A fusion iris feature extraction method based on fisher linear discriminant
Yong Zhang, Yan Wo · 2013
Iris in human's eyes contains enrich texture information which is useful for identity authentication. A key and still open issue in iris recognition is how best representing such textural information by using a set of feature vectors. This paper proposes a new method for iris feature recognition by fusing 2D and ID features. This iris recognition system consisted of four major stages: Iris Preprocessing, Feature Extraction, Matching and Combination. 2D Gabor and ID Log Gabor Filter extract phase information as 2D and ID features, Hamming Distance(HD) is used to to evaluate the effectiveness of the feature vectors. We also propose to apply the Fisher's Linear Discriminant(FLD) to determine the weights of the combination. Under experimental conditions, the fusion method obtains an encouraging and positive performance.