Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.395765
Title: Extraction of spatial information from stereoscopic SAR images
Author: Chen, Pu-Huai
ISNI:       0000 0001 3535 929X
Awarding Body: University of London
Current Institution: University College London (University of London)
Date of Award: 2000
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
Synthetic Aperture Radar (SAR) is now widely used for generating Digital Elevation Models (DEMs) and has advantages over optical data in terms of availability as it allows all-day and all-weather operations. The stereoscopic SAR method, which allows direct extraction of spatial information in three-dimensional space, has been established for decades. However, the traditional stereoscopic methods developed for SAR data depend on many human operations and need ground control points (GCPs), to set up geometric models. The aims of the thesis are not only to propose a refined rigorous stereoscopic SAR method and a new error model to predict theoretic errors, but also to achieve a higher level of automation and accuracy. By using a weighting matrix, which is derived by considering different observations in the space intersection algorithm, the minimal number of the GCPs required for the refined algorithm is only two. To achieve a high degree of automation, an optimized strategy of parameter selection for the pyramidal image correlation scheme employing a region-growing technique has been proposed. This avoids a trial-and-error approach to produce digital parallax data from the same-side SAR image pairs. A new method to derive GCPs automatically has been developed using a SAR image simulation technique, under the condition that a known DEM chip is available, to minimize human interventions and operator error. The proposed method for providing GCPs and the DEMs generated from space intersection have been incorporated into the procedures for geocoding SAR images to validate the proposed algorithms. The results derived show that the stereoscopic SAR data can be applied to geometric rectification in flat-to-moderate areas, and other applications of extraction of spatial information are promising.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.395765  DOI: Not available
Share: