Deriving Sparse Coefficients of Wavelet Pyramid Taking Clues from Hough Transform
Many applications like image compression requires sparse representation of image. To represent the image by sparse coefficients of transform is an old age technique. Still research is going on for better sparse representation of an image. A very recent technique is based on learning the basis for getting sparse coefficients. But learned basis are not guaranteed to span l 2 space, which is required for reconstruction. In this paper we are presenting a new technique to choose steerable basis of wavelet pyramid which gives sparse coefficients and better reconstruction. Here selection of steerable basis is based on clues from Hough transform.
KeywordsOrientation Direction Band Pass Sparse Representation Hough Transformation Sparse Coefficient
- 1.Tropp, J.A., Laska, J.N., Duarte, M.F., Romberg, J.K., Baraniuk, R.G.: Beyond nyquist: Efficient sampling of sparse bandlimited signals. CoRR (2009)Google Scholar
- 2.Li, Y., Cichocki, A., ichi Amari, S., Shishkin, S., Cao, J., Gu, F., Cao, J., Gu, F.: Sparse representation and its applications in blind source separation. In: Seventeenth Annual Conference on Neural Information Processing Systems, NIPS 2003 (2003)Google Scholar
- 4.Sallee, P., Olshausen, B.A., Lewicki, M.S.: Learning sparse image codes using a wavelet pyramid architecture 13, 887–893 (2001)Google Scholar
- 6.Karasaridis, A., Simoncelli, E.: A filter design technique for steerable pyramid image transforms, pp. 2387–2390 (1996)Google Scholar