The DOI matches its Crossref record; 32 of the 32 references it cites were found in a registry or an index.
Minimizers of Cost-Functions Involving Nonsmooth Data-Fidelity Terms. Application to the Processing of Outliers
Abstract
We present a theoretical study of the recovery of an unknown vector $x\in{\mathbb R}^p$ (such as a signal or an image) from noisy data $y\in{\mathbb R}^q$ by minimizing with respect to x a regularized cost-function ${\cal F}(x,y)=\Psi(x,y)+\alpha\Phi(x)$, where $\Psi$ is a data-fidelity term, $\Phi$ is a smooth regularization term, and $\alpha > 0$ is a parameter. Typically, $\Psi(x,y)=\|Ax-y \|^2$, where A is a linear operator. The data-fidelity terms $\Psi$ involved in regularized cost-functions are generally smooth functions; only a few papers make an exception to this and they consider restricted situations. Nonsmooth data-fidelity terms are avoided in image processing. In spite of this, we consider both smooth and nonsmooth data-fidelity terms. Our goal is to capture essential features exhibited by the local minimizers of regularized cost-functions in relation to the smoothness of the data-fidelity term. In order to fix the context of our study, we consider $\Psi(x,y)=\sum_i\psi(a_i^Tx-y_i)$, where $a_i^T$ are the rows of A and $\psi$ is ${\cal C}^m$ on ${\mathbb R}\setminus \{0\}$. We show that if $\psi\1(0^-) < \psi\1(0^+)$, then typical data y give rise to local minimizers $\hat x$ of ${\cal F}(.,y)$ which fit exactly a certain number of the data entries: there is a possibly large set $\hat h$ of indexes such that $a_i^T\hat x=y_i$ for every $i\in\hat h$. In contrast, if $\psi$ is smooth on ${\mathbb R}$, for almost every y, the local minimizers of ${\cal F}(.,y)$ do not fit any entry of y. Thus, the possibility that a local minimizer fits some data entries is due to the nonsmoothness of the data-fidelity term. This is a strong mathematical property which is useful in practice. By way of application, we construct a cost-function allowing aberrant data (outliers) to be detected and to be selectively smoothed. Our numerical experiments advocate the use of nonsmooth data-fidelity terms in regularized cost-functions for special purposes in image and signal processing.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Nonlinear total variation based noise removal algorithms
Rudin, L. I., Osher, S., & Fatemi, E. (1992) · Physica D: Nonlinear Phenomena · doi:10.1016/0167-2789(92)90242-f · Cited by 15,704
Scale-space and edge detection using anisotropic diffusion
Perona, P., & Malik, J. (1990) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.56205 · Cited by 12,119
Solutions of Ill-Posed Problems.
Bell, J. B., Tikhonov, A. N., & Arsenin, V. Y. (1978) · Mathematics of Computation · doi:10.2307/2006360 · Cited by 8,101
Principles of Computerized Tomographic Imaging
Kak, A. C., & Slaney, M. (2001) · doi:10.1137/1.9780898719277 · Cited by 5,809
On the Statistical Analysis of Dirty Pictures
Besag, J. (1986) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1986.tb01412.x · Cited by 4,191
Rockafellar, R. T., & Wets, R. J. B. (1998) · Grundlehren der mathematischen Wissenschaften · doi:10.1007/978-3-642-02431-3 · Cited by 4,086
Anisotropic diffusion in image processing
Weickert, J. (1998) · Cited by 2,457
Inverse Problem Theory: Methods for Data Fitting and Model Parameter Estimation
Tarantola, A. (1987) · International Linear Collider · Cited by 2,193
Principles of Computerized Tomographic Imaging
Kak, A. C., Slaney, M., & Wang, G. (2002) · Medical Physics · doi:10.1118/1.1455742 · Cited by 1,877
Image recovery via total variation minimization and related problems
Chambolle, A., & Lions, P.-L. (1997) · Numerische Mathematik · doi:10.1007/s002110050258 · Cited by 1,569
Markov Random Field Modeling in Computer Vision
Li, S. Z. (1995) · Computer Science Workbench · doi:10.1007/978-4-431-66933-3 · Cited by 1,276
Bayesian reconstructions from emission tomography data using a modified EM algorithm
Green, P. J. (1990) · IEEE Transactions on Medical Imaging · doi:10.1109/42.52985 · Cited by 1,241
Total variation blind deconvolution
Chan, T. F., & Wong, C.-K. (1998) · IEEE Transactions on Image Processing · doi:10.1109/83.661187 · Cited by 1,201
Constrained restoration and the recovery of discontinuities
Geman, D., & Reynolds, G. (1992) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.120331 · Cited by 1,085
Nonlinear image recovery with half-quadratic regularization
Geman, D., & Yang, C. (1995) · IEEE Transactions on Image Processing · doi:10.1109/83.392335 · Cited by 1,000 · cited 1×
“The concave function was shown to transform ramp-shaped data into a step-shaped minimizer [19].”
A generalized Gaussian image model for edge-preserving MAP estimation
Bouman, C., & Sauer, K. (1993) · IEEE Transactions on Image Processing · doi:10.1109/83.236536 · Cited by 884
Analysis of bounded variation penalty methods for ill-posed problems
Acar, R., & Vogel, C. R. (1994) · Inverse Problems · doi:10.1088/0266-5611/10/6/003 · Cited by 818
Black, M. J., & Rangarajan, A. (1996) · International Journal of Computer Vision · doi:10.1007/bf00131148 · Cited by 757
A view of three decades of linear filtering theory
Kailath, T. (1974) · IEEE Transactions on Information Theory · doi:10.1109/tit.1974.1055174 · Cited by 702
Image reconstruction and restoration: overview of common estimation structures and problems
Demoment, G. (1989) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/29.45551 · Cited by 647
Fast, robust total variation-based reconstruction of noisy, blurred images
Vogel, C. R., & Oman, M. E. (1998) · IEEE Transactions on Image Processing · doi:10.1109/83.679423 · Cited by 601
Statistical methods for tomographic image reconstruction
Geman, S., & McClure, D. E. (1987) · Medical Entomology and Zoology · Cited by 579
Bloomfield, P., & Steiger, W. L. (1984) · CERN Document Server (European Organization for Nuclear Research) · doi:10.1007/978-1-4684-8574-5 · Cited by 553
Least Squares Methods for Ill-Posed Problems with a Prescribed Bound
Miller, K. (1970) · SIAM Journal on Mathematical Analysis · doi:10.1137/0501006 · Cited by 504
A local update strategy for iterative reconstruction from projections
Sauer, K., & Bouman, C. (1993) · IEEE Transactions on Signal Processing · doi:10.1109/78.193196 · Cited by 494
A unified approach to statistical tomography using coordinate descent optimization
Bouman, C. A., & Sauer, K. (1996) · IEEE Transactions on Image Processing · doi:10.1109/83.491321 · Cited by 442
Maximum Entropy and the Nearly Black Object
Donoho, D. L., Johnstone, I. M., Hoch, J. C., & Stern, A. S. (1992) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1992.tb01864.x · Cited by 393
Local Strong Homogeneity of a Regularized Estimator
Nikolova, M. (2000) · SIAM Journal on Applied Mathematics · doi:10.1137/s0036139997327794 · Cited by 291
Self-citationRecovery of Blocky Images from Noisy and Blurred Data
Dobson, D. C., & Santosa, F. (1996) · SIAM Journal on Applied Mathematics · doi:10.1137/s003613999427560x · Cited by 260
Random fields and inverse problems in imaging
Geman, D. (1990) · Lecture Notes in Mathematics · doi:10.1007/bfb0103042 · Cited by 214
Digital filters as absolute norm regularizers
Alliney, S. (1992) · IEEE Transactions on Signal Processing · doi:10.1109/78.139258 · Cited by 116
Estimées localement fortement homogènes
Nikolova, M. (1997) · Comptes Rendus de l Académie des Sciences - Series I - Mathematics · doi:10.1016/s0764-4442(97)84780-5 · Cited by 27
Self-citation
Cited by
Salt-and-pepper noise removal by median-type noise detectors and detail-preserving regularization
Chan, R. H., Chung-Wa, & Nikolova, M. (2005) · IEEE Transactions on Image Processing · doi:10.1109/tip.2005.852196 · Cited by 1,152
Self-citationAlgorithms for Finding Global Minimizers of Image Segmentation and Denoising Models
Chan, T. F., Esedoglu, S., & Nikolova, M. (2006) · SIAM Journal on Applied Mathematics · doi:10.1137/040615286 · Cited by 1,049
Self-citationFast Global Minimization of the Active Contour/Snake Model
Bresson, X., Esedoḡlu, S., Vandergheynst, P., Thiran, J.-P., & Osher, S. (2007) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-007-0002-0 · Cited by 857
Split Bregman Methods and Frame Based Image Restoration
Cai, J.-F., Osher, S., & Shen, Z. (2010) · Multiscale Modeling & Simulation · doi:10.1137/090753504 · Cited by 670
Aspects of Total Variation Regularized L1Function Approximation
Chan, T. F., & Esedoglu, S. (2005) · SIAM Journal on Applied Mathematics · doi:10.1137/040604297 · Cited by 651
Metadata
- Authors
- Nikolova, M.
- Venue
- SIAM Journal on Numerical Analysis
- Published
- 2002-01
- Volume
- 40
- Issue
- 3
- Pages
- 965-994
- ISSN
- 0036-1429 (ISSN-L)
- 1095-7170
- OpenAlex
- W2111841405
- Cited by
- 326
- References
- 32
- FWCI
- 12.36
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.