Eavaluation Landuse Changes Using neural network classifiers and a support vector machine

Document Type : Original Article

Author

Education Expert ، DEPARTMENT OF MATEMATIC ، TABRIZ UNIVESITITY ،TABRIZ ، IRAN

Abstract

: Over time, patterns of land cover and land use change and subsequent changes are fundamental and human factor plays a most important role in this process. Ever, scientists have attempted to identify factors that cause land use changes and their impact on the environment. Therefore, in previous decades, researchers have different views collected from the field, as well as aerial photographs to detect land use changes resulting from the imposition of natural and human processes have been analyzed. Today, however, based on technological advances made in the field of remote sensing, satellite imagery can be used to more accurately evaluate the environmental changes during the process and the final results of the illustrated model. The main purpose of the ongoing monitoring of land use changes in river basins liqvan is 1985-2006-2013. Accordingly, to explore the changes occurring in the study area, Landsat TM and ETM + Landsat images of the years 1985-2006-2013 were analyzed. Accordingly, after applying atmospheric and geometric correction, image enhancement operations performed using the maximum likelihood method of supervised classification algorithms similar actions and thematic maps of land use of the basin has been designed to liqvan. In general, the overall accuracy of SVM method calculated in 1985 (96.20) and in 2006 (96.26) and in 2013 (99.64) by Mdkh highest accuracy than other methods. Finally moorland in the first place and then irrigated gardens and residential areas in the study area are eventually.

Keywords

Main Subjects

Volume 2, Issue 3 - Serial Number 3
September 2022
Pages 82-69
  • Receive Date: 31 July 2022
  • Revise Date: 23 October 2022
  • Accept Date: 06 September 2022