Cover Image

Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]

Currently the computations of solving problems are becoming complex. To make sure the results are accurate, large dataset should be included in the computation. It is not a good practice to process large dataset using normal computers because it is time consuming and will lead to system crash. To av...

Full description

Saved in:
Bibliographic Details
Main Authors: Sulaiman, Hanifah, Abdul Hadi, Normi, Ibrahim, Arsmah, Abd Halim, Muhammad Syawal
Format: Book Section
Language:English
Published: Research Innovation Business Unit 2014
Subjects:
TD Environmental technology. Sanitary engineering
Computer engineering. Computer hardware
Photoelectronic devices (General)
Online Access:https://ir.uitm.edu.my/id/eprint/70485/1/70485.pdf
https://ir.uitm.edu.my/id/eprint/70485/
http://www.iidex.com.my
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uitm.ir.70485
record_format eprints
spelling my.uitm.ir.704852023-02-24T02:01:05Z https://ir.uitm.edu.my/id/eprint/70485/ Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.] Sulaiman, Hanifah Abdul Hadi, Normi Ibrahim, Arsmah Abd Halim, Muhammad Syawal TD Environmental technology. Sanitary engineering Computer engineering. Computer hardware Photoelectronic devices (General) Currently the computations of solving problems are becoming complex. To make sure the results are accurate, large dataset should be included in the computation. It is not a good practice to process large dataset using normal computers because it is time consuming and will lead to system crash. To avoid system crash, parallel computation as a technique is becoming the current trend where the large dataset can be executed and carried out simultaneously thus reducing the processing time. In parallel computations, the operations on large problem can be divided into smaller ones which are then solved concurrently. ARS cluster is a mini parallel laboratory that was developed specially for surface reconstruction in image processing. This innovation of the parallel processing laboratory is a self-developed laboratory where it is designed by using eight processors and connected via internal local area network. The operating system used is open source LINUX Fedora 7. It also uses MATLAB Distributed Computing Server (MDCS) as the platform for parallel processing. Client PC is needed to run the parallel programs. The technology developed was specially designed to process a medical imaging such as Computerized Tomography (CT), mammogram and MRI. Medical usage of 3D reconstructions is being extensively implemented especially using MRI equipment. For example, doctors can see a 3D model of a person’s brain without performing any surgery and can easily detect the actual location of a brain’s tumour. Another example is the face of a human can be identified through the 3D reconstruction form CT scan multi-slice images. ARS cluster has proven that it can solve a problem with less time rather than solving the problem using single processor. Parallel computer is preferable since it provides faster processing time rather than sequential processing when it comes to doing repetitive calculations on vast amounts of data. Therefore, the problems can be solved in a shorter time. The parallel computer not only reduces the processing time, but also efficient and effective to produce the intended results. Parallel laboratory is very costly in term of its development. However, ARS cluster managed to be developed in a low-cost environment with the performance as good as the high cost parallel laboratory. The cluster is a green home-grown technology that approached 3R culture; reuse the old personal computer, Recycle the old computers and Reduce the processing time and easy to carry anywhere because the size for each processor is the same as the CPU of personal computer. The laboratory can benefit researchers, students and people who are interested in processing task which involves large dataset and huge repetitive computation. It can assist researchers to enhance the knowledge and experience in the parallel processing environment. This laboratory can be commercialised by opening it to the public. Users can come to the laboratory with their problem or can send it on the net (online). The price will be less than a normal parallel system available on the web since this laboratory is a low-cost laboratory. Therefore, this laboratory is an advantage in the parallel computing field. Research Innovation Business Unit 2014 Book Section NonPeerReviewed text en https://ir.uitm.edu.my/id/eprint/70485/1/70485.pdf Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]. (2014) In: IIDEX 2014: invention, innovation & design exposition. Research Innovation Business Unit, Shah Alam, Selangor, p. 59. (Submitted) http://www.iidex.com.my
institution Universiti Teknologi Mara
building Tun Abdul Razak Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Mara
content_source UiTM Institutional Repository
url_provider http://ir.uitm.edu.my/
language English
topic TD Environmental technology. Sanitary engineering
Computer engineering. Computer hardware
Photoelectronic devices (General)
spellingShingle TD Environmental technology. Sanitary engineering
Computer engineering. Computer hardware
Photoelectronic devices (General)
Sulaiman, Hanifah
Abdul Hadi, Normi
Ibrahim, Arsmah
Abd Halim, Muhammad Syawal
Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]
description Currently the computations of solving problems are becoming complex. To make sure the results are accurate, large dataset should be included in the computation. It is not a good practice to process large dataset using normal computers because it is time consuming and will lead to system crash. To avoid system crash, parallel computation as a technique is becoming the current trend where the large dataset can be executed and carried out simultaneously thus reducing the processing time. In parallel computations, the operations on large problem can be divided into smaller ones which are then solved concurrently. ARS cluster is a mini parallel laboratory that was developed specially for surface reconstruction in image processing. This innovation of the parallel processing laboratory is a self-developed laboratory where it is designed by using eight processors and connected via internal local area network. The operating system used is open source LINUX Fedora 7. It also uses MATLAB Distributed Computing Server (MDCS) as the platform for parallel processing. Client PC is needed to run the parallel programs. The technology developed was specially designed to process a medical imaging such as Computerized Tomography (CT), mammogram and MRI. Medical usage of 3D reconstructions is being extensively implemented especially using MRI equipment. For example, doctors can see a 3D model of a person’s brain without performing any surgery and can easily detect the actual location of a brain’s tumour. Another example is the face of a human can be identified through the 3D reconstruction form CT scan multi-slice images. ARS cluster has proven that it can solve a problem with less time rather than solving the problem using single processor. Parallel computer is preferable since it provides faster processing time rather than sequential processing when it comes to doing repetitive calculations on vast amounts of data. Therefore, the problems can be solved in a shorter time. The parallel computer not only reduces the processing time, but also efficient and effective to produce the intended results. Parallel laboratory is very costly in term of its development. However, ARS cluster managed to be developed in a low-cost environment with the performance as good as the high cost parallel laboratory. The cluster is a green home-grown technology that approached 3R culture; reuse the old personal computer, Recycle the old computers and Reduce the processing time and easy to carry anywhere because the size for each processor is the same as the CPU of personal computer. The laboratory can benefit researchers, students and people who are interested in processing task which involves large dataset and huge repetitive computation. It can assist researchers to enhance the knowledge and experience in the parallel processing environment. This laboratory can be commercialised by opening it to the public. Users can come to the laboratory with their problem or can send it on the net (online). The price will be less than a normal parallel system available on the web since this laboratory is a low-cost laboratory. Therefore, this laboratory is an advantage in the parallel computing field.
format Book Section
author Sulaiman, Hanifah
Abdul Hadi, Normi
Ibrahim, Arsmah
Abd Halim, Muhammad Syawal
author_facet Sulaiman, Hanifah
Abdul Hadi, Normi
Ibrahim, Arsmah
Abd Halim, Muhammad Syawal
author_sort Sulaiman, Hanifah
title Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]
title_short Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]
title_full Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]
title_fullStr Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]
title_full_unstemmed Surface reconstruction on mini parallel processing: green technology for image processing / Hanifah Sulaiman … [et al.]
title_sort surface reconstruction on mini parallel processing: green technology for image processing / hanifah sulaiman … [et al.]
publisher Research Innovation Business Unit
publishDate 2014
url https://ir.uitm.edu.my/id/eprint/70485/1/70485.pdf
https://ir.uitm.edu.my/id/eprint/70485/
http://www.iidex.com.my
_version_ 1758953520200941568
score 13.211869

Similar Items