Academic Journals Database
Disseminating quality controlled scientific knowledge

Gabor Wavelet based Face Recognition System using EWCVT and Bagging Adaboost Algorithm

ADD TO MY LIST
 
Author(s): Srinivasan A

Journal: International Journal of Computer Applications
ISSN 0975-8887

Volume: 16;
Issue: 3;
Start page: 49;
Date: 2011;
Original page

Keywords: Face recognition | Gabor Wavelets | Local Gabor Phase pattern | Global Gabor Phase Pattern | Adaptive Binning | Spatial Histograms and Tessellations

ABSTRACT
Facial recognition system is a computer application for identifying or verifying a person from a digital image automatically. One among this method involves comparison of the facial features from the test image database with trained facial database. Histogram Gabor Phase Pattern (HGPP) is an extended histogram feature which represents original image by combining Local Gabor Phase Patterns (LGPP) and Global Gabor Phase Patterns (GGPP). This method lacks in efficiency and computational complexity because it involves huge volume of data. To reduce the data, edge weighted centroidal voronoi tessellation (EWCVT) is used and to increase the efficiency a classifier called Bagging AdaBoost is used. Bagging-AdaBoost classifier bridges the semantic gap between the low-level feature vectors of the image and the high-level concepts. The proposed system incorporates the EWCVT and bagging technique to improve the accuracy, stability and robustness of the system. The results obtained prove that the proposed system has improved accuracy in recognition, more stability, less computational complexity and processing time.
Why do you need a reservation system?      Affiliate Program