Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/260200
Title: | Efficient algorithms and architecture for multi variant face recognition |
Researcher: | Viji A |
Guide(s): | Vaidehi V |
Keywords: | Algorithms Engineering and Technology,Computer Science,Computer Science Information Systems Face Recognition |
University: | Anna University |
Completed Date: | 2018 |
Abstract: | In today s digital age, face recognition plays a paramount role in computer vision applications. Although numerous methods for face recognition exist in literature, there is a demand for scale, pose, and illumination invariant algorithm. Automatic facial expression recognition is an important application in Human Computer Interaction. It is still difficult to develop a facial expression recognition system that is real-time, person-independent, camera and illumination robust due to the subtlety, complexity, and variety of facial expressions. In many security and intelligence scenarios Visible light Source (VIS) face images are used as gallery images for enrolment and Near Infra-Red (NIR) face image as the probe image of the same person. Matching face images from different imaging devices or different lighting conditions are called Heterogeneous Face Recognition (HFR). The most challenging issues in HFR are to reduce the modality gap between NIR and VIS face images taken at different devices of the same person under the unconstrained illumination condition. This proposed work focused to solve the above mentioned challenges. An efficient Illumination and Pose Invariant (IPI) methodology to recognize human faces under uncontrolled lighting is proposed and the age of the recognized faces is also found. Face recognition is based on robust preprocessing followed by fusion of Extended Curvature Gabor wavelets (ECG) and Local Binary Patterns (LBP) to extract the features of curvature information and the texture information of face image respectively. As both feature sets are higher in dimension, PCA is used to reduce the dimensionality prior to Z-Score normalization. Nearest Neighbour classifier is used to recognize the face. Support Vector Machine based Regression algorithm is used to estimate the age of a face in an image. newline newline newline |
Pagination: | xxxi, 183p. |
URI: | http://hdl.handle.net/10603/260200 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 6.76 kB | Adobe PDF | View/Open |
02_certificates.pdf | 63.24 kB | Adobe PDF | View/Open | |
03_abstract.pdf | 8.98 kB | Adobe PDF | View/Open | |
04_acknowledgement.pdf | 4.07 kB | Adobe PDF | View/Open | |
05_table of contents.pdf | 37.82 kB | Adobe PDF | View/Open | |
06_list_of_symbols and abbreviations.pdf | 68.81 kB | Adobe PDF | View/Open | |
07_chapter1.pdf | 220.57 kB | Adobe PDF | View/Open | |
08_chapter2.pdf | 160.55 kB | Adobe PDF | View/Open | |
09_chapter3.pdf | 358.49 kB | Adobe PDF | View/Open | |
10_chapter4.pdf | 449.47 kB | Adobe PDF | View/Open | |
11_chapter5.pdf | 280.72 kB | Adobe PDF | View/Open | |
12_chapter6.pdf | 209.84 kB | Adobe PDF | View/Open | |
13_conclusion.pdf | 23.06 kB | Adobe PDF | View/Open | |
14_references.pdf | 47.24 kB | Adobe PDF | View/Open | |
15_list_of_publications.pdf | 5.8 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: