Monitoring Urban Devlopment Using Satellite Images

Keeping track of urban development is an important task carried out by various municipal agencies. As resources are in general limited, priorities must be set as to where and when updates of data bases should take place. In this paper the results of a linear unmixing procedure, applied to two Landsat TM images of the City of Vienna from 1996 and 1999 will be analysed to highlight where building activities have taken place. In order to verify the results aerial photographs from 1996 and 1999 were available for three parts of the image. Each photograph covers an area of 5 x 5 km. In addition an Ikonos panchromatic image from 2000, covering one of the validation areas, will be used. Changes that were visible in the high-resolution imagery were highlighted and compared to a change map produced on the basis of the results of a linear unmixing procedure. In total more than 75 % of all changes identified in the high resolution imagery could be identified with the fraction images. Linear unmixing for highlighting where changes are most likely to have occurred can be used in a number of ways. One is to give an overview over large areas to show were most changes have taken place. The other is an in depth analysis of smaller subsets as to where these changes have actually occurred. It is a procedure that can be carried in shorter intervals than high-resolution campaigns are normally possible thus providing more up-to-date information.