Classification of Satellite Images in the Evaluation of Urban Land Use Change Using Scale Optimization in Objected Oriented Processing (Case Study: Ardabil City)

The purpose of the study was to compare the efficiency of the seven commonly used methods of satellite-controlled monitoring of land use changes in the evaluation of land use changes using TM and OLS Landsat, and IRS, and Spot 5 and Quick Birds bands, and different color combinations of this Images are intended for exploitation of agricultural land, residential areas and aquatic areas using object-oriented processing. Digital processing of satellite images was carried out in 1998 and 2016 using advanced methods. Educational examples in five user classes by eCognition software using segmentation scale optimization using different color combinations and coefficients of shape and compression, an appropriate scale for segmentation for arable land, scale 50, for human complications 8 and finally for aquatic areas 10 as appropriate scales. Then each image was classified separately using seven methods and extracted samples and the efficiency of each classification method was calculated by calculating the two general health and Kappa coefficients. The results indicate the accuracy of each classification method, which method of classification of the neural network with a total accuracy of 475/94 and Kappa coefficient 925/92 as the most accurate method among class methods Fetch selected. These results show that the sampling of educational samples with the proper precision of the classes in the images and the probability of belonging to each of the pixels of satellite images to these classes can well be classified in the group Available in the helpful area.