Robust regression of scattered data with adaptive spline-wavelets

A coarse-to-fine data fitting algorithm for irregularly spaced data based on boundary-adapted adaptive tensor-product semi-orthogonal spline-wavelets has been proposed in Castano and Kunoth, 2003. This method has been extended in Castano and Kunoth, 2005 to include regularization in terms of Sobolev...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on image processing 2006-06, Vol.15 (6), p.1621-1632
Main Authors: Castano, D., Kunoth, A.
Format: Article
Language:English
Subjects:
Adaptive algorithms
Adaptive wavelets
Algorithms
Applied sciences
Artificial Intelligence
Boundaries
Clouds
coarse-to-fine algorithm
Computational efficiency
Computer Simulation
Estimators
Exact sciences and technology
Government
Image Enhancement - methods
Image Interpretation, Computer-Assisted - methods
Image processing
Imaging, Three-Dimensional - methods
Information Storage and Retrieval - methods
Information, signal and communications theory
Laboratories
Least squares approximation
Least squares method
Mathematical analysis
Mathematical models
Models, Statistical
Norms
outlier detection
Regression
Regression Analysis
Regularization
robust estimation
Robustness
Sample Size
scattered data
Scattering
Signal processing
Signal Processing, Computer-Assisted
Spline
Telecommunications and information theory
Tensile stress
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A coarse-to-fine data fitting algorithm for irregularly spaced data based on boundary-adapted adaptive tensor-product semi-orthogonal spline-wavelets has been proposed in Castano and Kunoth, 2003. This method has been extended in Castano and Kunoth, 2005 to include regularization in terms of Sobolev and Besov norms. In this paper, we develop within this least-squares approach some statistical robust estimators to handle outliers in the data. Our wavelet scheme yields a numerically fast and reliable way to detect outliers.
ISSN:1057-7149
1941-0042
DOI:10.1109/TIP.2006.871164