A novel facial expression recognition method using bi-dimensional EMD based edge detection
- WCU Author/Contributor (non-WCU co-authors, if there are any, appear on document)
- Zijing Qin (Creator)
- Institution
- Western Carolina University (WCU )
- Web Site: http://library.wcu.edu/
- Advisor
- James Zhang
Abstract: Facial expressions provide an important channel of nonverbal communication. Facial
recognition techniques detect people’s emotions using their facial expressions and
have found applications in technical fields such as Human-Computer-Interaction (HCI)
and security monitoring. Technical applications generally require fast processing and decision
making. Therefore, it is imperative to develop innovative recognition methods that
can detect facial expressions effectively and efficiently.
Traditionally, human facial expressions are recognized using standard images. Existing
methods of recognition require subjective expertise and high computational costs.
This thesis proposes a novel method for facial expression recognition using image edge
detection based on Bi-dimensional Empirical Mode Decomposition (BEMD). In this research,
a BEMD based edge detection algorithm was developed, a facial expression measurement
metric was created, and an intensive database testing was conducted. The success
rates of recognition suggest that the proposed method could be a potential alternative to traditional methods for human facial expression recognition with substantially lower
computational costs. Furthermore, a possible blind-detection technique was proposed as
a result of this research. Initial detection results suggest great potential of the proposed
method for blind-detection that may lead to even more efficient techniques for facial expression
recognition.
A novel facial expression recognition method using bi-dimensional EMD based edge detection
PDF (Portable Document Format)
1850 KB
Created on 4/1/2010
Views: 4851
Additional Information
- Publication
- Thesis
- Language: English
- Date: 2010
- Keywords
- Edge detection, EMD, Facial recognition
- Subjects
- Facial expression -- Testing
- Facial expression -- Research
- Hilbert-Huang transform