Development Of Famous People Recognition System From Video Sequences
Video based face recognition has become an important task due to the huge demand on the surveillance system application such as monitoring activities of closed circuit TV (CCTV). There are some cases due to the security reason, an identity of interest (IoI) need to be searched manually from all the...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2015
|
Subjects: | |
Online Access: | http://eprints.usm.my/41328/1/YII_WEN_WEN_24_Pages.pdf http://eprints.usm.my/41328/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Universiti Sains Malaysia |
Language: | English |
id |
my.usm.eprints.41328 |
---|---|
record_format |
eprints |
spelling |
my.usm.eprints.41328 http://eprints.usm.my/41328/ Development Of Famous People Recognition System From Video Sequences Yii , Wen Wen TK7800-8360 Electronics Video based face recognition has become an important task due to the huge demand on the surveillance system application such as monitoring activities of closed circuit TV (CCTV). There are some cases due to the security reason, an identity of interest (IoI) need to be searched manually from all the captured video through CCTV devices. This task is tiring, tedious and wasting time by looking through video one by one. Therefore, the initial work of developing a basic system for video based face recognition on searching selected identity of interest automatically is proposed in this research. In this research, combination of Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are used as feature extractor. Three feature classifiers used in this system are Euclidean Distance, Manhattan Distance, and Learning Vector Quantization Neural Network (LVQNET). Comparison between the performance of three classifiers have been conducted on the overall recognition results of selected video for famous people recognition. Experimental results of the proposed method on selected video has the recognition accuracy up to 76.4% for Euclidean distance, 75.9% for Manhattan distance,and 64.5% for LVQNET Network with histogram equalization filtering technique is applied. The ability of the proposed system has been proven to be effective and has a significant value for intelligent applications such as automated video based face recognition on selected person. 2015 Thesis NonPeerReviewed application/pdf en http://eprints.usm.my/41328/1/YII_WEN_WEN_24_Pages.pdf Yii , Wen Wen (2015) Development Of Famous People Recognition System From Video Sequences. Masters thesis, Universiti Sains Malaysia. |
institution |
Universiti Sains Malaysia |
building |
Hamzah Sendut Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Sains Malaysia |
content_source |
USM Institutional Repository |
url_provider |
http://eprints.usm.my/ |
language |
English |
topic |
TK7800-8360 Electronics |
spellingShingle |
TK7800-8360 Electronics Yii , Wen Wen Development Of Famous People Recognition System From Video Sequences |
description |
Video based face recognition has become an important task due to the huge demand on the surveillance system application such as monitoring activities of closed circuit TV (CCTV). There are some cases due to the security reason, an identity of interest (IoI) need to be searched manually from all the captured video through CCTV devices. This task is tiring, tedious and
wasting time by looking through video one by one. Therefore, the initial work of developing
a basic system for video based face recognition on searching selected identity of interest automatically
is proposed in this research. In this research, combination of Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are used as feature extractor. Three feature classifiers used in this system are Euclidean Distance, Manhattan Distance, and Learning
Vector Quantization Neural Network (LVQNET). Comparison between the performance of three classifiers have been conducted on the overall recognition results of selected video for famous people recognition. Experimental results of the proposed method on selected video has the recognition accuracy up to 76.4% for Euclidean distance, 75.9% for Manhattan distance,and 64.5% for LVQNET Network with histogram equalization filtering technique is applied. The ability of the proposed system has been proven to be effective and has a significant value
for intelligent applications such as automated video based face recognition on selected person. |
format |
Thesis |
author |
Yii , Wen Wen |
author_facet |
Yii , Wen Wen |
author_sort |
Yii , Wen Wen |
title |
Development Of Famous People Recognition System From Video Sequences
|
title_short |
Development Of Famous People Recognition System From Video Sequences
|
title_full |
Development Of Famous People Recognition System From Video Sequences
|
title_fullStr |
Development Of Famous People Recognition System From Video Sequences
|
title_full_unstemmed |
Development Of Famous People Recognition System From Video Sequences
|
title_sort |
development of famous people recognition system from video sequences |
publishDate |
2015 |
url |
http://eprints.usm.my/41328/1/YII_WEN_WEN_24_Pages.pdf http://eprints.usm.my/41328/ |
_version_ |
1643710190384578560 |