourikhine, A.N.: Vectorized Image Segmentation via Trixel Agglomeration. Pattern Recognition 39, 501–514 (2006)
Martin, D., Fowlkes, C., Tal, D., Malik, J.: A Database of Human Segmented Natural Images and its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics. In: Proc. 8th Int’l Conf. Computer Vision, vol. 2, pp. 416–423 (2001)
Shi, J.B., Malik, J.: Normalized Cuts and Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 22(8), 888–905 (2000)
Godunov, S.K., Antonov, A.G., Kiriljuk, O.P., Kostin, V.I.: Guaranteed accuracy in numerical linear algebra. Kluwer Academic Publishers Group, Dordrecht (1993); ISBN 0-7923-2352-1. Translated and revised from the 1988 Russian original
Matsekh, A.M.: The Godunov-Inverse Iteration: a Fast and Accurate Solution to the Symmetric Tridiagonal Eigenvalue Problem. Applied Numerical Mathematics 54(2), 208–221 (2005)
Demmel, J.: Lecture Notes S267. U.C. Berkeley (1996), http://www.cs.berkeley.edu/~demmel/cs267
Comaniciu, D., Meer, P.: Mean Shift: A Robust Approach toward Feature Space Analysis. IEEE Trans. Pattern Analysis Machine Intell., 603–619 (2002)
Sturm, U., Weidner, U.: Further investigations on segmentation quality assessment for remote sensing applications. In: ISPRS Hannover Workshop 2009 High-Resolution Earth Imaging for Geospatial Information, Hannover, Germany (2009)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Matsekh, A., Skurikhin, A., Prasad, L., Rosten, E. (2012). Numerical Aspects of Spectral Segmentation on Polygonal Grids. In: Jónasson, K. (eds) Applied Parallel and Scientific Computing. PARA 2010. Lecture Notes in Computer Science, vol 7133. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-28151-8_19
Download citation
DOI: https://doi.org/10.1007/978-3-642-28151-8_19
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-28150-1
Online ISBN: 978-3-642-28151-8
eBook Packages: Computer ScienceComputer Science (R0)