Rotationally Invariant Time-Frequency Scattering Transforms

被引:6
|
作者
Czaja, Wojciech [1 ]
Li, Weilin [2 ]
机构
[1] Univ Maryland, Dept Math, College Pk, MD 20742 USA
[2] NYU, Courant Inst Math Sci, New York, NY 10003 USA
基金
美国国家科学基金会;
关键词
Scattering transform; Uniform covering frames; Time-frequency; Directional representations; Neural networks; Feature extraction; TEXTURE CLASSIFICATION; FRAMES;
D O I
10.1007/s00041-019-09705-w
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
The focus of this paper is on the mathematical construction of a transform that is invariant to a finite rotation group and is stable to small perturbations. A key step in our theory lies in the construction of directionally sensitive functions that are partially generated by rotations. We call such a family a rotational uniform covering frame and by studying rotations of the frame, we derive the desired operator, which we call the rotational Fourier scattering transform. We prove that the transformation is rotationally invariant to a finite rotation group, is bounded above and below, is non-expansive, and contracts small translations and additive diffeomorphisms. To address the numerical aspects of this theory, we also construct digital versions of the frame and show how to faithfully truncate the transform. We also discuss connections between this new family of directional representations with previously constructed ones.
引用
收藏
页数:23
相关论文
共 50 条
  • [21] JOINT TIME-FREQUENCY SCATTERING FOR AUDIO CLASSIFICATION
    Anden, Joakim
    Lostanlen, Vincent
    Mallat, Stephane
    [J]. 2015 IEEE INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING, 2015,
  • [22] A time-frequency analysis method for radar scattering
    Kritikos, HN
    Teti, JG
    [J]. IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, 1998, 46 (03) : 257 - 260
  • [23] MODEL OF ACOUSTIC-SIGNALS BY TIME-SCALE AND TIME-FREQUENCY TRANSFORMS
    COURBEBAISSE, G
    ESCUDIE, B
    GUILLEMAIN, P
    KRONLANDMARTINET, R
    [J]. JOURNAL DE PHYSIQUE IV, 1994, 4 (C5): : 1315 - 1318
  • [24] Applications of high-resolution time-frequency transforms to attenuation estimation
    Tary, Jean Baptiste
    van der Baan, Mirko
    Herrera, Roberto Henry
    [J]. GEOPHYSICS, 2017, 82 (01) : V7 - V20
  • [25] The estimation of spectra from time-frequency transforms for use in attenuation studies
    Beckwith, James
    Clark, Roger
    Hodgson, Linda
    [J]. GEOPHYSICAL PROSPECTING, 2017, 65 (01) : 204 - 220
  • [26] TIME-FREQUENCY ANALYSIS ASSOCIATED WITH K-HANKEL-WIGNER TRANSFORMS
    Boubatra, Mohamed Amine
    [J]. COMMUNICATIONS OF THE KOREAN MATHEMATICAL SOCIETY, 2022, 37 (02): : 521 - 535
  • [27] A correspondence between zeros of time-frequency transforms and Gaussian analytic functions
    Bardenet, Remi
    Chainais, Pierre
    Flamant, Julien
    Hardy, Adrien
    [J]. 2019 13TH INTERNATIONAL CONFERENCE ON SAMPLING THEORY AND APPLICATIONS (SAMPTA), 2019,
  • [28] Radon transforms and time-frequency representations for ISAR motion compensation and imaging
    Steeghs, P
    Kester, L
    Gelsema, S
    [J]. WAVELET AND INDEPENDENT COMPONENET ANALYSIS APPLICATIONS IX, 2002, 4738 : 252 - 263
  • [29] Aspects of Change Detection in Vibrational Processes Based on Time-Frequency Transforms
    Aiordachioaie, Dorel
    [J]. 2017 5TH INTERNATIONAL SYMPOSIUM ON ELECTRICAL AND ELECTRONICS ENGINEERING (ISEEE), 2017,
  • [30] Time-Frequency and Wavelet Transforms of EMI Dynamic Spectrum in Chaotic Converter
    Yang, Ru
    Zhang, Bo
    Qiu, Dongyuan
    Liu, Zuolian
    [J]. IEEE TRANSACTIONS ON POWER ELECTRONICS, 2009, 24 (3-4) : 1083 - 1092