A linear manifold color descriptor for medicine package recognition

Kenjiro Sugimoto, Koji Inoue, Yoshimitsu Kuroki, Seiichiro Kamata

Research output: Contribution to journalArticle

Abstract

This paper presents a color-based method for medicine package recognition, called a linear manifold color descriptor (LMCD). It describes a color distribution (a set of color pixels) of a color package image as a linear manifold (an affine subspace) in the color space, and recognizes an anonymous package by linear manifold matching. Mainly due to low dimensionality of color spaces, LMCD can provide more compact description and faster computation than description styles based on histogram and dominant-color. This paper also proposes distance-based dissimilarities for linear manifold matching. Specially designed for color distribution matching, the proposed dissimilarities are theoretically appropriate more than J-divergence and canonical angles. Experiments on medicine package recognition validates that LMCD outperforms competitors including MPEG-7 color descriptors in terms of description size, computational cost and recognition rate.

Original languageEnglish
Pages (from-to)1264-1271
Number of pages8
JournalIEICE Transactions on Information and Systems
VolumeE95-D
Issue number5
DOIs
Publication statusPublished - 2012 May

Fingerprint

Medicine
Color
Pixels

Keywords

  • Linear manifold
  • Low-level color descriptor
  • Medicine package recognition

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

A linear manifold color descriptor for medicine package recognition. / Sugimoto, Kenjiro; Inoue, Koji; Kuroki, Yoshimitsu; Kamata, Seiichiro.

In: IEICE Transactions on Information and Systems, Vol. E95-D, No. 5, 05.2012, p. 1264-1271.

Research output: Contribution to journalArticle

@article{1ab3e75e0841422b974b5db97fd1947b,
title = "A linear manifold color descriptor for medicine package recognition",
abstract = "This paper presents a color-based method for medicine package recognition, called a linear manifold color descriptor (LMCD). It describes a color distribution (a set of color pixels) of a color package image as a linear manifold (an affine subspace) in the color space, and recognizes an anonymous package by linear manifold matching. Mainly due to low dimensionality of color spaces, LMCD can provide more compact description and faster computation than description styles based on histogram and dominant-color. This paper also proposes distance-based dissimilarities for linear manifold matching. Specially designed for color distribution matching, the proposed dissimilarities are theoretically appropriate more than J-divergence and canonical angles. Experiments on medicine package recognition validates that LMCD outperforms competitors including MPEG-7 color descriptors in terms of description size, computational cost and recognition rate.",
keywords = "Linear manifold, Low-level color descriptor, Medicine package recognition",
author = "Kenjiro Sugimoto and Koji Inoue and Yoshimitsu Kuroki and Seiichiro Kamata",
year = "2012",
month = "5",
doi = "10.1587/transinf.E95.D.1264",
language = "English",
volume = "E95-D",
pages = "1264--1271",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "5",

}

TY - JOUR

T1 - A linear manifold color descriptor for medicine package recognition

AU - Sugimoto, Kenjiro

AU - Inoue, Koji

AU - Kuroki, Yoshimitsu

AU - Kamata, Seiichiro

PY - 2012/5

Y1 - 2012/5

N2 - This paper presents a color-based method for medicine package recognition, called a linear manifold color descriptor (LMCD). It describes a color distribution (a set of color pixels) of a color package image as a linear manifold (an affine subspace) in the color space, and recognizes an anonymous package by linear manifold matching. Mainly due to low dimensionality of color spaces, LMCD can provide more compact description and faster computation than description styles based on histogram and dominant-color. This paper also proposes distance-based dissimilarities for linear manifold matching. Specially designed for color distribution matching, the proposed dissimilarities are theoretically appropriate more than J-divergence and canonical angles. Experiments on medicine package recognition validates that LMCD outperforms competitors including MPEG-7 color descriptors in terms of description size, computational cost and recognition rate.

AB - This paper presents a color-based method for medicine package recognition, called a linear manifold color descriptor (LMCD). It describes a color distribution (a set of color pixels) of a color package image as a linear manifold (an affine subspace) in the color space, and recognizes an anonymous package by linear manifold matching. Mainly due to low dimensionality of color spaces, LMCD can provide more compact description and faster computation than description styles based on histogram and dominant-color. This paper also proposes distance-based dissimilarities for linear manifold matching. Specially designed for color distribution matching, the proposed dissimilarities are theoretically appropriate more than J-divergence and canonical angles. Experiments on medicine package recognition validates that LMCD outperforms competitors including MPEG-7 color descriptors in terms of description size, computational cost and recognition rate.

KW - Linear manifold

KW - Low-level color descriptor

KW - Medicine package recognition

UR - http://www.scopus.com/inward/record.url?scp=84860628740&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860628740&partnerID=8YFLogxK

U2 - 10.1587/transinf.E95.D.1264

DO - 10.1587/transinf.E95.D.1264

M3 - Article

VL - E95-D

SP - 1264

EP - 1271

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 5

ER -