Block-median pyramidal transform: Analysis and denoising applications

Vladimir P. Melnik; Ilya Shmulevich; Karen Egiazarian; Jaakko Astola

doi:10.1109/78.902119

Block-median pyramidal transform: Analysis and denoising applications

Vladimir P. Melnik
, Ilya Shmulevich
, Karen Egiazarian
, Jaakko Astola

Research output: Contribution to journal › Article › peer-review

13 Scopus citations

Abstract

A nonlinear multiscale pyramidal transform based on nonoverlapping block decompositions using the median operation and a polynomial approximation is considered. It is shown that this structure can be useful for denoising of one- and two-dimensional (1-D and 2-D) signals. Various denoising techniques are analyzed, including methods based on spatially adaptive thresholding and partial cycle-spinning algorithms. An analytical method for deriving the distribution function of the transform coefficients is also presented. This, in turn, can be used for the selection of thresholds for denoising applications.

Original language	English
Pages (from-to)	364-372
Number of pages	9
Journal	IEEE Transactions on Signal Processing
Volume	49
Issue number	2
DOIs	https://doi.org/10.1109/78.902119
State	Published - Feb 2001
Externally published	Yes

Access to Document

10.1109/78.902119

Cite this

@article{4ac4f300217e40ca9ce94c639187da38,

title = "Block-median pyramidal transform: Analysis and denoising applications",

abstract = "A nonlinear multiscale pyramidal transform based on nonoverlapping block decompositions using the median operation and a polynomial approximation is considered. It is shown that this structure can be useful for denoising of one- and two-dimensional (1-D and 2-D) signals. Various denoising techniques are analyzed, including methods based on spatially adaptive thresholding and partial cycle-spinning algorithms. An analytical method for deriving the distribution function of the transform coefficients is also presented. This, in turn, can be used for the selection of thresholds for denoising applications.",

author = "Melnik, \{Vladimir P.\} and Ilya Shmulevich and Karen Egiazarian and Jaakko Astola",

note = "Funding Information: This conclusion is supported by the results of X-ray di€ractionstudiesonthesethreecompounds[14± 16]. Each compound forms molecular pairs in the crystalline state, where two molecules are aligned parallel across the inversion centre, but the extent of the overlap of the core parts is di€ erent. In the crystalline states of 3-BBCN",

year = "2001",

month = feb,

doi = "10.1109/78.902119",

language = "English",

volume = "49",

pages = "364--372",

journal = "IEEE Transactions on Signal Processing",

issn = "1053-587X",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - Block-median pyramidal transform

T2 - Analysis and denoising applications

AU - Melnik, Vladimir P.

AU - Shmulevich, Ilya

AU - Egiazarian, Karen

AU - Astola, Jaakko

N1 - Funding Information: This conclusion is supported by the results of X-ray di€ractionstudiesonthesethreecompounds[14± 16]. Each compound forms molecular pairs in the crystalline state, where two molecules are aligned parallel across the inversion centre, but the extent of the overlap of the core parts is di€ erent. In the crystalline states of 3-BBCN

PY - 2001/2

Y1 - 2001/2

N2 - A nonlinear multiscale pyramidal transform based on nonoverlapping block decompositions using the median operation and a polynomial approximation is considered. It is shown that this structure can be useful for denoising of one- and two-dimensional (1-D and 2-D) signals. Various denoising techniques are analyzed, including methods based on spatially adaptive thresholding and partial cycle-spinning algorithms. An analytical method for deriving the distribution function of the transform coefficients is also presented. This, in turn, can be used for the selection of thresholds for denoising applications.

AB - A nonlinear multiscale pyramidal transform based on nonoverlapping block decompositions using the median operation and a polynomial approximation is considered. It is shown that this structure can be useful for denoising of one- and two-dimensional (1-D and 2-D) signals. Various denoising techniques are analyzed, including methods based on spatially adaptive thresholding and partial cycle-spinning algorithms. An analytical method for deriving the distribution function of the transform coefficients is also presented. This, in turn, can be used for the selection of thresholds for denoising applications.

UR - https://www.scopus.com/pages/publications/0035249389

U2 - 10.1109/78.902119

DO - 10.1109/78.902119

M3 - Article

AN - SCOPUS:0035249389

SN - 1053-587X

VL - 49

SP - 364

EP - 372

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

IS - 2

ER -

Block-median pyramidal transform: Analysis and denoising applications

Abstract

Access to Document

Other files and links

Fingerprint

Cite this