Department of Computer Engineering
S E M I N A R
Hierarchical Classification of Remotely Sensed Images
Huseyin Gokhan Akcay
Remote sensing is the science of acquiring information about the Earth's surface using satellites, aircraft etc.. Automatic content extraction and classification of remotely sensed images have become highly desired goals by the advances in satellite technology and computing power. The usual choice for the level of processing image data has been pixel-based analysis. However, spatial information is an important element to interpret the land cover because pixels alone do not give much information about image content. In this talk, we give an overview of the remote sensing image analysis techniques. We also present a hierarchical method for classification of remotely sensed imagery. We use pixel level data only for specific classes which are easily separable from others by their spectral features. However, some land cover/use structures (e.g. buildings and roads) cannot be detected by analyzing pixels but require structure and shape analysis. There are several approach es for extraction of these structures from the image data. However, most of the previous approaches try to solve the problem on specific images such as images of the same type of area and images where these structures are isolated. Our goal is to develop a generic model that can be applied to different types of images.
DATE: April 10, 2006, Monday@ 15:40
PLACE: EA 502