The Effects of Haze on the Accuracy of Maximum Likelihood Classification

Asmala, A. and Shaun, Quegan (2016) The Effects of Haze on the Accuracy of Maximum Likelihood Classification. Applied Mathematical Sciences, 10 (39). pp. 1935-1944. ISSN 1312-885X

Text ahmadAMS37-40-2016 effects of haze on accuracy of ML published.pdf Download (556kB)

Abstract

This study aims to investigate the effects of haze on the accuracy of Maximum Likelihood classification. Data containing eleven land covers recorded from Landsat 5 TM satellite were used. Two ways of selecting training pixels were considered which are choosing from the haze-affected and haze-free data. The accuracy of Maximum Likelihood classification was computed based on confusion matrices where the accuracy of the individual classes and the overall accuracy were determined. The result of the study shows that classification accuracies declines with faster rate as visibility gets poorer when using training pixels from clear compared to hazy data.

Item Type:	Article
Uncontrolled Keywords:	Haze, Landsat, Classification Accuracy, Training Pixels
Subjects:	Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science T Technology > T Technology (General)
Divisions:	Faculty of Information and Communication Technology > Department of Industrial Computing
Depositing User:	Dr. Asmala Ahmad
Date Deposited:	30 Jun 2016 02:06
Last Modified:	05 Sep 2021 16:32
URI:	http://eprints.utem.edu.my/id/eprint/16729
Statistic Details:	View Download Statistic

Actions (login required)

View Item