File Entry: The Evaluation of the Usage of the Fuzzy Algorithms in Increasing the Accuracy of the Extracted Land Use Maps

Abstract:

This study aimed to evaluate the accuracy of object-oriented classification with and without the use of fuzzy algorithms, compared with pixel based algorithm in land use/ land cover classification. In order to compare different methods, AVNIR2 sensor images obtained from ALOS satellite were used to classify land use in Maragheh County. The results indicated that object-oriented classification approach produced more accurate results than pixel-based, So that the fuzzy knowledge based method yielded the highest overall accuracy of 93.28 percent as compared to the object-oriented method without the use of the fuzzy algorithms (88.06 percent accuracy) and pixel based algorithm with 83.79 percent accuracy. Based on the results of this research, it is recommended that the researchers use higher spatial resolution images with the proper algorithms in the extraction of features of land use classes.

Keywords: remote sensing, object-oriented classification methods, pixel base, fuzzy algorithms, land use map