Invariant Feature Extraction and Object Shape Matching Using Gabor Filtering
Gabor filter-based feature extraction and its use in object shape matching are addressed. For the feature extraction multi-scale Gabor filters are used. From the analysis of the properties of the Gabor-filtered image, we know isolated dominant points generally exist on the object contour, when the filter design parameters are properly selected. The dominant points thus extracted are robust to the image noise, scaling, rotation, translation, and the minor projection deformation. Object shape matching in terms of a two-stage point matching is presented. First, a feature vector representation of the dominant point is used for initial matching. Secondly, the compatibility constraints on the distances and angles between point pairs are used for the final matching. Computer simulations with synthetic and real object images are included to show the feasibility of the proposed method.
KeywordsGabor Filter Gabor Wavelet Compatibility Constraint Gabor Function Object Contour
Unable to display preview. Download preview PDF.
- 2.A.K. Jain, S. Prabhakar, and L. Hong. A multichannel approach to fingerprint classification. IEEE Trans. PAMI, vol. 21, no.4, pp.348–359, 1999.Google Scholar
- 3.J. Chen, Y. Sato, and S. Tamura. Orientation space filtering for multiple orientation line segmentation. IEEE Trans, PAMI, vol. 22, no. 5, pp. 417–429, 2000.Google Scholar
- 4.Z. Wang and M. Jenkin. Using complex Gabor filters to detect and localize edges and bars. In: C. Archibald and E. Petriu, (eds.): Advanced in Machine Vision: Strategies and Applications, vol. 32, River Edge, NJ: World Scientific (1992) pp. 151–170.Google Scholar
- 8.B.S. Manjunath and W.Y. Ma. Texture features for browsing and retrieval of image data. IEEE Trans. PAMI, vol. 18, no. 8, pp. 837–842, 1996.Google Scholar
- 9.A.P. Witkin. Scale-space filtering. In proc 8th Int. Joint Conf. Artificial Intell., pp. 1019–1021, 1983.Google Scholar
- 11.J.G. Daugman. Uncertainty relation for resolution in space, spatial frequency, and orientation optimized by two-dimensional visual cortical filters. J. Optical Soc. Amer., vol. 2, no.7, pp. 1169–1179, 1988.Google Scholar
- 13.T. S. Lee. Image representation using Gabor wavelets. IEEE Trans. PAMI, vol 18, no. 10, pp. 959–970 1996.Google Scholar