Posterior probability-based optimization of texture window size for image classification

Jinxiu Liu, Huiping Liu, Janne Heiskanen, Matti Mõttus, Petri Pellikka

Research output: Contribution to journalArticleScientificpeer-review

3 Citations (Scopus)

Abstract

Texture provides spatial features complementary to spectral information in land cover classification of high spatial resolution imagery. In texture classification, window size is an important factor influencing classification accuracy, but selecting the optimal window size is difficult. In this paper, we propose an optimized window size texture classification method which can solve the window size selection problem. In order to validate the new method, we designed four classification experiments with different input features based on SPOT-5 imagery: (1) spectral features, (2) spectral features and single window size texture features, (3) spectral features and multiple window size texture features and (4) spectral features and optimized window size texture features based on posterior probabilities. Overall, the highest accuracy was obtained using the optimized window size texture classification, which does not require window size selection before classification. Furthermore, the results imply that optimized window size varies with land cover type.

Original languageEnglish
Pages (from-to)753-762
Number of pages10
JournalRemote Sensing Letters
Volume5
Issue number8
DOIs
Publication statusPublished - 3 Aug 2014
MoE publication typeA1 Journal article-refereed

Fingerprint

Image classification
image classification
Textures
texture
size selection
land cover
imagery
SPOT
spatial resolution

Cite this

Liu, Jinxiu ; Liu, Huiping ; Heiskanen, Janne ; Mõttus, Matti ; Pellikka, Petri. / Posterior probability-based optimization of texture window size for image classification. In: Remote Sensing Letters. 2014 ; Vol. 5, No. 8. pp. 753-762.
@article{d94a3c9d5bbb4b09b643222bcfd2b1ed,
title = "Posterior probability-based optimization of texture window size for image classification",
abstract = "Texture provides spatial features complementary to spectral information in land cover classification of high spatial resolution imagery. In texture classification, window size is an important factor influencing classification accuracy, but selecting the optimal window size is difficult. In this paper, we propose an optimized window size texture classification method which can solve the window size selection problem. In order to validate the new method, we designed four classification experiments with different input features based on SPOT-5 imagery: (1) spectral features, (2) spectral features and single window size texture features, (3) spectral features and multiple window size texture features and (4) spectral features and optimized window size texture features based on posterior probabilities. Overall, the highest accuracy was obtained using the optimized window size texture classification, which does not require window size selection before classification. Furthermore, the results imply that optimized window size varies with land cover type.",
author = "Jinxiu Liu and Huiping Liu and Janne Heiskanen and Matti M{\~o}ttus and Petri Pellikka",
year = "2014",
month = "8",
day = "3",
doi = "10.1080/2150704X.2014.963895",
language = "English",
volume = "5",
pages = "753--762",
journal = "Remote Sensing Letters",
issn = "2150-704X",
publisher = "Taylor & Francis",
number = "8",

}

Posterior probability-based optimization of texture window size for image classification. / Liu, Jinxiu; Liu, Huiping; Heiskanen, Janne; Mõttus, Matti; Pellikka, Petri.

In: Remote Sensing Letters, Vol. 5, No. 8, 03.08.2014, p. 753-762.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Posterior probability-based optimization of texture window size for image classification

AU - Liu, Jinxiu

AU - Liu, Huiping

AU - Heiskanen, Janne

AU - Mõttus, Matti

AU - Pellikka, Petri

PY - 2014/8/3

Y1 - 2014/8/3

N2 - Texture provides spatial features complementary to spectral information in land cover classification of high spatial resolution imagery. In texture classification, window size is an important factor influencing classification accuracy, but selecting the optimal window size is difficult. In this paper, we propose an optimized window size texture classification method which can solve the window size selection problem. In order to validate the new method, we designed four classification experiments with different input features based on SPOT-5 imagery: (1) spectral features, (2) spectral features and single window size texture features, (3) spectral features and multiple window size texture features and (4) spectral features and optimized window size texture features based on posterior probabilities. Overall, the highest accuracy was obtained using the optimized window size texture classification, which does not require window size selection before classification. Furthermore, the results imply that optimized window size varies with land cover type.

AB - Texture provides spatial features complementary to spectral information in land cover classification of high spatial resolution imagery. In texture classification, window size is an important factor influencing classification accuracy, but selecting the optimal window size is difficult. In this paper, we propose an optimized window size texture classification method which can solve the window size selection problem. In order to validate the new method, we designed four classification experiments with different input features based on SPOT-5 imagery: (1) spectral features, (2) spectral features and single window size texture features, (3) spectral features and multiple window size texture features and (4) spectral features and optimized window size texture features based on posterior probabilities. Overall, the highest accuracy was obtained using the optimized window size texture classification, which does not require window size selection before classification. Furthermore, the results imply that optimized window size varies with land cover type.

UR - http://www.scopus.com/inward/record.url?scp=84907879664&partnerID=8YFLogxK

U2 - 10.1080/2150704X.2014.963895

DO - 10.1080/2150704X.2014.963895

M3 - Article

VL - 5

SP - 753

EP - 762

JO - Remote Sensing Letters

JF - Remote Sensing Letters

SN - 2150-704X

IS - 8

ER -