Mohammad Karim Sirat; Gomroki Masoomeh
Abstract
Today, land use change detection in urban areas have key role in city management, resource management and city development. So, in this research the land use change detection is investigated. The purpose of this research is to investigate the land use changes in Herat using Landsat 8 satellite images ...
Read More
Today, land use change detection in urban areas have key role in city management, resource management and city development. So, in this research the land use change detection is investigated. The purpose of this research is to investigate the land use changes in Herat using Landsat 8 satellite images of OLI sensor during 2015 and 2022. After geometrical, radiometric and atmospheric correction in these four-use satellite images; Soil, plant, city and water were identified in the study area. Two methods of maximum likelihood and artificial neural network have been used to classify satellite images to identify land use changes. In general, the classification of the images used by the maximum likelihood method has provided better accuracy, which has a good kappa coefficient and overall accuracy. In the image of 2015, the maximum likelihood method with a kappa coefficient of 0.75, overall accuracy of 0.85 was used, and in the image of 2022, classification was done using the maximum likelihood method. which has a kappa coefficient of 0.96 and an overall accuracy of 0.97. Based on the results of the classification, during the period of 2015-2022, a decrease in the level of land and water and an increase in the level of city and plant land were observed. After classifying the number of changes in the time period of 2015-2022, the land use area has decreased by 4.00 square kilometers, water by 1.62 square kilometers, and the city has increased by 1.39 square kilometers and plants by 4.59 square kilometers.
