Narrow Results

This paper proposes an intelligent 2ν-support vector machine based match score fusion algorithm to improve the performance of face and iris recognition by integrating the quality of images. The proposed algorithm applies redundant discrete wavelet transform to evaluate the underlying linear and non-linear features present in the image. A composite quality score is computed to determine the extent of smoothness, sharpness, noise, and other pertinent features present in each subband of the image. The match score and the corresponding quality score of an image are fused using 2ν-support vector machine to improve the verification performance. The proposed algorithm is experimentally validated using the FERET face database and the CASIA iris database. The verification performance and statistical evaluation show that the proposed algorithm outperforms existing fusion algorithms.

The brain activity observed on EEG electrodes is influenced by volume conduction and functional connectivity of a person performing a task. When the task is a biometric test the EEG signals represent the unique “brain print”, which is defined by the functional connectivity that is represented by the interactions between electrodes, whilst the conduction components cause trivial correlations. Orthogonalization using autoregressive modeling minimizes the conduction components, and then the residuals are related to features correlated with the functional connectivity. However, the orthogonalization can be unreliable for high-dimensional EEG data. We have found that the dimensionality can be significantly reduced if the baselines required for estimating the residuals can be modeled by using relevant electrodes. In our approach, the required models are learnt by a Group Method of Data Handling (GMDH) algorithm which we have made capable of discovering reliable models from multidimensional EEG data. In our experiments on the EEG-MMI benchmark data which include 109 participants, the proposed method has correctly identified all the subjects and provided a statistically significant ($p<0.01$) improvement of the identification accuracy. The experiments have shown that the proposed GMDH method can learn new features from multi-electrode EEG data, which are capable to improve the accuracy of biometric identification.

Face recognition is a rapidly growing research area due to it is contactless, most natural and most humanized. Many issues needed to do research exist in face recognition. A novel wavelet-based and SVM framework for face training and recognizing is proposed. FMT and PCA are also used for improving the accuracy and performance. In order for searching for the best performance, many experiments conducted with this method show that face recognition is largely affected by wavelet decomposing level, and at the same time by SVM type and kernel. So for high recognizing accuracy, proper methods and parameters are needed. The results prove that our scheme is an efficient method as 2D facial image recognition.

Cryptographic key is the central problem in cryptographic techniques. The key based on password protection is not enough secure because of the low entropy in user chosen passwords that can be exploited to launch password-guessing attacks. And the length of cryptographic key generated from user's biometric features directly is limited. Differ from prior methods, a novel scheme of cryptographic key generation based on biometric features and secret sharing is proposed. Biometric feature vectors is transformed through orthogonal matrix, and mapped to integer spaces. Then, a steady integer vector is obtained, and can be utilized to bind with cryptographic key by Shamir's secret sharing scheme. This method may realize the security of key storing, and can produce the key of the arbitrary length.

With the development of the technology of information security, iris recognition based on biometrics has become more and more important. In an iris recognition system, preprocessing, especially iris localization plays a very important role. So far, there are many iris localization algorithms having been proposed. In this paper, we propose an iris localization algorithm, in which we localize iris through detecting the edge points and improved integral differential operator and curve fitting. All the procedures of the algorithm are proved to be valid through our experiment on 648 iris images from CASIA (The Institute of Automation, Chinese Academy of Sciences) database.

User always needs to remember passwords in many systems. But a low-entropy password is easy to be attacked, and a high-entropy password is not easy to be remembered. A scheme of key generation from face features is proposed. This scheme is based on FCS (Fuzzy Commitment Scheme) and utilizes the error-correct encoding technique and Hash algorithm to hide random key into biometric features. It realized secure key reconstruction and authentication. This scheme not only protects the user's key but also prevents user's biometric features revealed effectively. The design is provably effective for biometric key generation and secure management through the experiment. The scheme is easy to extend to other biometrics since the extracting process of biometric features can not affect it.