A comparative study of multiband mamdani fuzzy classification methods for west of Iraq satellite image

In our paper, performance of four fuzzy membership function generation methods was studied. These were studied in the context implementing Mamdani classification on a set satellite images for western Iraqi territory. The first method generate triangulate functions using mean, minimum (min) and maximum (max) histogram attribute values (AV), while peak standard deviation (STD) these AV used second. On other hand, third fourth methods, Gaussian are generated same mentioned second respectively. goal to type inference system (MF) each territory training data sets. A pixel-by-pixel comparison with traditional likelihood (ML) made sets comprising six bands imagery region taken by Landsat-5 satellite. Simulation results comparisons singled out that MFs together STD as best achiever similarity 83.16 percent band (3) area.