Illumination influence on iris identification algorithms

Iris recognition is accepted as one of the best biometric methods. Implementing this method into the practical system requires the special attention is paid to the iris images, which are being analysed. Current methods for iris identification can be distinguished with regard to two types of images: taken under near infrared illumination and white visible light. The choice of illumination type has a great influence on preprocessing methods and iris feature extraction algorithms as well. In this paper sensitivity of wavelet methods applied to iris feature extraction to illumination type is discussed and the results for both illumination types are compared.