Prediction of Urban Area Expansion with Implementation of MLC, SAM and SVMs’ Classifiers Incorporating Artificial Neural Network Using Landsat Data

A reliable land cover (LC) map is essential for planners, as missing proper maps may deviate a project. This study focusing on classification and prediction using three well known classifiers remote sensing data. Maximum Likelihood classifier (MLC), Spectral Angle Mapper (SAM), Support Vector Machines (SVMs) algorithms are used the representatives parametric, non-parametric subpixel capable methods change detection of Urmia City (Iran) its suburbs. Landsat images 2000, 2010, 2020 have been to provide information. The results demonstrated 0.93–0.94 overall accuracies MLC SVMs’ algorithms, but it was around 0.79 SAM algorithm. performed slightly better than classifier. Cellular Automata Artificial neural network method predict changes. Overall accuracy higher others at about 0.94 accuracy, although, SVMs were more accurate large area segments. Land predicted 2030, which demonstrate city’s expansion from 5500 ha in 2000 9000 2030.