We present a functional framework for the adaptive design of dictionaries where the invariance to translation, dilation, and rotation is built upfront into the primary representation space. Our key idea is to build an invariant signal representation prior to the learning stage. By doing so, we focus our effort on adapting the dictionary to the distinctive features of the signal, rather than to ...

Density conditions for wavelet systems with arbitrary sampling points to be frames are studied. We show that for a wavelet system generated by admissible functions and irregular affine lattices to be a frame, the sampling points must have a positive lower Beurling density. The same is true for wavelet systems with arbitrary sampling points and nice generating functions.

An irregular wavelet frame has the form W(ψ,Λ) = {aψ( a − b)}(a,b)∈Λ, where ψ ∈ L (R) and Λ is an arbitrary sequence of points in the affine group A = R ×R. Such irregular wavelet frames are poorly understood, yet they arise naturally, e.g., from sampling theory or the inevitability of perturbations. This paper proves that irregular wavelet frames satisfy a Homogeneous Approximation Property, w...

Finite affine groups are given by groups of translations and di- lations on finite cyclic groups. For cyclic groups of prime order we develop a time-scale (wavelet) analysis and show that for a large class of non-zero window signals/vectors, the generated full cyclic wavelet system constitutes a frame whose canonical dual is a cyclic wavelet frame.

In this technical report we build Discrete Wavelet Frames on the sphere S, discretizing the existing Spherical Continuous Wavelet Transform (CWT). We first explore the spherical half-continuous frames, i.e. where the position remains a continuous variable; and then we proceed to the fully discrete frames. We introduce the notion of controlled frames, which reflects the particular nature of the ...