Evaluation of different satellite image enhancement techniques in separating of geological units

Lithological studies and geological unit's mapping are generally applicable to the many fields of natural resources management. Satellite remote sensing images have been widely used for separating of geological units which can generate accurate results, as well as time and cost saving. This research aims at comparing the performance of different enhancement techniques in order to separate four geological units in the study area of Taft watershed, Yazd. In this study, Landsat 8 satellite images (OLI sensor) were used, analyzed and classified. Geological maps were also up to dated using google earth images. A comparison of the results of the satellites classified images and base geological maps, indicates the Kappa coefficient and overall accuracy of the classified supervised maps of the ration of b7/b5 have the highest values of 0.78 and 86.7%, respectively. To verify the accuracy of the comparison results, QQ plots were used to compare the percentage of areas. Comparing the percentage of areas, the oblique angle of the b7/b5 bands ratio was lower (i.e. better) than the other different ratio combinations. Finally, the values of accuracy assessment for the reflection of lithological units in the observation and estimation range of supervised classification map of the ratio bands of the b7/b5 in the geological units Ks, gd, Kt-1, and Qal were 0.993, 0.980, 0.948, and 0.985, respectively, indicates that the b7/b5 band ratio separate four different geological units of the study area more precisely, and easily than the other classification methods.