Improved Personal Identification Using Face and Hand Geometry Fusion and Support Vector Machines

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Scopus citations


Multimodal biometric authentication, either verification or identification, systems are known to offer relatively better security solutions than single modality systems. In this paper, we present an identification system based on the fusion of face and hand geometry at the feature level. For face images, we use two-dimensional Discrete Cosine Transform (DCT) to extract discriminant face features which are then combined with hand geometric features. The augmented feature vectors are classified using support vector machines. We compare the performance of the proposed solution with two of the popular classifiers: rule based and decision trees. We also study the impact of feature normalization and selection on the performance. The experimental work shows that the proposed system can lead to great improvement in person identification as compared to other approaches with more than 99% accuracy and very low false acceptance and false rejection rates.

Original languageEnglish
Title of host publicationNetworked Digital Technologies- 4th International Conference, NDT 2012, Proceedings
EditorsRachid Benlamri
Number of pages9
StatePublished - 2012

Publication series

NameCommunications in Computer and Information Science
Volume294 PART 2
ISSN (Print)1865-0929


  • Access Control
  • Authentication
  • Biometrics
  • Discrete Cosine Transform
  • Face
  • Hand Geometry
  • Identification
  • Machine Learning
  • Support Vector Machine
  • Verification

ASJC Scopus subject areas

  • General Computer Science
  • General Mathematics


Dive into the research topics of 'Improved Personal Identification Using Face and Hand Geometry Fusion and Support Vector Machines'. Together they form a unique fingerprint.

Cite this