Linear Multiscale Transforms Based on Even-Reversible Subdivision Operators

Multiscale transforms for real-valued data, based on interpolatory subdivision operators, have been studied in recent years. They are easy to define and can be extended other types of example, manifold-valued data. In this chapter, we linear multiscale transforms, certain linear, non-interpolatory termed “even-reversible.” For such prove, using Wiener’s lemma, the existence an inverse operator defined by even part mask term it “even-inverse.” We show that with spline or pseudo-spline masks, even-reversible derive explicitly, quadratic cubic symbols corresponding even-inverse operators. also analyze properties particular, their stability rate decay details.