No abstract

No abstract

No abstract

No abstract

Multispectral palmprint recognition has been investigated for many problems and applications over the last decade. It has become one of the most well-known biometric recognition systems. Its success is due to the rich features that can be extracted and exploited from the multispectral images of palmprint captured within specific wavelength ranges across the electromagnetic spectrum. This paper provides an overview of recent state-of-the-art multispectral palmprint approaches for person recognition. The approaches surveyed are discussed by describing, in particular, their feature extraction, feature fusion, matching and decision algorithms. Finally, a comparative study to evaluate their performances for both verification and identification modes is addressed.