Face classification using PCA and K-nearest neighbor method

Principle Component Analysis is one of the most useful method using for human face recognition, our aim in this research is to implement an understandable recognition program with language R; furthermore, comparing K-Nearest Neighbor classifier with two different distance measurement and Support Vec...

Full description

Bibliographic Details
Main Author: Liang, Wen-Wei
Format: Dissertation (University of Nottingham only)
Language:English
Published: 2014
Online Access:https://eprints.nottingham.ac.uk/30757/
_version_ 1848794052143611904
author Liang, Wen-Wei
author_facet Liang, Wen-Wei
author_sort Liang, Wen-Wei
building Nottingham Research Data Repository
collection Online Access
description Principle Component Analysis is one of the most useful method using for human face recognition, our aim in this research is to implement an understandable recognition program with language R; furthermore, comparing K-Nearest Neighbor classifier with two different distance measurement and Support Vector Machines on their efficiency. PCA indeed reduce the run time of processing massive face images data and reach an ideal accuracy in controlled circumstances. We found that Euclidean K-NN classifier generally has better accuracy than Manhattan K-NN classifier; however, it suggested that they are suitable to use for smaller k value, especially one. The eigenfaces applied to construct the face space is essential as well, we reached best accuracies when selecting around 10% to 20% of eigens. However, the reconstruction suggests in an opposite way, more eigenfaces help the rebuilt much ideally; more than half of eigenvectors selection can reconstruct the face to be easily recognised as the original people.
first_indexed 2025-11-14T19:10:03Z
format Dissertation (University of Nottingham only)
id nottingham-30757
institution University of Nottingham Malaysia Campus
institution_category Local University
language English
last_indexed 2025-11-14T19:10:03Z
publishDate 2014
recordtype eprints
repository_type Digital Repository
spelling nottingham-307572017-10-19T15:08:54Z https://eprints.nottingham.ac.uk/30757/ Face classification using PCA and K-nearest neighbor method Liang, Wen-Wei Principle Component Analysis is one of the most useful method using for human face recognition, our aim in this research is to implement an understandable recognition program with language R; furthermore, comparing K-Nearest Neighbor classifier with two different distance measurement and Support Vector Machines on their efficiency. PCA indeed reduce the run time of processing massive face images data and reach an ideal accuracy in controlled circumstances. We found that Euclidean K-NN classifier generally has better accuracy than Manhattan K-NN classifier; however, it suggested that they are suitable to use for smaller k value, especially one. The eigenfaces applied to construct the face space is essential as well, we reached best accuracies when selecting around 10% to 20% of eigens. However, the reconstruction suggests in an opposite way, more eigenfaces help the rebuilt much ideally; more than half of eigenvectors selection can reconstruct the face to be easily recognised as the original people. 2014-12-09 Dissertation (University of Nottingham only) NonPeerReviewed application/pdf en https://eprints.nottingham.ac.uk/30757/1/WLiang_odledata_temp_turnitintool_576361323._13264_1411071324_98240.pdf Liang, Wen-Wei (2014) Face classification using PCA and K-nearest neighbor method. [Dissertation (University of Nottingham only)]
spellingShingle Liang, Wen-Wei
Face classification using PCA and K-nearest neighbor method
title Face classification using PCA and K-nearest neighbor method
title_full Face classification using PCA and K-nearest neighbor method
title_fullStr Face classification using PCA and K-nearest neighbor method
title_full_unstemmed Face classification using PCA and K-nearest neighbor method
title_short Face classification using PCA and K-nearest neighbor method
title_sort face classification using pca and k-nearest neighbor method
url https://eprints.nottingham.ac.uk/30757/