Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.650054
Title: An evolvable hardware system for automatic optical inspection
Author: Evans, Jonathan
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 2004
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
This thesis investigates the use of System-On-Chip technology as the basis for a low cost Automatic Optical Inspection system. Novel object recognition and image registration algorithms are developed and targeted for a System-On-Chip platform. Execution time analysis of the novel algorithms is performed. This analysis shows that the application's performance criteria can be met through enhancing the platform using a Digital Signal Processor. Fast and accurate object recognition is an important class of algorithms for inspection systems. This thesis shows that techniques based on the Hough Transform are too computationally expensive for the recognition of Integrated Circuits. The thesis presents a novel low complexity technique based on Region Growing which is robust and efficient for complex images. Effective and efficient image registration is a fundamental task in inspection systems. This thesis presents a new approach to detecting placement errors of Integrated Circuits. The approach is based on the use of a Genetic Algorithm to derive transformations for matching a captured image to a reference image. This thesis presents execution time studies of the novel algorithms on an enhanced target platform. The image registration system is partitioned across the enhanced platform. The Digital Signal Processor executes the fitness function of the Genetic Algorithm. The rest of the algorithm runs on the System-On-Chip platform. The techniques developed in this work can be used to detect further types of faults on test samples such as placement errors of resistors and capacitors. The work therefore forms the basis of a high functionality low cost Automatic Optical Inspection system.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.650054  DOI: Not available
Share: