L0-Norm-Based Sparse Representation Through Alternate Projections

作者: Luis Mancera , Javier Portilla

DOI: 10.1109/ICIP.2006.312819

关键词:

摘要: We present a simple and robust method for finding sparse representations in overcomplete transforms, based on minimization of the L0-norm. Our is better than current solutions L1-norm terms energy compaction. These results strongly question equivalence minimizing both norms real conditions. also show application to in-painting (interpolation lost pixels).

参考文章(21)
R. R. Coifman, D. L. Donoho, Translation-Invariant De-Noising Wavelets and Statistics. pp. 125- 150 ,(1995) , 10.1007/978-1-4612-2544-7_9
Brian A. Wandell, Foundations of vision ,(1995)
Filip Rooms, Wilfried R. Philips, Javier Portilla, Parametric PSF estimation via sparseness maximization in the wavelet domain Proc.of SPIE - Wavelet applications in industrial processing II. ,vol. 5607, pp. 26- 33 ,(2004) , 10.1117/12.571539
D. Youla, Generalized Image Restoration by the Method of Alternating Orthogonal Projections IEEE Transactions on Circuits and Systems. ,vol. 25, pp. 694- 702 ,(1978) , 10.1109/TCS.1978.1084541
Emmanuel Candes, Justin Romberg, Stable signal recovery from incomplete observations Proceedings of SPIE. ,vol. 5914, pp. 239- 244 ,(2005) , 10.1117/12.620143
J.-L. Starck, Y. Moudden, J. Bobin, M. Elad, D. L. Donoho, Morphological component analysis Proceedings of SPIE. ,vol. 5914, pp. 209- ,(2005) , 10.1117/12.615237
David L. Donoho, For most large underdetermined systems of linear equations the minimal 1-norm solution is also the sparsest solution Communications on Pure and Applied Mathematics. ,vol. 59, pp. 797- 829 ,(2006) , 10.1002/CPA.20132
B OLSHAUSEN, D FIELD, Sparse coding of sensory inputs. Current Opinion in Neurobiology. ,vol. 14, pp. 481- 487 ,(2004) , 10.1016/J.CONB.2004.07.007
Scott Shaobing Chen, David L. Donoho, Michael A. Saunders, Atomic Decomposition by Basis Pursuit SIAM Review. ,vol. 43, pp. 129- 159 ,(2001) , 10.1137/S003614450037906X