Satellite image classification using proposed singular value decomposition method

Main Article Content

Noor Zubair Kouder

Abstract

In this work, satellite images for Razaza Lake and the surrounding area
district in Karbala province are classified for years 1990,1999 and
2014 using two software programming (MATLAB 7.12 and ERDAS
imagine 2014). Proposed unsupervised and supervised method of
classification using MATLAB software have been used; these are
mean value and Singular Value Decomposition respectively. While
unsupervised (K-Means) and supervised (Maximum likelihood
Classifier) method are utilized using ERDAS imagine, in order to get
most accurate results and then compare these results of each method
and calculate the changes that taken place in years 1999 and 2014;
comparing with 1990. The results from classification indicated that
water and hills are decreased, while vegetation, wet land and barren
land are increased for years 1999 and 2014; comparable with 1990.
The classification accuracy was done by number of random points
chosen on the study area in the field work and geographical data then
compared with the classification results, the classification accuracy for
the proposed SVD method are 92.5%, 84.5% and 90% for years
1990,1999,2014, respectivety, while the classification accuracies for
unsupervised classification method based mean value are 92%, 87%
and 91% for years 1990,1999,2014 respectivety.

Article Details

How to Cite
1.
Satellite image classification using proposed singular value decomposition method. IJP [Internet]. 2019 Feb. 4 [cited 2024 Mar. 28];13(28):52-67. Available from: https://ijp.uobaghdad.edu.iq/index.php/physics/article/view/243
Section
Articles

How to Cite

1.
Satellite image classification using proposed singular value decomposition method. IJP [Internet]. 2019 Feb. 4 [cited 2024 Mar. 28];13(28):52-67. Available from: https://ijp.uobaghdad.edu.iq/index.php/physics/article/view/243

Similar Articles

You may also start an advanced similarity search for this article.