Analysis of Land Use/ Land Cover Changes in Jalalabad City Using Landsat 8 Imagery and Comparison of Maximum Likelihood and Support Vector Machine Classification Methods in 2015 and 2023

تحلیل تغییرات با استفاده از تصاویر ماهواره ای

Authors

  • Mohammad Karim Sirat Bamyan University https://orcid.org/0009-0000-8463-1042
  • Mohammad Nasir Haidari Graduate of Department of Engineering Geodesy, Faculty of Geosciences, Bamyan University
  • Sayed Jawad Tamim Graduate of Department of Engineering Geodesy, Faculty of Geosciences, Bamyan University
  • Mohammad Jawad Madadi Department of Engineering Geodesy, Faculty of Geosciences, Bamyan University

DOI:

https://doi.org/10.64226/sarj.v3i02.176

Keywords:

Land Use/Land Cover, Image Classification, Remote Sensing, maximum Likelihood Classification, Support Vector Machine, Accuracy Assessment, Change Detection

Abstract

Urban growth and changes in human activities can influence the land use and land cover (LULC) pattern of Jalalabad City; therefore, monitoring and quantifying these changes are important for sustainable urban planning and management. This study aimed to analyze LULC changes from 2015 to 2023 using Landsat 8 imagery and two supervised classification methods: Maximum Likelihood Classification (MLC) and Support Vector Machine (SVM). The study area comprised the urban area of Jalalabad City, and Landsat 8 images from 2015 and 2023 were used to extract and compare LULC classes. Following image preprocessing, four land-use/land-cover classes, including vegetation, bare soil, urban areas, and water bodies, were defined and classified. Training and validation samples were independently prepared and used to evaluate the performance of the two classification algorithms. The accuracy assessment showed that MLC achieved an overall accuracy of 96.5% and a Kappa coefficient of 0.95 in 2015, and an overall accuracy of 98.0% and a Kappa coefficient of 0.96 in 2023, whereas SVM achieved an overall accuracy of 95.6% and a Kappa coefficient of 0.94 in 2015, and an overall accuracy of 97.0% and a Kappa coefficient of 0.95 in 2023. Based on the MLC results, urban areas increased from 20.27 to 21.19 km², and vegetation cover increased from 1.45 to 3.86 km², whereas bare-soil areas decreased from 7.14 to 3.83 km², and water bodies decreased from 0.07 to 0.043 km². Overall, MLC showed slightly higher performance than SVM in both years, although the differences in classification accuracy were relatively small. The observed LULC changes indicate a decrease in bare-soil areas and an increase in urban areas and vegetation cover within the study area. These findings can support the monitoring of spatial changes and contribute to sustainable urban planning and management in Jalalabad City.

Downloads

Published

2026-10-19

How to Cite

Sirat, M. K., Haidari, M. N., Tamim, S. J., & Madadi, M. J. (2026). Analysis of Land Use/ Land Cover Changes in Jalalabad City Using Landsat 8 Imagery and Comparison of Maximum Likelihood and Support Vector Machine Classification Methods in 2015 and 2023: تحلیل تغییرات با استفاده از تصاویر ماهواره ای . Samangan Academic & Research Journal, 3(02), 47–66. https://doi.org/10.64226/sarj.v3i02.176

Issue

Section

Articles

Similar Articles

<< < 1 2 

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