ECCV 2004: Computer Vision - ECCV 2004 pp 14-24

# Support Blob Machines

The Sparsification of Linear Scale Space
• Marco Loog
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3024)

## Abstract

A novel generalization of linear scale space is presented. The generalization allows for a sparse approximation of the function at a certain scale.

To start with, we first consider the Tikhonov regularization viewpoint on scale space theory [15]. The sparsification is then obtained using ideas from support vector machines [22] and based on the link between sparse approximation and support vector regression as described in [4] and [19].

In regularization theory, an ill-posed problem is solved by searching for a solution having a certain differentiability while in some precise sense the final solution is close to the initial signal. To obtain scale space, a quadratic loss function is used to measure the closeness of the initial function to its scale σ image.

We propose to alter this loss function thus obtaining our generalization of linear scale space. Comparable to the linear ε-insensitive loss function introduced in support vector regression [22], we use a quadratic ε-insensitive loss function instead of the original quadratic measure. The ε-insensitivity loss allows errors in the approximating function without actual increase in loss. It penalizes errors only when they become larger than the a priory specified constant ε. The quadratic form is mainly maintained for consistency with linear scale space.

Although the main concern of the article is the theoretical connection between the foregoing theories, the proposed approach is tested and exemplified in a small experiment on a single image.

## Keywords

Support Vector Machine Support Vector Loss Function Quadratic Programming Support Vector Regression
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

## References

1. 1.
van den Boomgaard, R., Smeulders, A.W.M.: The morphological structure of images, the differential equations of morphological scale-space. IEEE Transactions on Pattern Analysis and Machine Intelligence 16(11), 1101–1113 (1994)
2. 2.
Duits, R., Felsberg, M., Florack, L.M.J., Platel, B.: α Scale spaces on a bounded domain. In: Griffin, L.D., Lillholm, M. (eds.) Scale-Space 2003. LNCS, vol. 2695, pp. 494–510. Springer, Heidelberg (2003)
3. 3.
Florack, L.M.J., Maas, R., Niessen, W.J.: Pseudo-linear scale-space theory. International Journal of Computer Vision 31(2/3), 247–259 (1999)
4. 4.
Girosi, F.: An equivalence between sparse approximation and support vector machines. Neural Computation 10(6), 1455–1480 (1998)
5. 5.
Girosi, F., Jones, M., Poggio, T.: Prior stabilizers and basis functions: From regularization to radial, tensor and additive splines. Technical Report AI Memo 1430, CBCL Paper 75. MIT, Cambridge, MA (1993)Google Scholar
6. 6.
Griffin, L.D.: Local image structure, metamerism, norms, and natural image statistics. Perception 31(3) (2002)Google Scholar
7. 7.
ter Haar Romeny, B.M. (ed.): Geometry-Driven Diffusion. Kluwer, Dordrecht (1996)Google Scholar
8. 8.
Huber, P.J.: Robust Statistics. John Wiley & Sons, New York (1981)
9. 9.
Jackway, P.R.: Morphological scale-space. In: 11th IAPR International Conference on Pattern Recognition, pp. 252–255. The Hague, The Netherlands (1992)Google Scholar
10. 10.
Koenderink, J.J.: The structure of images. Biological Cybernetics 50, 363–370 (1984)
11. 11.
Koenderink, J.J., van Doorn, A.J.: Metamerism in complete sets of image operators. In: Advances in Image Understanding 1996, pp. 113–129 (1996)Google Scholar
12. 12.
Lillholm, M., Nielsen, M., Griffin, L.D.: Feature-based image analysis. International Journal of Computer Vision 52, 73–95 (2003)
13. 13.
Lindeberg, T.: Scale-Space Theory in Computer Vision. Kluwer Academic Press, Boston (1994)Google Scholar
14. 14.
Loog, M., Lillholm, M., Nielsen, M., Viergever, M.A.: Gaussian scale space from insufficient image information. In: Griffin, L.D., Lillholm, M. (eds.) Scale-Space 2003. LNCS, vol. 2695, pp. 757–769. Springer, Heidelberg (2003)
15. 15.
Nielsen, M., Florack, L.M.J., Deriche, R.: Regularization, scale-space, and edge detection filters. Journal of Mathematical Imaging and Vision 7(4), 291–307 (1997)
16. 16.
Park, K.-R., Lee, C.-N.: Scale-space using mathematical morphology. IEEE Transactions on Pattern Analysis and Machine Intelligence 18(11), 1121–1126 (1996)
17. 17.
Perona, P., Malik, J.: Scale space and edge detection using anisotropic diffusion. IEEE Transactions on Pattern Analysis and Machine Intelligence 12, 429–439 (1990)
18. 18.
Scherzer, O., Weickert, J.: Relations between regularization and diffusion filtering. Journal of Mathematical Imaging and Vision 12, 43–63 (2000)
19. 19.
Smola, J., Schölkopf, B.: A tutorial on support vector regression. Technical Report NC-TR-98-030, Royal Holloway College, University of London, UK (1998)Google Scholar
20. 20.
Smola, J., Schölkopf, B., Müller, K.-R.: The connection between regularization operators and support vector kernels. Neural Networks 11, 637–650 (1998)
21. 21.
Tikhonov, N., Arseninn, V.Y.: Solution of Ill-Posed Problems. W. H. Winston, Washington (1977)Google Scholar
22. 22.
Vapnik, V.: The Nature of Statistical Learning Theory. Springer, New York (1995)
23. 23.
Weickert, J.: Anisotropic Diffusion in Image Processing. B. G. Teubner, Stuttgart (1998)
24. 24.
Witkin, P.: Scale-space filtering. In: Proceedings of IJCAI, Germany (1983)Google Scholar