Image Reconstruction Using NMF with Sparse Constraints Based on Kurtosis Measurement Criterion
A novel image reconstruction method using non-negative matrix factorization (NMF) with sparse constraints based on the kurtosis measurement is proposed by us. This NMF algorithm with sparse constraints exploited the Kurtosis as the maximizing sparse measure criterion of feature coefficients. The experimental results show that the natural images’ feature basis vectors can be successfully extracted by using our algorithm. Furthermore, compared with the standard NMF method, the simulation results show that our algorithm is indeed efficient and effective in performing image reconstruction task.
KeywordsNMF Sparse constraints Kurtosis Image reconstruction
Unable to display preview. Download preview PDF.
- 1.Julian, E., Edgar, K.: Sparse coding and NMF. In: proceedings of 2004 IEEE International Joint Conference on Neural Networks, vol. 4, pp. 2529–2533 (2004)Google Scholar
- 2.Hoyer, P.: Non-negative matrix factorization with sparseness constraints. Journal of Machine Learning Research 5, 1427–1469 (2004)Google Scholar
- 7.Hyvärinen, A., Oja, E., Hoyer, P., Horri, J.: Image Feature Extraction by Sparse Coding and Independent Component Analysis. In: Proc. Int. Conf. on Pattern Recognition (ICPR 1998), Brisbane, Australia, pp. 1268–1273 (1998)Google Scholar