A Lapped Directional Transform for Spectral Image Analysis and Its Application to Restoration and Enhancement

Til Aach; Dietmar Kunz

doi:10.1016/S0165-1684(00)00122-5

A Lapped Directional Transform for Spectral Image Analysis and Its Application to Restoration and Enhancement

Til Aach, Dietmar Kunz

Institut für Signalverarbeitung

Philips Research Laboratories

Abstract

Wir beschreiben eine neue reellwertige überlappende Transformation für 2D-Signal- und Bildverarbeitung. Überlappende Transformationen sind besonders geeignet für blockweise Verarbeitung, da sie Blockartefakte reduzieren oder vermeiden. Die neue Transformation wird aus der reellwertigen und separierbaren Modulierten Überlappenden Transformation (MLT) abgeleitet. Im Gegensatz zur Diskreten Fourier-Transformation (DFT) erlauben die MLT wie auch die Diskrete Cosinus-Transformation (DCT) keine eindeutige Detektion von Orientierung allein auf der Basis spektraler Energieverteilung. Die neue überlappende Transformation ist reellwertig, und ermöglicht eindeutige Orientierungsdetektion aus der spektralen Energieverteilung. Ein schneller, separierbarer Algorithmus für die Transformation existiert. Die Transformation wird in spektralen Ansätzen zur Bildrestauration und Verbesserung eingesetzt, und mit der DFT verglichen.

Originalsprache	Englisch
Zeitschrift	Signal Processing
Jahrgang	80
Ausgabenummer	11
Seiten (von - bis)	2347-2364
Seitenumfang	18
ISSN	0165-1684
DOIs	https://doi.org/10.1016/S0165-1684(00)00122-5
Publikationsstatus	Veröffentlicht - 2000

UN SDGs

Dieser Output leistet einen Beitrag zu folgendem(n) Ziel(en) für nachhaltige Entwicklung

Dokumente

10.1016/S0165-1684(00)00122-5

Weitere Links

semanticscholar.org

Zitieren

@article{ee796d9efefb4874aaf1153961796c13,

title = "A Lapped Directional Transform for Spectral Image Analysis and Its Application to Restoration and Enhancement",

abstract = "We describe a new real-valued lapped transform for 2D-signal and image processing. Lapped transforms are particularly useful in block-based processing, since their overlapping basis functions reduce or prevent block artifacts. Our transform is derived from the modulated lapped transform (MLT), which is a real-valued and separable transform. Like the discrete cosine transform, the MLT does not allow to unambiguously identify spatial orientation from modulus spectra or spectral energy. This is in marked contrast to the complex-valued discrete Fourier transform (DFT). The new lapped transform is real valued, and at the same time allows unambiguous detection of spatial orientation from spectral energy. Furthermore, a fast and separable algorithm for this transform exists. As an application example, we investigate the transform's performance in anisotropic spectral approaches to image restoration and enhancement, and compare it to the DFT.",

author = "Til Aach and Dietmar Kunz",

year = "2000",

doi = "10.1016/S0165-1684(00)00122-5",

language = "English",

volume = "80",

pages = "2347--2364",

journal = "Signal Processing",

issn = "0165-1684",

number = "11",

}

TY - JOUR

T1 - A Lapped Directional Transform for Spectral Image Analysis and Its Application to Restoration and Enhancement

AU - Aach, Til

AU - Kunz, Dietmar

PY - 2000

Y1 - 2000

N2 - We describe a new real-valued lapped transform for 2D-signal and image processing. Lapped transforms are particularly useful in block-based processing, since their overlapping basis functions reduce or prevent block artifacts. Our transform is derived from the modulated lapped transform (MLT), which is a real-valued and separable transform. Like the discrete cosine transform, the MLT does not allow to unambiguously identify spatial orientation from modulus spectra or spectral energy. This is in marked contrast to the complex-valued discrete Fourier transform (DFT). The new lapped transform is real valued, and at the same time allows unambiguous detection of spatial orientation from spectral energy. Furthermore, a fast and separable algorithm for this transform exists. As an application example, we investigate the transform's performance in anisotropic spectral approaches to image restoration and enhancement, and compare it to the DFT.

AB - We describe a new real-valued lapped transform for 2D-signal and image processing. Lapped transforms are particularly useful in block-based processing, since their overlapping basis functions reduce or prevent block artifacts. Our transform is derived from the modulated lapped transform (MLT), which is a real-valued and separable transform. Like the discrete cosine transform, the MLT does not allow to unambiguously identify spatial orientation from modulus spectra or spectral energy. This is in marked contrast to the complex-valued discrete Fourier transform (DFT). The new lapped transform is real valued, and at the same time allows unambiguous detection of spatial orientation from spectral energy. Furthermore, a fast and separable algorithm for this transform exists. As an application example, we investigate the transform's performance in anisotropic spectral approaches to image restoration and enhancement, and compare it to the DFT.

UR - https://www.semanticscholar.org/paper/A-lapped-directional-transform-for-spectral-image-Aach-Kunz/9951270975e352ba15ffccddde1f8ff983bae075

U2 - 10.1016/S0165-1684(00)00122-5

DO - 10.1016/S0165-1684(00)00122-5

M3 - Journal articles

SN - 0165-1684

VL - 80

SP - 2347

EP - 2364

JO - Signal Processing

JF - Signal Processing

IS - 11

ER -

A Lapped Directional Transform for Spectral Image Analysis and Its Application to Restoration and Enhancement

Abstract

UN SDGs

Dokumente

Weitere Links

Fingerprint

Zitieren