Urban land use Change detection using image processing of Satellite Imagery based on neural net work (case study: Taabriz city)

Tabriz as the largest metropolis in the North West has seen faced significant change and growth in last three decades. In this study, by using satellite imagery such as IRS LISS III and ETM within the interval of 20 years and using the technique of comparing the classification (PCC) changes within the city of Tabriz was discussed. As the accuracy of extracted changes in this technique depend on the classification method, so an artificial neural network (ANN) was used for classification. Firstly, by using back propagation Neural Network was designed and then images classified and finally data were evaluated by PCC method. Results show that, built-up area in this period increased from 4707 to 8322 hectare and also road area 1416 to 31128 hectare. Also, vegetation area and barren land decreased about 937 and 4379 hectare respectively. The result of performance of back propagation Neural Network shows that extracted data in barren land, built-up and road have more accuracy than other land uses.