An MRF model-based approach to the detection of rectangular shape objects in color images

Yangxing Liu, Takeshi Ikenaga, Satoshi Goto

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

Rectangular shape object detection in color images is a critical step of many image recognition systems. However, there are few reports on this matter. In this paper, we proposed a hierarchical approach, which combines a global contour-based line segment detection algorithm and an Markov random field (MRF) model, to extract rectangular shape objects from real color images. Firstly, we use an elaborate edge detection algorithm to obtain image edge map and accurate edge pixel gradient information (magnitude and direction). Then line segments are extracted from the edge map and some neighboring parallel segments are merged into a single line segment. Finally all segments lying on the boundary of unknown rectangular shape objects are labeled via an MRF model built on line segments. Experimental results show that our method is robust in locating multiple rectangular shape objects simultaneously with respect to different size, orientation and color.

Original languageEnglish
Pages (from-to)2649-2658
Number of pages10
JournalSignal Processing
Volume87
Issue number11
DOIs
Publication statusPublished - 2007 Nov

Fingerprint

Color
Image recognition
Edge detection
Pixels
Object detection

Keywords

  • Edge detection
  • Line segment detection
  • MRF model
  • Randomized Hough transform
  • Rectangle detection

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

An MRF model-based approach to the detection of rectangular shape objects in color images. / Liu, Yangxing; Ikenaga, Takeshi; Goto, Satoshi.

In: Signal Processing, Vol. 87, No. 11, 11.2007, p. 2649-2658.

Research output: Contribution to journalArticle

@article{6ac0f9c1e92f44a2af235e66b33230da,
title = "An MRF model-based approach to the detection of rectangular shape objects in color images",
abstract = "Rectangular shape object detection in color images is a critical step of many image recognition systems. However, there are few reports on this matter. In this paper, we proposed a hierarchical approach, which combines a global contour-based line segment detection algorithm and an Markov random field (MRF) model, to extract rectangular shape objects from real color images. Firstly, we use an elaborate edge detection algorithm to obtain image edge map and accurate edge pixel gradient information (magnitude and direction). Then line segments are extracted from the edge map and some neighboring parallel segments are merged into a single line segment. Finally all segments lying on the boundary of unknown rectangular shape objects are labeled via an MRF model built on line segments. Experimental results show that our method is robust in locating multiple rectangular shape objects simultaneously with respect to different size, orientation and color.",
keywords = "Edge detection, Line segment detection, MRF model, Randomized Hough transform, Rectangle detection",
author = "Yangxing Liu and Takeshi Ikenaga and Satoshi Goto",
year = "2007",
month = "11",
doi = "10.1016/j.sigpro.2007.04.018",
language = "English",
volume = "87",
pages = "2649--2658",
journal = "Signal Processing",
issn = "0165-1684",
publisher = "Elsevier",
number = "11",

}

TY - JOUR

T1 - An MRF model-based approach to the detection of rectangular shape objects in color images

AU - Liu, Yangxing

AU - Ikenaga, Takeshi

AU - Goto, Satoshi

PY - 2007/11

Y1 - 2007/11

N2 - Rectangular shape object detection in color images is a critical step of many image recognition systems. However, there are few reports on this matter. In this paper, we proposed a hierarchical approach, which combines a global contour-based line segment detection algorithm and an Markov random field (MRF) model, to extract rectangular shape objects from real color images. Firstly, we use an elaborate edge detection algorithm to obtain image edge map and accurate edge pixel gradient information (magnitude and direction). Then line segments are extracted from the edge map and some neighboring parallel segments are merged into a single line segment. Finally all segments lying on the boundary of unknown rectangular shape objects are labeled via an MRF model built on line segments. Experimental results show that our method is robust in locating multiple rectangular shape objects simultaneously with respect to different size, orientation and color.

AB - Rectangular shape object detection in color images is a critical step of many image recognition systems. However, there are few reports on this matter. In this paper, we proposed a hierarchical approach, which combines a global contour-based line segment detection algorithm and an Markov random field (MRF) model, to extract rectangular shape objects from real color images. Firstly, we use an elaborate edge detection algorithm to obtain image edge map and accurate edge pixel gradient information (magnitude and direction). Then line segments are extracted from the edge map and some neighboring parallel segments are merged into a single line segment. Finally all segments lying on the boundary of unknown rectangular shape objects are labeled via an MRF model built on line segments. Experimental results show that our method is robust in locating multiple rectangular shape objects simultaneously with respect to different size, orientation and color.

KW - Edge detection

KW - Line segment detection

KW - MRF model

KW - Randomized Hough transform

KW - Rectangle detection

UR - http://www.scopus.com/inward/record.url?scp=34447270250&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34447270250&partnerID=8YFLogxK

U2 - 10.1016/j.sigpro.2007.04.018

DO - 10.1016/j.sigpro.2007.04.018

M3 - Article

VL - 87

SP - 2649

EP - 2658

JO - Signal Processing

JF - Signal Processing

SN - 0165-1684

IS - 11

ER -