Parsimonious representation of signals based on scattering transform - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2008

Parsimonious representation of signals based on scattering transform

Abstract

A parsimonious representation of signals is a mathematic model parametrized with a small number of parameters. Such models are useful for analysis, interpolation, filtering, feature extraction, and data compression. A new parsimonious model is presented in this paper based on scattering transforms. It is closely related to the eigenvalues and eigenfunctions of the linear Schrödinger equation. The efficiency of this method is illustrated in this paper with examples of both synthetic and real signals.

Dates and versions

hal-00854784 , version 1 (28-08-2013)

Identifiers

Cite

Michel Sorine, Qinghua Zhang, Taous-Meriem Laleg, Emmanuelle Crépeau. Parsimonious representation of signals based on scattering transform. 17th IFAC World Congress, Jul 2008, Seoul, South Korea. pp.12430-12435, ⟨10.3182/20080706-5-KR-1001.02104⟩. ⟨hal-00854784⟩
157 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More