AutocorrelationShell is package for using autocorrelation wavelets in julia.

Signal representations using autocorrelation wavelets are redundant and non-orthogonal. Some desirable properties of autocorrelation wavelet transforms are symmetry without losing vanishing moments, edge detection and characterization capabilities, and shift invariance. Autocorrelation wavelets can be used as a tool for data analysis such as time series analysis and image analysis.

The Autocorrelation Shell is heavily inspired by the paper Wavelets, their autocorrelation functions, and multiresolution representation of signals.


You can install the package at the Pkg REPL-mode with:

(v1.4) > add AutocorrelationShell


  • 1D autocorrelation wavelet transform.
  • 2D autocorrelation wavelet transform.
  • 1D autocorrelation wavelet packet transform.
  • A variety of utility functions for image de-noising such as functions for Signal to Noise Ratio(SNR) and Structural Similarity Index(SSIM).
  • The package is completely documented.


This package was authored by Rishi Subramanian, Christina Chang, and Shozen Dan under the supervision of Professor Naoki Saito at University of California, Davis.