Algorithms For An Integrated Vision-Based Home Security System

This thesis delves into the algorithmic design of two security systems - face recognition system and human motion detection system. The novelty of this scheme is twofold. First, we have developed an eigenedge technique, which employs Principal Component Analysis (PCA) characterisation on the face ed...

全面介紹

Saved in:
書目詳細資料
主要作者: See, John Su Yang
格式: Thesis
出版: 2005
主題:
標簽: 添加標簽
沒有標簽, 成為第一個標記此記錄!
id my-mmu-ep.858
record_format uketd_dc
spelling my-mmu-ep.8582010-07-06T04:11:14Z Algorithms For An Integrated Vision-Based Home Security System 2005-04 See, John Su Yang TH Building construction This thesis delves into the algorithmic design of two security systems - face recognition system and human motion detection system. The novelty of this scheme is twofold. First, we have developed an eigenedge technique, which employs Principal Component Analysis (PCA) characterisation on the face edge map derived from a novel fuzzy-based Gaussian-like edge detector. Secondly, a three-tier cascaded classification structure has been developed to enable reclassification using different features or classifiers to improve recognition rate, while maintaining efficiency and computational cost at an acceptable level. 2005-04 Thesis http://shdl.mmu.edu.my/858/ http://myto.perpun.net.my/metoalogin/logina.php masters Multimedia University Research Library
institution Multimedia University
collection MMU Institutional Repository
topic TH Building construction
spellingShingle TH Building construction
See, John Su Yang
Algorithms For An Integrated Vision-Based Home Security System
description This thesis delves into the algorithmic design of two security systems - face recognition system and human motion detection system. The novelty of this scheme is twofold. First, we have developed an eigenedge technique, which employs Principal Component Analysis (PCA) characterisation on the face edge map derived from a novel fuzzy-based Gaussian-like edge detector. Secondly, a three-tier cascaded classification structure has been developed to enable reclassification using different features or classifiers to improve recognition rate, while maintaining efficiency and computational cost at an acceptable level.
format Thesis
qualification_level Master's degree
author See, John Su Yang
author_facet See, John Su Yang
author_sort See, John Su Yang
title Algorithms For An Integrated Vision-Based Home Security System
title_short Algorithms For An Integrated Vision-Based Home Security System
title_full Algorithms For An Integrated Vision-Based Home Security System
title_fullStr Algorithms For An Integrated Vision-Based Home Security System
title_full_unstemmed Algorithms For An Integrated Vision-Based Home Security System
title_sort algorithms for an integrated vision-based home security system
granting_institution Multimedia University
granting_department Research Library
publishDate 2005
_version_ 1747829233353228288