GEOMETRIC MODELING AND RECTIFICATION OF SATELLITE SCANNER IMAGERY AND INVESTIGATION OF RELATED CRITICAL ISSUES (LANDSAT)
Abstract
Acquisition of timely information about the surface of the earth is now feasible using satellites carrying passive sensors such as electro optical scanners. Imageries acquired by these scanners need to be transformed or rectified to the earth's surface to facilitate the extraction and storage of information contained in these imageries. This requires a thorough understanding of the geometry of the relevant image acquisition process. A comprehensive model describing the geometry of satellite scanner imageries which is suitable for both image rectification and simulation is derived. The model is then validated using real data, and once validated, it is used as a powerful tool for understanding, through simulation studies, the geometric process involved in scanner imaging. A comparative study of different models and their variations is performed using simulated data to determine the best model for different conditions. The results showed that the derived model is well-suited for high accuracy applications. Once the imagery and the model to be used for analysis are chosen, one major remaining factor affecting rectification accuracy is concerned with control data characteristics which is investigated next. Alternatives to control points such as edges and tie points (i.e. points in block adjustment) are explored. Other related topics such as the major rectification/registration sequence and blunder analysis are also addressed.
Degree
Ph.D.
Subject Area
Civil engineering|Remote sensing
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.