Morphology of the galaxy distribution from wavelet denoising
Preprint
Abstract
We have developed a method based on wavelets to obtain the true underlying smooth density from a point distribution. The goal has been to reconstruct the density field in an optimal way ensuring that the morphology of the reconstructed field reflects the true underlying morphology of the point field which, as the galaxy distribution, has a genuinely multiscale structure, with near-singular behavior on sheets, filaments and hotspots. If the discrete distributions are smoothed using Gaussian filters, the morphological properties tend to be closer to those expected for a Gaussian field. The use of wavelet denoising provide us with a unique and more accurate morphological description.Keywords
All Related Versions
- Version 1, 2005-08-15, ArXiv
- Published version: The Astrophysical Journal, 634 (2), 744.
This publication has 3 references indexed in Scilit:
- Image Processing and Data Analysis, The Multiscale ApproachJournal of the American Statistical Association, 1999
- Density Estimation for Statistics and Data AnalysisTechnometrics, 1987
- The Geometry of Random Fields.Journal of the American Statistical Association, 1982