Electronic Journal of Statistics
- Electron. J. Statist.
- Volume 6 (2012), 231-268.
Statistical multiresolution Dantzig estimation in imaging: Fundamental concepts and algorithmic framework
In this paper we are concerned with fully automatic and locally adaptive estimation of functions in a “signal + noise”-model where the regression function may additionally be blurred by a linear operator, e.g. by a convolution. To this end, we introduce a general class of statistical multiresolution estimators and develop an algorithmic framework for computing those. By this we mean estimators that are defined as solutions of convex optimization problems with ℓ∞-type constraints. We employ a combination of the alternating direction method of multipliers with Dykstra’s algorithm for computing orthogonal projections onto intersections of convex sets and prove numerical convergence. The capability of the proposed method is illustrated by various examples from imaging and signal detection.
Electron. J. Statist., Volume 6 (2012), 231-268.
First available in Project Euclid: 29 February 2012
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
alternating direction method of multipliers (ADMM) biophotonics Dantzig selector Dykstra’s projection algorithm local adaption signal detection statistical imaging statistical multiscale analysis statistical regularization
Frick, Klaus; Marnitz, Philipp; Munk, Axel. Statistical multiresolution Dantzig estimation in imaging: Fundamental concepts and algorithmic framework. Electron. J. Statist. 6 (2012), 231--268. doi:10.1214/12-EJS671. https://projecteuclid.org/euclid.ejs/1330524559