A new algorithm for texture segmentation based on edge detection

Yu Xiaohan, Juha Ylä-Jääski, Yuan Baozong

Research output: Contribution to journalArticleScientificpeer-review

14 Citations (Scopus)

Abstract

A new unsupervised algorithm for texture segmentation is proposed in this paper. The new scheme is based on the idea that texture features change abruptly near boundaries between different textures, and the segmentation can be carried out by detecting the feature changes or so-called feature edges. In this algorithm, the image is first projected onto a hyperplane called the characteristic image, in which the value of each pixel is not a grey level but a vector value of the local textural features. An edge detection algorithm is then extended to the vector space and applied to the hyperplane to detect the feature edges. In this way, textural boundaries in the image are detected. The validity of our algorithm is also confirmed by experimental results.

Original languageEnglish
Pages (from-to)1105 - 1112
Number of pages8
JournalPattern Recognition
Volume24
Issue number11
DOIs
Publication statusPublished - 1991
MoE publication typeA1 Journal article-refereed

Fingerprint

Edge detection
Textures
Vector spaces
Pixels

Cite this

Xiaohan, Yu ; Ylä-Jääski, Juha ; Baozong, Yuan. / A new algorithm for texture segmentation based on edge detection. In: Pattern Recognition. 1991 ; Vol. 24, No. 11. pp. 1105 - 1112.
@article{2375f7b2681c49bebcee4e0015c1f8e2,
title = "A new algorithm for texture segmentation based on edge detection",
abstract = "A new unsupervised algorithm for texture segmentation is proposed in this paper. The new scheme is based on the idea that texture features change abruptly near boundaries between different textures, and the segmentation can be carried out by detecting the feature changes or so-called feature edges. In this algorithm, the image is first projected onto a hyperplane called the characteristic image, in which the value of each pixel is not a grey level but a vector value of the local textural features. An edge detection algorithm is then extended to the vector space and applied to the hyperplane to detect the feature edges. In this way, textural boundaries in the image are detected. The validity of our algorithm is also confirmed by experimental results.",
author = "Yu Xiaohan and Juha Yl{\"a}-J{\"a}{\"a}ski and Yuan Baozong",
note = "Project code: GRAT435",
year = "1991",
doi = "10.1016/0031-3203(91)90125-O",
language = "English",
volume = "24",
pages = "1105 -- 1112",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier",
number = "11",

}

Xiaohan, Y, Ylä-Jääski, J & Baozong, Y 1991, 'A new algorithm for texture segmentation based on edge detection', Pattern Recognition, vol. 24, no. 11, pp. 1105 - 1112. https://doi.org/10.1016/0031-3203(91)90125-O

A new algorithm for texture segmentation based on edge detection. / Xiaohan, Yu; Ylä-Jääski, Juha; Baozong, Yuan.

In: Pattern Recognition, Vol. 24, No. 11, 1991, p. 1105 - 1112.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - A new algorithm for texture segmentation based on edge detection

AU - Xiaohan, Yu

AU - Ylä-Jääski, Juha

AU - Baozong, Yuan

N1 - Project code: GRAT435

PY - 1991

Y1 - 1991

N2 - A new unsupervised algorithm for texture segmentation is proposed in this paper. The new scheme is based on the idea that texture features change abruptly near boundaries between different textures, and the segmentation can be carried out by detecting the feature changes or so-called feature edges. In this algorithm, the image is first projected onto a hyperplane called the characteristic image, in which the value of each pixel is not a grey level but a vector value of the local textural features. An edge detection algorithm is then extended to the vector space and applied to the hyperplane to detect the feature edges. In this way, textural boundaries in the image are detected. The validity of our algorithm is also confirmed by experimental results.

AB - A new unsupervised algorithm for texture segmentation is proposed in this paper. The new scheme is based on the idea that texture features change abruptly near boundaries between different textures, and the segmentation can be carried out by detecting the feature changes or so-called feature edges. In this algorithm, the image is first projected onto a hyperplane called the characteristic image, in which the value of each pixel is not a grey level but a vector value of the local textural features. An edge detection algorithm is then extended to the vector space and applied to the hyperplane to detect the feature edges. In this way, textural boundaries in the image are detected. The validity of our algorithm is also confirmed by experimental results.

U2 - 10.1016/0031-3203(91)90125-O

DO - 10.1016/0031-3203(91)90125-O

M3 - Article

VL - 24

SP - 1105

EP - 1112

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 11

ER -