Farzad Moradi - Dept. of RS & GIS, Science and Research Branch, Islamic Azad university, Yazd, Iran.
M.H. Mokhtari - Dept. of RS & GIS, Science and Research Branch, Islamic Azad university, Yazd, Iran.
Ali Ardakani - Dept. of RS & GIS, Science and Research Branch, Islamic Azad university, Yazd, Iran.

Remote sensing satellite imagery represent an important source of information for urban analysis. Gather information about Land Cover (LC) changes is fundamental for a better understanding the relationships and interactions between humans and the natural environment. Remote sensing (RS) data have been one of the most important data sources for studied of LC spatial and temporal changes. In fact, multi-temporal RS datasets, opportunely processed and elaborated, allow to map and identify landscape changes, giving an effective effort to sustainable landscape planning and management. In this paper, a change detection analysis has been done using Landsat TM satellite image with 11 years time interval. In preprocessing stage, geometric correction, image rectification, selection of study area, radiometric correction and spectral calibration to convert landsat TM digital numbers to radiance and exoatmospheric reflectance have been done. Then the resulting images have been imported to a change detection system. For this section at first each of two images classified with ARTIFICIAL NEURAL NETWORK methods. Then accuracy assessment have been done with more than 32points which obtained by GPS and overall accuracy and Kappa coefficient extracted about %68. Then two thematic maps in physical space have been obtained. Then change area and from-to classes have been provided by means of post classification method. This study conclude that the expansion of the city is about 32 percent compared to the first data image, and the city expansion is mainly due to the lands changed from bare land or vegetation around the city to built-up area.

change detection, landsat, neural network, remote sensing