Human Face Recognition in the Wavelet Compressed Domain

Abstract

Automated recognition of human faces is a rapidly growing field. To reduce the amount of disk space required in maintaining an image database, the images are compressed. To reduce computation time, face recognition is performed in the compressed domain. But altering the image through compression could affect the accuracy of the recognition process. An experiment is described in which the Eigenface approach to face recognition is performed in the compressed domain. The results of wavelet compressed images versus their uncompressed original images are presented. Experimental results indicate that it is possible, even preferable, to perform face recognition in the compressed domain.

Department(s)

Computer Science

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 The Authors, All rights reserved.

Publication Date

01 Dec 1997

This document is currently not available here.

Share

 
COinS