TY  - JOUR
T1  - Experimental Analysis of Face Recognition from Infrared Camera using
Retinex Algorithm and Linear Discriminant Analysis (LDA)
AU - Nurfakhrian Aziz, Mohamad AU - Luhur Prasasti, Anggunmeka AU - Waluyo Purboyo, Tito 
JO  - Journal of Engineering and Applied Sciences
VL  - 14
IS  - 8
SP  - 2594
EP  - 2609
PY  - 2019
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2019.2594.2609
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2019.2594.2609
KW  - Infrared camera
KW  -face recognition
KW  -image quality enhancement
KW  -linear discriminant analysis
KW  -retinex
KW  -MSR
AB  - Face recognition using infrared technology is increasingly being used for the security purpose, it is
Closed Circuit Television (CCTV) camera. CCTV camera is generally using infrared technology to monitor the
surrounding environment in every condition whether it’s clear or even dark one without any lighting. Any
evidence of crime captured by CCTV camera will be used for legal purposes. As we know that the image from
infrared camera has poor quality, especially in terms of illumination and image detail, so, it’s hard to recognize
someone from infrared camera. Therefore, the image quality from infrared camera needs to be enhanced. Retinex
algorithm works like retina (in the eyes) and cortex (in the brain) to adapt the illumination. SSR (Single-Scale
Retinex) and MSR (Multi-Scale Retinex) methods has been used in this research to be analyzed their
effectiveness in face recognition using LDA (Linear Discriminant Analysis) method. In the tests that have been
done, people face recognition in very dark conditions can still be recognized compared to face recognition
without retinex and MSR has better accuracy than SSR one.
ER  -