Optimization of Face Recognition Algorithm using Artificial Neural Network

Simple item page
dc.contributor.authorNwoye, E. O.
dc.contributor.authorChukwuma, V. M.
dc.contributor.authorOronti, I. B.
dc.contributor.authorNwaneri, S. C.
dc.date.accessioned2019-09-20T15:22:49Z
dc.date.available2019-09-20T15:22:49Z
dc.date.issued2012-03-01
dc.description.abstract2. Background: The rapid growth in cellular radio communications necessitates more efficient utilization of spectrum. This is more so in medicine where very large volume of data and information are involved. The increased sharing of spectrum due to this growth translates into a higher likelihood of users interfering with one another. So cell capacity is inherently interference limited, particularly by co-channel interference (CCI) and adjacent channel interference (ACI). Objective: one of the solutions to combat these interferences is to control of power. Two types of power control are used in wireless network communication systems: centralised power control (CPC) and distributed power control (DPC). Centralized power control is computationally very complex for large systems as it assumes that all information about the link gains is only local information to adjust power levels of each transmitted signal. DPC is therefore more realistic when the number of mobiles grows and will be used as the foundational system model in this work. Methods: Many optimization techniques have been used for distributed power control system models. However, the techniques have been restricted to the traditional optimisation methods which use the characteristics of the problem to determine the next sampling point. In this work Genetic Algorithm GA, an evolutionary approach is used. It follows the concept of evolution by stochastically developing generations of power-solution populations based on a fitness score. Result: it can be seen that the convergence speed of GA is not as fast as that of the conventional method sand consequently a higher outage probability. The simulation results show that GA is more robust. More realistic performance, very proactive to noise, fading or shadowing Conclusion: A genetic algorithm responds and adapts to this change on the fly but many traditi9onal optimization procedures must restart refresh which is computationally expensive.en_US
dc.identifier.issn0795-2333
dc.identifier.urihttps://ir.unilag.edu.ng/handle/123456789/5941
dc.language.isoenen_US
dc.publisherJournal of Engineering Researchen_US
dc.relation.ispartofseries17;01
dc.subjectBackpropagationen_US
dc.subjectNeural Networken_US
dc.subjectFace Recognitionen_US
dc.subjectDatabaseen_US
dc.subjectPCAen_US
dc.subjectEigenfaceen_US
dc.titleOptimization of Face Recognition Algorithm using Artificial Neural Networken_US
dc.typeArticleen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
4_Optimization of Face Recognition Algorithm using Artificial Neural Network.pdf
Size:
6.86 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Biomedical Engineering- Scholarly Publications