Extraction and evaluation of land use changes by using of the accuracy of SVM algorithm with polynomial kernel and maximum likelihood method in Ojan Chay basin

Land is one of the inputs of the production sector. It plays an important role not only in the agricultural economy and natural resources but also in the economy of the whole country and it is necessary to pay attention to the land and the changes that have taken place in it. The present study was conducted to investigate the trend of land use change in Ojan Chay catchment using remote sensing and GIS. Multi-timer images of 1987 TM, ETM + 2002, and ETM + 2015 were used in this study, and land use maps were prepared based on digital processing of maximum probability and backup vector machine. The type of land use changes in the region was obtained. According to the results, the SVM method is more efficient for estimating land use changes in the study area so that the extracted results have a higher percentage of accuracy and kappa coefficient. From the results of evaluations, it can be deduced that the trend of land use change in some land uses such as agricultural lands and pastures is high. Agricultural lands have fluctuated from 33% to 37% over a period of 28 years and rangelands have decreased from 51% to 49%, with these two uses having the most changes. Finally, due to the fact that most of the changes in rangeland change have been related to agricultural lands and vice versa, therefore, the need to focus land management and improvement activities on this type of land use increases.