Monday, 15 August 2011

Wseas Transactions

New Subscription to Wseas Transactions

The following information was submitted:

Transactions: INTERNATIONAL JOURNAL of COMPUTERS AND COMMUNICATIONS
Transactions ID Number: 17-303
Full Name: Siti arpah Ahmad
Position: Ph.D. Candidate
Age: ON
Sex: Female
Address: Faculty of Computer and Mathematical Science, Universiti Teknologi MARA, Shah Alam, 40450, Malaysia
Country: MALAYSIA
Tel: 55211103
Tel prefix: 603
Fax: 5534 5077
E-mail address: arpah@tmsk.uitm.edu.my
Other E-mails: elaiza@tmsk.uitm.edu.my, noorhayati@tmsk.uitm.edu.my
Title of the Paper: Analysis of Parallel Multicore Performance on Application of Sobel Edge Detector
Authors as they appear in the Paper: N.e.a.Khalid, S.a.Ahmad, N.m.Noor, A.f.a.Fadzil and M.n.Taib
Email addresses of all the authors: arpah@tmsk.uitm.edu.my,elaiza@tmsk.uitm.edu.my,noorhayati@tmsk.uitm.edu.my,ahmadfirdausfadzil@gmail.com ,dr.nasir@ieee.org
Number of paper pages: 9
Abstract: This paper presents the parallel multicore Sobel edge algorithm which parallelizes the traditional sequential Sobel edge detection algorithm on a parallel multicore platform. Dealing with images, the algorithm inherits repetitive instructions that depend on the image size, thus may slow down the processing speed. The multicore architecture is a ready available resource on ordinary personal computer but often not fully utilized to its utmost potential. Align with this hidden opportunity, Sobel edge algorithm can be implemented on parallel programming paradigm by focusing on the thread operations. This work presents the parallel multicore Sobel edge algorithm which parallelizes the traditional sequential Sobel edge detection algorithm on a parallel multicore platform via Parallel communication software named Message Passing Interface (MPI). The test is being done on ten different images with each image tested in the varying size of 1KxK, 2KxK and 3KxK pixels. Variou!
s threads, ranging from two to ten had been performed on Duo and Quad cores. An interesting result shows that in sequential processing Duo core overcome Quad core speed. As for parallel processing, two thread is the best used for Duo core and eight threads is finest for Quad core.
Keywords: Parallel programming, Sobel edge algorithm, Message passing interface , Multicore.
EXTENSION of the file: .pdf
Special (Invited) Session: Analysis of Parallel Multicore Performance on Sobel Edge Detector
Organizer of the Session: 659-467
How Did you learn about congress: yazid@umt.edu.my
IP ADDRESS: 60.50.83.60