A Fast Dictionary Learning Algorithm for Image Denoising

Abstract

Abstract. The K-SVD is one of the well-known and effective methods to learn a universal and overcomplete dictionary. However, K-SVD is very expensive because many iteration steps are needed. What's more, when it converts 2D data patches into 1D vectors for training or learning, K-SVD breaks down the inherent geometric structure of the data. To overcome these limitations, employing a subspace partition technique, we propose an efficient and fast algorithm, the fast top-bottom two-dimensional subspace partition algorithm, for learning overcomplete dictionaries. Experimental simulations demonstrate that our dictionary learning approach is effective for image denoising.

Adaptive Multiscale Methods for Sparse Image Representation and Dictionary Learning

“…In Liu et al (2018) (see also Li et al (2018)) another procedure for denoising of twodimensional patches extracted from images is proposed which also follows the iterative scheme of Algorithm (9). However here the coding operator is again linear: the dictionary learning procedure works with the two-dimensional patches, but once this is learned the patches and dictionary atoms are vectorized into matrices Y = [vec(Y 1 )| .…”