Characterizing the Structure Tensor Using Gamma Distributions

Magnus Oskarsson

doi:10.1109/ICPR.2016.7899727

Characterizing the Structure Tensor Using Gamma Distributions

Magnus Oskarsson

Forskningsoutput: Kapitel i bok/rapport/Conference proceeding › Konferenspaper i proceeding › Peer review

237 Nedladdningar (Pure)

Sammanfattning

The structure tensor is a powerful tool describing the local intensity structure of an image or image sequence. In this paper we give a model for the noise distribution of the components of the tensor. In order to do so we have also investigated some properties of the gamma distribution. We show that, given an input image corrupted with Gaussian noise, the noise in the structure tensor can be modeled well by gamma distributions. We apply our model to automatic contrast enhancement of images taken under poor illumination. We show how our noise model can be used for automatic parameter selection in the filtering process, giving powerful results without the need for cumbersome parameter tuning.

Originalspråk	engelska
Titel på värdpublikation	Pattern Recognition (ICPR), 2016 23rd International Conference on
Förlag	IEEE - Institute of Electrical and Electronics Engineers Inc.
Sidor	763-768
Antal sidor	6
ISBN (elektroniskt)	978-1-5090-4847-2
DOI	https://doi.org/10.1109/ICPR.2016.7899727
Status	Published - 2017 apr. 24
Evenemang	2016 23rd International Conference on Pattern Recognition (ICPR 2016) - Cancún Center, Cancún, Mexiko Varaktighet: 2016 dec. 4 → 2016 dec. 8 http://www.icpr2016.org/site/

Konferens

Konferens	2016 23rd International Conference on Pattern Recognition (ICPR 2016)
Förkortad titel	ICPR 2016
Land/Territorium	Mexiko
Ort	Cancún
Period	2016/12/04 → 2016/12/08
Internetadress	http://www.icpr2016.org/site/

Ämnesklassifikation (UKÄ)

Sannolikhetsteori och statistik
Signalbehandling
Datorseende och robotik (autonoma system)

Tillgång till dokument

10.1109/ICPR.2016.7899727

structure_stats_finalPublicerad version, 8,43 MB

Andra filer och länkar

Link to publication in Scopus

Citera det här

@inproceedings{3a7667c86add41f6a31ff4601968c163,

title = "Characterizing the Structure Tensor Using Gamma Distributions",

abstract = "The structure tensor is a powerful tool describing the local intensity structure of an image or image sequence. In this paper we give a model for the noise distribution of the components of the tensor. In order to do so we have also investigated some properties of the gamma distribution. We show that, given an input image corrupted with Gaussian noise, the noise in the structure tensor can be modeled well by gamma distributions. We apply our model to automatic contrast enhancement of images taken under poor illumination. We show how our noise model can be used for automatic parameter selection in the filtering process, giving powerful results without the need for cumbersome parameter tuning. ",

keywords = "Enhancement, restoration and filtering, Signal, image and video processing, Computational photography",

author = "Magnus Oskarsson",

year = "2017",

month = apr,

day = "24",

doi = "10.1109/ICPR.2016.7899727",

language = "English",

pages = "763--768",

booktitle = "Pattern Recognition (ICPR), 2016 23rd International Conference on",

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

address = "United States",

note = "2016 23rd International Conference on Pattern Recognition (ICPR 2016), ICPR 2016 ; Conference date: 04-12-2016 Through 08-12-2016",

url = "http://www.icpr2016.org/site/",

}

TY - GEN

T1 - Characterizing the Structure Tensor Using Gamma Distributions

AU - Oskarsson, Magnus

PY - 2017/4/24

Y1 - 2017/4/24

N2 - The structure tensor is a powerful tool describing the local intensity structure of an image or image sequence. In this paper we give a model for the noise distribution of the components of the tensor. In order to do so we have also investigated some properties of the gamma distribution. We show that, given an input image corrupted with Gaussian noise, the noise in the structure tensor can be modeled well by gamma distributions. We apply our model to automatic contrast enhancement of images taken under poor illumination. We show how our noise model can be used for automatic parameter selection in the filtering process, giving powerful results without the need for cumbersome parameter tuning.

AB - The structure tensor is a powerful tool describing the local intensity structure of an image or image sequence. In this paper we give a model for the noise distribution of the components of the tensor. In order to do so we have also investigated some properties of the gamma distribution. We show that, given an input image corrupted with Gaussian noise, the noise in the structure tensor can be modeled well by gamma distributions. We apply our model to automatic contrast enhancement of images taken under poor illumination. We show how our noise model can be used for automatic parameter selection in the filtering process, giving powerful results without the need for cumbersome parameter tuning.

KW - Enhancement

KW - restoration and filtering

KW - Signal, image and video processing

KW - Computational photography

UR - http://www.scopus.com/inward/record.url?scp=85019092215&partnerID=8YFLogxK

U2 - 10.1109/ICPR.2016.7899727

DO - 10.1109/ICPR.2016.7899727

M3 - Paper in conference proceeding

SP - 763

EP - 768

BT - Pattern Recognition (ICPR), 2016 23rd International Conference on

PB - IEEE - Institute of Electrical and Electronics Engineers Inc.

T2 - 2016 23rd International Conference on Pattern Recognition (ICPR 2016)

Y2 - 4 December 2016 through 8 December 2016

ER -

Characterizing the Structure Tensor Using Gamma Distributions

Sammanfattning

Konferens

Ämnesklassifikation (UKÄ)

Tillgång till dokument

Andra filer och länkar

Fingeravtryck

Citera det här