Image quality-based adaptive illumination normalisation for face recognition

Harin Sellahewa; Sabah A. Jassim

doi:10.1117/12.819087

5 May 2009 Image quality-based adaptive illumination normalisation for face recognition

Harin Sellahewa, Sabah A. Jassim

Proceedings Volume 7306, Optics and Photonics in Global Homeland Security V and Biometric Technology for Human Identification VI; 73061V (2009) https://doi.org/10.1117/12.819087
Event: SPIE Defense, Security, and Sensing, 2009, Orlando, Florida, United States

Abstract

Automatic face recognition is a challenging task due to intra-class variations. Changes in lighting conditions during enrolment and identification stages contribute significantly to these intra-class variations. A common approach to address the effects such of varying conditions is to pre-process the biometric samples in order normalise intra-class variations. Histogram equalisation is a widely used illumination normalisation technique in face recognition. However, a recent study has shown that applying histogram equalisation on well-lit face images could lead to a decrease in recognition accuracy. This paper presents a dynamic approach to illumination normalisation, based on face image quality. The quality of a given face image is measured in terms of its luminance distortion by comparing this image against a known reference face image. Histogram equalisation is applied to a probe image if its luminance distortion is higher than a predefined threshold. We tested the proposed adaptive illumination normalisation method on the widely used Extended Yale Face Database B. Identification results demonstrate that our adaptive normalisation produces better identification accuracy compared to the conventional approach where every image is normalised, irrespective of the lighting condition they were acquired.

Citation Download Citation

Harin Sellahewa and Sabah A. Jassim "Image quality-based adaptive illumination normalisation for face recognition", Proc. SPIE 7306, Optics and Photonics in Global Homeland Security V and Biometric Technology for Human Identification VI, 73061V (5 May 2009); https://doi.org/10.1117/12.819087

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available