Use this URL to cite or link to this record in EThOS:
Title: Recognizing faces : an approach based on Gabor wavelets
Author: Shen, LinLin
Awarding Body: University of Nottingham
Current Institution: University of Nottingham
Date of Award: 2005
Availability of Full Text:
Access from EThOS:
Access from Institution:
As a hot research topic over the last 25 years, face recognition still seems to be a difficult and largely problem. Distortions caused by variations in illumination, expression and pose are the main challenges to be dealt with by researchers in this field. Efficient recognition algorithms, robust against such distortions, are the main motivations of this research. Based on a detailed review on the background and wide applications of Gabor wavelet, this powerful and biologically driven mathematical tool is adopted to extract features for face recognition. The features contain important local frequency information and have been proven to be robust against commonly encountered distortions. To reduce the computation and memory cost caused by the large feature dimension, a novel boosting based algorithm is proposed and successfully applied to eliminate redundant features. The selected features are further enhanced by kernel subspace methods to handle the nonlinear face variations. The efficiency and robustness of the proposed algorithm is extensively tested using the ORL, FERET and BANCA databases. To normalize the scale and orientation of face images, a generalized symmetry measure based algorithm is proposed for automatic eye location. Without the requirement of a training process, the method is simple, fast and fully tested using thousands of images from the BioID and BANCA databases. An automatic user identification system, consisting of detection, recognition and user management modules, has been developed. The system can effectively detect faces from real video streams, identify them and retrieve corresponding user information from the application database. Different detection and recognition algorithms can also be easily integrated into the framework.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: QA299 Analysis