The LDA-Based Face Recognition at a Distance Using Multiple Distance Image

Author

Hae-Min Moon ; Sung Bum Pan

Author_Institution

Dept. of Inf. & Commun. Eng., Chosun Univ., Gwangju, South Korea

fYear

2013

fDate

3-5 July 2013

Firstpage

249

Lastpage

255

Abstract

These days surveillance system has been developed into intelligent system that can judge and handle using human recognition technology that can recognize the people. In this paper, LDA-based face recognition algorithm for the intelligent surveillance system is proposed. While the existing face recognition algorithm uses the short distance images for training images but the proposed algorithm uses face images by distance extracted from 1m to 5m for training images. In the proposed LDA-based long distance face recognition algorithm, bilinear interpolation uses for the normalization and Euclidean Distance measure method is used for similarity measure. As a result of experiment, it is confirmed that the proposed algorithm showed improved face recognition performance 6.1% in short distance and 31% in long distance.

Keywords

computational geometry; face recognition; interpolation; principal component analysis; surveillance; Euclidean distance measure method; LDA-based long distance face recognition algorithm; PCA; bilinear interpolation; human recognition technology; intelligent surveillance system; linear discriminant analysis; multiple distance image; normalization distance measure method; principal component analysis; similarity measure; Cameras; Convolution; Face; Face recognition; Interpolation; Training; Vectors; image interpolation; intelligent surveillance system; linear discriminant analysis (LDA); long distance face recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Innovative Mobile and Internet Services in Ubiquitous Computing (IMIS), 2013 Seventh International Conference on

Conference_Location

Taichung

Type

conf

DOI

10.1109/IMIS.2013.50

Filename

6603681