Algorithms for assessing the quality of facial images

Mohamed Abdel-Mottaleb, Mohammad H. Mahoor

Research output: Contribution to journalArticlepeer-review

30 Scopus citations

Abstract

The National Institute of Standards and Technology (NIST) published the NIST Fingerprint Image Quality algorithm in August, 2004, to address the problem of poor quality of facial images generated by video surveillance cameras, installed in public places. These algorithms can be used in the Quality Assessment (QA) module of face recognition systems and video cameras, to solve the problem of poor quality images. It is also needed to measure the level or intensity of the factors that affect the quality of facial images, to develop these algorithms for assessing the quality of these images. Two significant strategies have also been developed to deal with these problems. One of the strategies is concerned with improving blurring, lighting conditions, and head pose effects. The other strategy is involved with improving facial expressions of the facial images.

Original languageEnglish (US)
Pages (from-to)10-17
Number of pages8
JournalIEEE Computational Intelligence Magazine
Volume2
Issue number2
DOIs
StatePublished - May 1 2007

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Algorithms for assessing the quality of facial images'. Together they form a unique fingerprint.

Cite this