1 AIT Asian Institute of Technology

Optimizing the performance of an image processing algorithm using the PVM system

AuthorPrapa Lertsirisathaporn
Call NumberAIT Thesis no. CS-96-33
Subject(s)Image processing
Parallel processing (Electronic computers)
NoteA thesis submitted in the partial fulfillment of the requirements for the degree of Master of Science.
PublisherAsian Institute of Technology
Series StatementThesis ; no. CS-96-33
AbstractParallel processing is one of the most prom1smg solutions to the computing requirements of the future. A wide variety of applications demonstrate that solving problems with parallel algorithms is a significant approach for achieving faster processing. This is especially true in fields involved with intensive computation such as image processing. In this study, a two part method to enhance the speed of the image matching algorithm was developed and employed. First a parallel algorithm employing the divide-and-conquer technique was developed. Secondly, some forms of load balancing were considered in order to produce various task scheduling algorithms for distributing the tasks to processors. From implementation using the PVM (Parallel Virtual Machine) as an environment, it was found that faster processing depended on both the computational power of the machines forming the system and the policy for managing the system workload.
Year1996
Corresponding Series Added EntryAsian Institute of Technology. Thesis ; no. CS-96-33
TypeThesis
SchoolSchool of Engineering and Technology (SET)
DepartmentDepartment of Information and Communications Technologies (DICT)
Academic Program/FoSComputer Science (CS)
Chairperson(s)Yulu, Qi
Examination Committee(s)Stueart, Robert D.;Batanov, D. N.
Scholarship Donor(s)His Majesty King Bhumipol Adulyadej
DegreeThesis (M.Sc.) - Asian Institute of Technology, 1996


Usage Metrics
View Detail0
Read PDF0
Download PDF0