Analysis Accruing of Sentinel 2A Image’s Classification Methods Based on Object Base and Pixel Base in Flood Area Zoning of Taleqan River

Flood zonation mapping is one of the priorities for the soil and water management, which Remote Sensing (RS) capabilities are very applicable to this issue. The main objective of this research was study of accuracy of the Object oriented and Pixel based methods for flood zonation mapping in the Taleghan River basin. Therefore, the Sentinel 2A satellite image of the study area classified using supervised classification method based on the both Object oriented and Pixel based approaches, with the Random Forrest (RF) and Support Vector Machine (SVM) algorithms. The overall precision and Kapa Coefficient of the RF algorithm based on Object oriented method were 89 percent and 0.82 and based on Pixel based method 88.5 percent and 0.75 respectively. The results of the overall precision and Kapa Coefficient for the SVM algorithm based on Pixel based method were 84.5 percent and 0.73 and based on Object oriented method were 86 percent and 0.78 respectively. In order to verification of the generated flooding zone maps, the flooding zone maps resulted from the hydraulic flood zonation mapping method with 2,5,10, 25 and 100 year return periods have been used. The results shown that, the generated maps of the both RF and SVM algorithms based on object oriented method; have the most suitable overlaps with the all-return periods flooding zone maps