Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/513354Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Md. Jan Nordin, Prof. Dr. | - |
| dc.contributor.author | Amir Jamshid Nezhad (P47852) | - |
| dc.date.accessioned | 2023-10-16T04:35:49Z | - |
| dc.date.available | 2023-10-16T04:35:49Z | - |
| dc.date.issued | 2012-08-02 | - |
| dc.identifier.other | ukmvital:120067 | - |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/513354 | - |
| dc.description | In recent years, computer technology has led to remarkable increase in use of classification for intelligent systems. Classification of facial expressions opens a new direction to increase the interaction between computers and human. The major issue which divides the facial expressions from the other classification domains is natural based behavior of human as the objects to express the emotions which should be recognized with the classifier model. Existing research recognize the emotions using a range of classification techniques. However, low accuracy rate, large training set, large extracted features or priority for sequence images are the main drawbacks of those works. One of the recent techniques to address the facial expressions problem is Fuzzy Rule Based System (FRBS) which is used as a successful method to model and solve the natural based problems. However, FRBS is poor to adapt the existing knowledge with the diverse conditions. Therefore, the knowledge base which is created for the FRBS by experts such as the Fuzzy membership parameters is not optimum to use in the facial expressions classification model. Furthermore, selecting a proper type of Fuzzy membership functions such as triangular, trapezoidal and bell shaped or combination of them is a boring and uncertain work. With regarding to such problems, this research is an attempt to develop the classification model with small size of extracted features, robustness and optimum performance. The hybrid Genetic-Fuzzy rule based model as the proposed classification not only drives the simplicity of Fuzzy logic on a complicate domain but also optimizes the performance of Fuzzy classification while the limited raw input data as the features are used. In order to improve the Fuzzy knowledge base, membership parameters are tuned in a learning process while the constant rules are defined based on the psychological studies and statistical analysis of emotional states. In this model, the proposed Genetic Algorithm simulates and improves the honey bees offspring generation process called Bee Royalty Offspring Algorithm (BROA) to improve the training process of classic Genetic Algorithms. Therefore, Fuzzy membership parameters are tuned faster also present higher accuracy rate in the classification model. As a result, the main contributions of this research are using less extracted points from the static facial images compared with the existing works. This help to reduce the computation process in the classification, presenting a novel hybrid Genetic-Fuzzy model for the expressions recognition problem. Modification of the Genetic Algorithm to improve its training process and presenting higher accuracy rate than the Fuzzy rule based models and the corresponding existing classification techniques. Furthermore, the accuracy, reliability and validity of the proposed model have been measured and evaluated with several experiments. The outcomes have been compared with the result of previous techniques under the comparison criteria which influence the accuracy rate of classification. The comparison results illustrate that the Genetic-Fuzzy classification model improves considerably the accuracy rate and performance of FRBS while the BROA modify the training process of Genetic Algorithms. Moreover, the results showed that the proposed model is reliable, valid and robustness in the small size of facial feature points, using static images and the database which included non-controlled emotional subjects.,Certification of Master's / Doctoral Thesis" is not available | - |
| dc.language.iso | eng | - |
| dc.publisher | UKM, Bangi | - |
| dc.relation | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | - |
| dc.rights | UKM | - |
| dc.subject | Fuzzy systems | - |
| dc.subject | Facial expression | - |
| dc.subject | Image processing | - |
| dc.subject | Universiti Kebangsaan Malaysia -- Dissertations | - |
| dc.subject | Dissertations, Academic -- Malaysia | - |
| dc.title | A hybrid genetic-fuzzy model for classification of facial expressions | - |
| dc.type | Theses | - |
| dc.format.pages | 193 | - |
| dc.identifier.callno | TA1637.N449 2012 3 tesis | - |
| dc.identifier.barcode | 004051(2019) | - |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_120067+SOURCE1+SOURCE1.0.PDF Restricted Access | 541.66 kB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.