iRepository at Perpustakaan UniMAP >
Final Year Project Papers & Reports >
School of Mechatronic Engineering (FYP) >

Please use this identifier to cite or link to this item: http://dspace.unimap.edu.my:80/xmlui/handle/123456789/3802

Title: Face recognition using neural network
Authors: Azlan Zainul
???metadata.dc.contributor.advisor???: Mohd Rizon Mohamed Juhari, Prof. Dr. ( Advisor)
Keywords: Face recognition;Face perception -- Computer simulation;Neural networks (Computer science);Human face recognition (Computer science);Image processing
Issue Date: Mar-2008
Publisher: Universiti Malaysia Perlis
???metadata.dc.publisher.department???: School of Mechatronic Engineering
Abstract: This dissertation provides a study of camera-based face recognition research. There are two underlying motivations to conduct this project: the first is to study and analyze type of feature extraction method, and the second is to use Artificial Neural Network as method recognition of faces. This paper presents a technique for recognizing human faces in digital color images which has been changed into binary. The system relies on two step processes which first do some pre-processing to detect feature in the image and then extracts information from these feature which might indicate the location of a face in the image. The feature extraction is performed using three different methods such as Eigen, SVD (Singular Value Decomposition) and Component Labeling and the comparison is done to choose best method for feature classification. The face detection is performed on a binary image containing only the detected face areas. A combination of threshold and mathematical tools are used to extract object features that would indicate the presence of a face. The face recognition process works predictably and reliably. Although remarkably robust, face recognition is not perfectly invariant to pose and viewpoint changes. Back-Propagation Neural Network is applied in this project as it is much suitable compared with other type of neural network for feature classification. The system not perfectly works at the moment due to the result from feature extraction have not been normalized and randomized before it could be train in the network
URI: http://hdl.handle.net/123456789/3802
Appears in Collections:School of Mechatronic Engineering (FYP)
Mohd. Rizon Mohamed Juhari, Prof. Ir. Dr.

Files in This Item:

File Description SizeFormat
References and appendix.pdf776.62 kBAdobe PDFView/Open
Conclusion.pdf132.31 kBAdobe PDFView/Open
Results and discussion.pdf299.61 kBAdobe PDFView/Open
Methodology.pdf215.49 kBAdobe PDFView/Open
Literature review.pdf176.3 kBAdobe PDFView/Open
Introduction.pdf134.79 kBAdobe PDFView/Open
Abstract, Acknowledgement.pdf160.19 kBAdobe PDFView/Open
View Statistics

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0! Perpustakaan Tuanku syed Faizuddin Putra, Kampus Pauh Putra, Universiti Malaysia Perlis, 02600, Arau Perlis
TEL: +604-9885420 | FAX: +604-9885405 | EMAIL: rujukan@unimap.edu.my Feedback