TY  - JOUR
T1  - Improved Face Recognition using a Modified PSO Based Self-Weighted
Linear Collaborative Discriminant Regression Classification
AU - Shailaja, K. AU - Anuradha, B. 
JO  - Journal of Engineering and Applied Sciences
VL  - 12
IS  - 23
SP  - 7234
EP  - 7241
PY  - 2017
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2017.7234.7241
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2017.7234.7241
KW  - Deep learning
KW  -face recognition
KW  -linear collaborative discriminant regression classification
KW  -modified particle swarm optimization
KW  -reconstruction error
AB  - Biometric authentication utilizing Face Recognition (FR) is emerging as a significant research field.
In this scenario, Linear Collaborative Discriminant Regression Classification (LCDRC) scheme is undertaken
for experimental examination. Whereas, LCDRC could not able to categorize the samples that scattered around
the intersections and also it gives a poor outcome in severe lighting variations. In order to overcome this
difficulties, an effective weight function along with Deep Learning (DL) is included in LCDRC. Respective
weight function is selected based on Modified-Particle Swarm Optimization (MPSO) algorithm. This proposed
methodology significantly maximize the Reconstruction Error (RE) between the classes and also it minimize the
RE within the class. Though, the proposed methodology not only out-performs LCDRC, also it provides
superior outcome in terms of accuracy.
ER  -