عنوان مقاله [English]
Distinguishing the type of agricultural production usingfor optimal water resource management the Remote Sensing, despite the low availability of direct access to agricultural land, would significantly reduce costs in the agricultural area. However, agricultural classification is not very accurate due to the high similarity of different products using one image of multi-spectral. One of the ways to dominate this problem is to use the data time series of the satellite. The purpose of this study is to increase the precision of agricultural product separation by using time series data. In this study after the process of optical data, various vegetation indices, as well as albedo and surface temperature of the optical time series data were calculated and using the TIMESAT model, the key phonological parameters of the plant were extracted in part of the Miandoab plain. Given the availability of ground truth data and, vegetation status such as distribution and vegetation characteristics were examined. Then, using all of these features, a map of agricultural products was created using the support vector machine classification algorithm. The support vector machine classification algorithm is due to the high flexibility of this algorithm for different situations and purposes with a total accuracy of 92% and a kappa of 0.91 if the classification process involves the combination of bands, vegetation indices GNDVI and the ALBEDO, LST index, made the most careful distinction between crops.