Total Variation and Group Sparse Total Variation Denoising
Julia
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.

README.md

TotalVariation

An implementation of Total Variation Denoising and Group Sparse Total Variation Denoising.

Build Status

Total Variation (TV) minimization uses the TV norm to reduce excess variation in 1D signals. Using TV for denoising will result in a piecewise constant function with fewer pieces at higher levels of denoising.

Group sparse TV is an extension on the TV norm which models signals which have several localized transitions. Larger group sizes help model smoother signals with slow transitions.

For more information see src/example.jl and the source publication:

``Total Variation Denoising With Overlapping Group Sparsity'' by Ivan Selesnick and Po-Yu Chen (2013)