Use this URL to cite or link to this record in EThOS:
Title: A knowledge-based three-dimensional modelling system
Author: Tung, D-K.
Awarding Body: University College of Swansea
Current Institution: Swansea University
Date of Award: 1994
Availability of Full Text:
Access from EThOS:
As manufacturers strive towards high-quality production, automated industrial inspection is a potent resource in the design of cost-effective systems which can ensure that products meet all their design specifications. However, in reality, only in well-controlled applications are economically usable systems being taken into daily use. Where such systems are being used, they are seen to be primarily addressing 2-dimensional inspection problems. This is not surprising, given the highly complex problems which must be dealt with in practical, real-world environments. However, there is an urgent need to move towards acceptable machine-vision systems which not only can operate in industrial environments, but also offer the benefits of 3-dimensional visual representation - so vital in any real inspection situation. A fundamental aspect of any inspection system is the development of inspection models - to be used in subsequent inspection procedures. The generation of these models is a non-trivial task, and one which is increasingly being seen to be best done using operator assistance - as shown, for example, in the work of Chen [34]. However, most current work in such model generation has been tackled in the 2-D arena. This thesis addresses the problem of providing high-quality, visually meaningful, representations of 3-dimensional bodies, drawing information from 2 simple, but industrially-rugged, 2-dimensional images, and using operator assistance to determine the final models. When combined, the resulting 3-dimensional representation provides a valuable reference to an object's total physical structure. The models themselves allow for accurate inspection of the objects' physical parameters.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available