Multiscale Image Decompositions and Wavelets

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

This chapter introduces the basic concepts of multiscale image decompositions and wavelets. It focuses on three main techniques-Gaussian pyramids, Laplacian pyramids, and wavelets. The Gaussian pyramid provides a representation of the same image at multiple scales, using simple low-pass filtering and decimation techniques. The Laplacian pyramid provides a coarse representation of the image as well as a set of detailed images at different scales. Both the Gaussian and the Laplacian representations are over-complete in the sense that the total number of pixels is approximately 33% higher than in the original image. Wavelet decompositions are a more recent addition to the arsenal of multiscale signal processing techniques. Unlike the Gaussian and Laplacian pyramids, they provide a complete image representation and perform a decomposition according to both scale and orientation. They are implemented using cascaded filter banks in which the lowpass and highpass filters satisfy certain specific constraints. Although classical signal processing concepts provide an operational understanding of such systems, there exist remarkable connections with work in applied mathematics and in psychophysics, providing a deeper understanding of wavelet decompositions and their role in vision. From a mathematical standpoint, wavelet decompositions are equivalent to signal expansions in a wavelet basis.

Original languageEnglish (US)
Title of host publicationHandbook of Image and Video Processing
PublisherElsevier Inc.
Pages347-359
Number of pages13
ISBN (Print)9780121197926
DOIs
StatePublished - Dec 1 2005

Fingerprint

Wavelet decomposition
Signal processing
Arsenals
Filter banks
Pixels

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Moulin, P. (2005). Multiscale Image Decompositions and Wavelets. In Handbook of Image and Video Processing (pp. 347-359). Elsevier Inc.. https://doi.org/10.1016/B978-012119792-6/50084-X

Multiscale Image Decompositions and Wavelets. / Moulin, Pierre.

Handbook of Image and Video Processing. Elsevier Inc., 2005. p. 347-359.

Research output: Chapter in Book/Report/Conference proceedingChapter

Moulin, P 2005, Multiscale Image Decompositions and Wavelets. in Handbook of Image and Video Processing. Elsevier Inc., pp. 347-359. https://doi.org/10.1016/B978-012119792-6/50084-X
Moulin P. Multiscale Image Decompositions and Wavelets. In Handbook of Image and Video Processing. Elsevier Inc. 2005. p. 347-359 https://doi.org/10.1016/B978-012119792-6/50084-X
Moulin, Pierre. / Multiscale Image Decompositions and Wavelets. Handbook of Image and Video Processing. Elsevier Inc., 2005. pp. 347-359
@inbook{2d3cad277e554d288e68041dd9cefee0,
title = "Multiscale Image Decompositions and Wavelets",
abstract = "This chapter introduces the basic concepts of multiscale image decompositions and wavelets. It focuses on three main techniques-Gaussian pyramids, Laplacian pyramids, and wavelets. The Gaussian pyramid provides a representation of the same image at multiple scales, using simple low-pass filtering and decimation techniques. The Laplacian pyramid provides a coarse representation of the image as well as a set of detailed images at different scales. Both the Gaussian and the Laplacian representations are over-complete in the sense that the total number of pixels is approximately 33{\%} higher than in the original image. Wavelet decompositions are a more recent addition to the arsenal of multiscale signal processing techniques. Unlike the Gaussian and Laplacian pyramids, they provide a complete image representation and perform a decomposition according to both scale and orientation. They are implemented using cascaded filter banks in which the lowpass and highpass filters satisfy certain specific constraints. Although classical signal processing concepts provide an operational understanding of such systems, there exist remarkable connections with work in applied mathematics and in psychophysics, providing a deeper understanding of wavelet decompositions and their role in vision. From a mathematical standpoint, wavelet decompositions are equivalent to signal expansions in a wavelet basis.",
author = "Pierre Moulin",
year = "2005",
month = "12",
day = "1",
doi = "10.1016/B978-012119792-6/50084-X",
language = "English (US)",
isbn = "9780121197926",
pages = "347--359",
booktitle = "Handbook of Image and Video Processing",
publisher = "Elsevier Inc.",

}

TY - CHAP

T1 - Multiscale Image Decompositions and Wavelets

AU - Moulin, Pierre

PY - 2005/12/1

Y1 - 2005/12/1

N2 - This chapter introduces the basic concepts of multiscale image decompositions and wavelets. It focuses on three main techniques-Gaussian pyramids, Laplacian pyramids, and wavelets. The Gaussian pyramid provides a representation of the same image at multiple scales, using simple low-pass filtering and decimation techniques. The Laplacian pyramid provides a coarse representation of the image as well as a set of detailed images at different scales. Both the Gaussian and the Laplacian representations are over-complete in the sense that the total number of pixels is approximately 33% higher than in the original image. Wavelet decompositions are a more recent addition to the arsenal of multiscale signal processing techniques. Unlike the Gaussian and Laplacian pyramids, they provide a complete image representation and perform a decomposition according to both scale and orientation. They are implemented using cascaded filter banks in which the lowpass and highpass filters satisfy certain specific constraints. Although classical signal processing concepts provide an operational understanding of such systems, there exist remarkable connections with work in applied mathematics and in psychophysics, providing a deeper understanding of wavelet decompositions and their role in vision. From a mathematical standpoint, wavelet decompositions are equivalent to signal expansions in a wavelet basis.

AB - This chapter introduces the basic concepts of multiscale image decompositions and wavelets. It focuses on three main techniques-Gaussian pyramids, Laplacian pyramids, and wavelets. The Gaussian pyramid provides a representation of the same image at multiple scales, using simple low-pass filtering and decimation techniques. The Laplacian pyramid provides a coarse representation of the image as well as a set of detailed images at different scales. Both the Gaussian and the Laplacian representations are over-complete in the sense that the total number of pixels is approximately 33% higher than in the original image. Wavelet decompositions are a more recent addition to the arsenal of multiscale signal processing techniques. Unlike the Gaussian and Laplacian pyramids, they provide a complete image representation and perform a decomposition according to both scale and orientation. They are implemented using cascaded filter banks in which the lowpass and highpass filters satisfy certain specific constraints. Although classical signal processing concepts provide an operational understanding of such systems, there exist remarkable connections with work in applied mathematics and in psychophysics, providing a deeper understanding of wavelet decompositions and their role in vision. From a mathematical standpoint, wavelet decompositions are equivalent to signal expansions in a wavelet basis.

UR - http://www.scopus.com/inward/record.url?scp=84879242569&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879242569&partnerID=8YFLogxK

U2 - 10.1016/B978-012119792-6/50084-X

DO - 10.1016/B978-012119792-6/50084-X

M3 - Chapter

AN - SCOPUS:84879242569

SN - 9780121197926

SP - 347

EP - 359

BT - Handbook of Image and Video Processing

PB - Elsevier Inc.

ER -