Comparative studies on multispectral palm image fusion for biometrics

  • Authors:
  • Ying Hao;Zhenan Sun;Tieniu Tan

  • Affiliations:
  • National Laboratory of Pattern Recognition, Institute of Automation, CAS;National Laboratory of Pattern Recognition, Institute of Automation, CAS;National Laboratory of Pattern Recognition, Institute of Automation, CAS

  • Venue:
  • ACCV'07 Proceedings of the 8th Asian conference on Computer vision - Volume Part II
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Hand biometrics, including fingerprint, palmprint, hand geometry and hand vein pattern, have obtained extensive attention in recent years. Physiologically, skin is a complex multi-layered tissue consisting of various types of components. Optical research suggests that different components appear when the skin is illuminated with light sources of different wavelengths. This motivates us to extend the capability of camera by integrating information from multispectral palm images to a composite representation that conveys richer and denser pattern for recognition. Besides, usability and security of the whole system might be boosted at the same time. In this paper, comparative study of several pixel level multispectral palm image fusion approaches is conducted and several well-established criteria are utilized as objective fusion quality evaluation measure. Among others, Curvelet transform is found to perform best in preserving discriminative patterns from multispectral palm images.