object-based classification methods

Remote sensing technology is one of the efficient and innovative technologies for agricultural land use/cover mapping. In this regard, the object-based Image Analysis (OBIA) is new method of satellite image processing which integrates spatial and spectral information in satellite image process domain. This approach make use of spectral, environmental, physical and geometrical characteristics (texture, shape) context of images for modeling of land use/cover classes. Current study, aims to classify micro land use/cover of Meyandoab County by applying appropriate algorithms and parameters in the object based approach. For this goal, Quick Bird and Aster satellite images were used for processing and land use modeling. Accordingly, land use map was classified in 9 class based on spectral and spatial characteristic. The segmentation was performed in the scale of 10, shape parameter of 0.7 as well as the compactness of 0.3. In order to apply classification, fuzzy based algorithm and operators (AND, OR) was applied to detriment the membership functionality of segments for each class as well as classifying the related objects. We also applied textures, geometric, NDVI, GLCM, Braitnese algortims based on fuzzy operators. In order to validate results, the accuracy assessment step was performed and the finally overall accuracy of 93.6 was obtained for derived map. The Kappa coefficient was also detriment to be 0.92. The area under cultivation included respectively for lands of wheat and barley, prunes and plums, apples, vineyards and alfalfa hay2622.42, 4505, 4354.55, 4457.85, 14110.58 hectares.