آشكار سازي تغييرات پوشش اراضي شهر اراك با استفاده از RS و GIS

چكيده انگليسي :

Land cover change detection of Arak city using Remote Sensing GIS Mozhgan Ahmadi Nadoushan ahmadi@na iut ac ir Natural Resources Department Isfahan university of technology Isfahan 8415 9311 Iran Degree M Sc Language Farsi Dr Alireza Soffianian Soffianian@cc iut ac irAbstractLand cover changes are among the most important alterations of the earth s land surface which affectenvironment and environmental processes Nowadays with regard to population growth and peopletendency to live in urban centers most of the land cover changes take place in urban areas Urbangrowth in population and spatial extent has occurred during recent decades Changes in urban extentsettings may have some environmental impacts in local regional national and international scales Having access to up to date and accurate information about land cover and its changes especially inurban areas is important for decision making and urban planning Besides detecting changes occurredin an area in the past predicting and modeling future changes is important for recognizing quality andquantity of changes Determining past changes will help planners and managers to compensatedamages and losses and not to repeat their mistakes Future change prediction and modeling is criticalfor long term planning and preventing some undesirable environmental changes Post classificationcomparison is one of the most commonly used methods which generates information on the nature andarea of changes At the present situation using tools which are capable for environmental changedetection assessment and modeling is essential Remote sensing and GIS methods have high potentialsfor detection evaluation and modeling environmental changes The objective of the present study is touse remote sensing and GIS techniques in conjunction with modeling method in order to detect andpredict land cover changes of Arak metropolitan area and its periphery Arak is one of the cities which has undergone fast urban expansion during recent decades due toindustrialization and population immigration This expansion has some influences on Arak speriphery In this study for land cover change detection and evaluation of Arak during 1956 2006 aerial photos from 1956 and 1972 Landsat TM and ETM and IRS P6 LISS III images acquired in1980 1990 and 2006 and post classification method were used At first in order to prepare aerialphotos and satellite images geometric correction was done For this purpose topographic maps atscale of 1 50000 and 1 25000 and nearest neighbor method resampling method were applied Rootmean square error for all aerial photos and satellite images were less than one pixel After that allgeorefrenced photos were mosaiced and land cover maps with 4 classes urban areas vegetated areas barren lands and rocks were generated using visual interpretation of aerial photos After satelliteimages geometric correction topographic correction was applied to images using DEM and Lambertmodel In next step after producing false color composite images and image fusion supervised unsupervised artificial neural networks and V I S classification method were implemented Finally considering their good results artificial neural network and V I S classification were used forproducing final land cover maps Land cover maps in four classes with more than 90 percent overallaccuracy were generated For detecting land cover changes during 6 periods between 1956 2006 landcover maps of 1956 1972 1980 1990 and 2006 were compared and change maps and tables weremade The results showed significant urban expansion vegetated and barren lands losses and stabilityin rocks and mountainous areas during 1956 2006 In order to change prediction CA markov andmarkov chain models were applied and land cover map of 1404 was simulated For determining modelvalidation in land cover change prediction real and simulated maps of 1980 and 2006 were compared Results showed that models which were used in this study have good predictive power in urban areas Keyword

احمدي ندوشن، مژگان

آشكار سازي تغييرات پوشش اراضي شهر اراك با استفاده از RS و GIS

شبكه عصبي مصنوعي , V - I - S , مدل ماركوف