Invertible orientation bundles on 2D scalar images
A general approach for multiscale orientation analysis of 2D scalar images is proposed. A scale-dependent orientation bundle (map of the visual space into function of two arguments: position and orientation) is constructed from the local Gaussian-derivatives jet of a scalar image in 2D. It is shown that there exists a class of orientation filters exhibiting an invertible relation between the orientation bundle and the original image in space domain. This invertible transformation is used to regain the original acuity in the spatial domain after analyzing orientation features at any given scale. The approach turns out to be highly effective for the detection of elongated structures.
KeywordsIntegral Transformations Orientation Analysis Scale Space
Unable to display preview. Download preview PDF.
- 1.William T. Freeman and Edward H. Adelson. The design and use of steerable filters. IEEE Trans. Pattern Analysis and Machine Intelligence, 13(9):891–906, September 1991.Google Scholar
- 2.J. J. Koenderink. The structure of images. Biol. Cybern., 50:363–370, 1984.Google Scholar
- 3.Tai Sing Lee. Image representation using 2D Gabor wavelets. IEEE Trans. Pattern Analysis and Machine Intelligence, 18(10):959–971, October 1996.Google Scholar
- 4.T. Lindeberg. Scale-Space Theory in Computer Vision. The Kluwer International Series in Engineering and Computer Science. Kluwer Academic Publishers, Dordrecht, the Netherlands, 1994.Google Scholar
- 5.M. Michalis. Low level image processing using steerable filters. Institute for Medical Informatics, 1995. Munich, PhD thesis.Google Scholar
- 6.P. Perona. Deformable kernels for early vision. In IEEE CVPR, pages 222–227, June 1994.Google Scholar
- 7.E. P. Simoncelli. A rotation invariant pattern signature. In Proc. of the 3rd IEEE Int. Conf. on Image Processing. IEEE, 1996. to appear.Google Scholar
- 8.E. P. Simoncelli and H. Farid. Steerable wedge filters for local orientation analysis. IEEE Transactions on Image Processing, July 1996. to appear in.Google Scholar
- 9.Eero Simoncelli. A rotation-invariant pattern signature. In International Conference on Image Pprocessing, pages 185–188, 1996.Google Scholar