Iris verification system based on curvelet transform

Abstract

International audienceThe performance of the iris verification process highly depends on its extractor of iris features. So, to reduce the dimensionality of the iris image and improve the recognition rate, an iris features extraction method based on curvelet transform is proposed and presented in this paper. Thus, our paper focuses on presenting of our curvelet-based iris features extraction method. This method consists of two steps: decompose images into set of sub-bands by the curvelet transform and automatic extraction of the most discriminative features of these sub-bands. An extensive experimental results show that the proposed method is effective and encouraging

Hand verification system based on multi-features fusion

“…They are defined as: m(x, y) = J (L(x + 1, y) -L(x -1, y)) 2 + (L(x, y + 1) -L(x, y -1))2 -1)) ) x,y = at an (L(x + 1, y) -L(x -1, y)) (5) The orientation of one key point is getting according to the orientation histogram created with 36 orientations. The peak orientation is considered as the main orientation of this key point.…”

“…Biometric technology is considered as the safest means for identity authentication. In fact, several biometric modalities have been studied and adopted such as fingerprint [6], fa ce [8], [9], hand shape [18], iris [5], palmprint [19], handwriting [33] ... Despite the fa ct that hand shape trait cannot furnish competitive matching performances compared to some other biometric modalities like iris and fingerprint, it is still considered as a good choice in terms of personal identification and verification process thanks to its user acceptance, contactless, simplicity of acquisition and low cost device properties.…”

Intensity Separation based Iris Recognition Method using Curvelets and PCA

“…Each coefficient within a given C mR/I is then assigned to its respective matrix f ∈ F to create the desired frame. Each resulting f is then mean adjusted to create a corresponding f using (5):…”

“…Statistical methods extract some features from the input iris image and compare these features directly by some sort of distance measures to assess and find its closest match. Example of these techniques include Principal Component Analysis (PCA) [3], Wavelet [4], Curvelet [5], Gabor filter [6], and non-orthogonal transform [7] based techniques. Learning based methods use one more statistical based feature extraction algorithms.…”

Personal recognition system using hand modality based on local features

“…Biometric technology is considered as the safest means for identity authentication. In fact, several biometric modalities have been studied and adopted such as fingerprint [8], face [9], [10], hand shape [12], iris [7], handwriting [25] etc. Despite the fact that hand geometry trait cannot furnish competitive matching performances compared to some different biometric modalities like iris and fingerprint, it is still considered as a good choice in terms of personal identification and verification process thanks to its user acceptance, contactless, simplicity of acquisition and low cost device properties.…”