Generalized mixture ratio based blind image separation

Hara Kenji, Kohei Inoue, Kiichi Urahama

Research output: Contribution to journalArticle

Abstract

Iterative approaches have been mostly used for blind source separation. The success of iterative methods generally depends upon appropriate selection of the values of the input parameters such as initial conditions and control parameters. We propose a direct non-iterative method for blindly separating linearly mixed images using the boundary values of the determinant of the relative transformation matrix between the mixtures. The algorithm is a non-trivial extension of the mixture ratio based algorithm originally proposed by Diamantaras and Papadimitriou (2005) and allows for a finite arbitrary number of source images. We demonstrate the performance of our method on synthetic images.

Original languageEnglish
Article number6522461
Pages (from-to)743-746
Number of pages4
JournalIEEE Signal Processing Letters
Volume20
Issue number8
DOIs
Publication statusPublished - Jun 24 2013

Fingerprint

Blind source separation
Iterative methods
Transformation Matrix
Blind Source Separation
Boundary Value
Control Parameter
Determinant
Initial conditions
Linearly
Iteration
Arbitrary
Demonstrate

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Generalized mixture ratio based blind image separation. / Kenji, Hara; Inoue, Kohei; Urahama, Kiichi.

In: IEEE Signal Processing Letters, Vol. 20, No. 8, 6522461, 24.06.2013, p. 743-746.

Research output: Contribution to journalArticle

@article{825b3480a1054aee886015c1fc1422aa,
title = "Generalized mixture ratio based blind image separation",
abstract = "Iterative approaches have been mostly used for blind source separation. The success of iterative methods generally depends upon appropriate selection of the values of the input parameters such as initial conditions and control parameters. We propose a direct non-iterative method for blindly separating linearly mixed images using the boundary values of the determinant of the relative transformation matrix between the mixtures. The algorithm is a non-trivial extension of the mixture ratio based algorithm originally proposed by Diamantaras and Papadimitriou (2005) and allows for a finite arbitrary number of source images. We demonstrate the performance of our method on synthetic images.",
author = "Hara Kenji and Kohei Inoue and Kiichi Urahama",
year = "2013",
month = "6",
day = "24",
doi = "10.1109/LSP.2013.2265274",
language = "English",
volume = "20",
pages = "743--746",
journal = "IEEE Signal Processing Letters",
issn = "1070-9908",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - Generalized mixture ratio based blind image separation

AU - Kenji, Hara

AU - Inoue, Kohei

AU - Urahama, Kiichi

PY - 2013/6/24

Y1 - 2013/6/24

N2 - Iterative approaches have been mostly used for blind source separation. The success of iterative methods generally depends upon appropriate selection of the values of the input parameters such as initial conditions and control parameters. We propose a direct non-iterative method for blindly separating linearly mixed images using the boundary values of the determinant of the relative transformation matrix between the mixtures. The algorithm is a non-trivial extension of the mixture ratio based algorithm originally proposed by Diamantaras and Papadimitriou (2005) and allows for a finite arbitrary number of source images. We demonstrate the performance of our method on synthetic images.

AB - Iterative approaches have been mostly used for blind source separation. The success of iterative methods generally depends upon appropriate selection of the values of the input parameters such as initial conditions and control parameters. We propose a direct non-iterative method for blindly separating linearly mixed images using the boundary values of the determinant of the relative transformation matrix between the mixtures. The algorithm is a non-trivial extension of the mixture ratio based algorithm originally proposed by Diamantaras and Papadimitriou (2005) and allows for a finite arbitrary number of source images. We demonstrate the performance of our method on synthetic images.

UR - http://www.scopus.com/inward/record.url?scp=84879112227&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879112227&partnerID=8YFLogxK

U2 - 10.1109/LSP.2013.2265274

DO - 10.1109/LSP.2013.2265274

M3 - Article

VL - 20

SP - 743

EP - 746

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

SN - 1070-9908

IS - 8

M1 - 6522461

ER -