Upgrading the precision of face recognition using the gradient of a facet function

Hee-Sung Kim; Jun Hee Cho

doi:10.1117/12.664486

20 February 2006 Upgrading the precision of face recognition using the gradient of a facet function

Hee-Sung Kim, Jun Hee Cho

Proceedings Volume 6041, ICMIT 2005: Information Systems and Signal Processing; 60412X (2006) https://doi.org/10.1117/12.664486
Event: ICMIT 2005: Merchatronics, MEMS, and Smart Materials, 2005, Chongqing, China

Abstract

One of the major characteristics of human face is shown in the spatial curvature properties of face surface. The spatial curvature in object images can be represented by a set of gradient directions of the sub surfaces of the images. The image is appropriately divided into the same size of patches. Each patch is the sub surface of the image. A facet function on a patch can be obtained using the gray values of the pixels in the patch. A set of gradient directions of a group of the facet functions reflects one of the inherent curvatures of a face image. Two coefficients of the facet function indicate the gradient at the center of the patch. These coefficients form feature vectors for face discrimination or recognition. Computer computation suggests that the patch size of 5x5 yields the most precise recognition rate of 96.5.

Citation Download Citation

Hee-Sung Kim and Jun Hee Cho "Upgrading the precision of face recognition using the gradient of a facet function", Proc. SPIE 6041, ICMIT 2005: Information Systems and Signal Processing, 60412X (20 February 2006); https://doi.org/10.1117/12.664486

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