Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract. Sentinel 2 is the newest satellite that has a high spatial resolution and a 10 day equator with one satellite, and 5 day equator with 2 satellites. Sentinel is one of the data providers in Landsat and Modis satellite group and working in a time interval between these two satellites. The identification and exploration of vegetation indices on this satellite seems to be necessary. The aim of this study was to compare sentinel and Landsat satellite imagery in date palm identification. In the study area, three main types of land cover (alfalfa, date palm, and soil) were identified to compare in the images of these two satellites. To evaluate these aerial imageries based on previous literatures, four supervised classification methods of Neural Network (NN), Maximum Likelihood Classifier (MLC), Support Vector Machines (SVM) and Mahalanobis Distance Classifier (MDC) were used. The classification results showed that the overall accuracy of the classification in sentinel images by using NN, MLC, SVM and MDC were 99.57% (Kapa 0.99), 98.47% (Kappa 0.97), 99.63% (Kappa 0.99) and 98.37% (Kappa 0.97) respectively. For Landsat images by using NN, MLC, SVM and MDC the overall accuracy and kappa were 98.86% (Kappa 0.98), 93.37% (Kappa 0.87), 99.14% (Kappa 0.98) and 97.33% (Kappa 0.94), respectively. The results of satellite imagery comparison showed that Sentinel images with an average overall accuracy of 99.63% and Landsat with 99.14% can identify the areas under date palm cultivation. Both satellites were also able to classify other land covers satisfactorily.