Modified 9DLT Matrix for Similarity Retrieval of Line-Drawing Images
An attempt towards perception of spatial relationships existing among the generic components of line-drawing images is made for their similarity retrieval. The proposed work is based on modified 9DLT matrix representation. The conventional concept of 9DLT matrix has been tuned to accommodate multiple occurrences of identical components in images. A novel similarity measure to estimate the degree of similarity between two 9DLT matrices representing images is introduced and exploited for retrieval of line-drawing images. A database of 118 line-drawing images has been created to corroborate the effectiveness of the proposed model for similarity retrieval through an extensive experimentation.
- 2.Chang, C.C.: Spatial Match Retrieval of Symbolic Pictures. J. Information Science and Engineering 7 17(3), 405–422 (1991)Google Scholar
- 4.Chang, S.K., Yan, C.W., Dimitroff, D.C., Arndt, T.: An Intelligent Image Database System. IEEE Tran. Software Engineering 14(3) (1988)Google Scholar
- 6.Franti, P., Mednonogov, A., Kalviainen, H.: Hough transform for rotation invariant matching of line-drawing images. In: Proceedings of the International Conf. on Pattern Recognition (ICPR), pp. 389–392 (2000)Google Scholar
- 10.Kerbyson, D.J., Atherton, T.J.: Circle detection using Hough transform filters. In: Proceedings of Fifth International Conf. on Image Processing and its Applications, pp. 370–374 (1995)Google Scholar
- 14.Punitha, P.: IARS: Image Archival and Retrieval Systems, Ph. D. Thesis, Dept. of Studies in Computer Science, University of Mysore, India (2006)Google Scholar