International Journal of Biometrics http://www.inderscience.com/browse/index.php?journalID=285&year=2013&vol=5&issue=2 Inderscience Publishers Ltd en-uk support@inderscience.com International Journal of Biometrics 1755-8301 1755-831X © 2013 Inderscience Enterprises Ltd. © 2013 Inderscience Publishers Ltd editor@inderscience.com https://www.inderscience.com/images/files/coverImgs/ijbm_scoverijbm.jpg http://www.inderscience.com/browse/index.php?journalID=285&year=2013&vol=5&issue=2 http://www.inderscience.com/link.php?id=52943 While external body parts such as the face, fingerprints, or retina are often used for biometric identification, it can also be reasonably assumed that internal organs imaged with biomedical imaging devices can also allow biometric identification. Here we studied the use of MRI images for the purpose of biometric identification, and show that the accuracy of person identification using knee MRIs is significantly higher than random. Knee MRI images of 2,686 different patients were used in the experiment, and analysed using the wndchrm image classification scheme. Experimental results show that the rank-10 identification accuracy using the MRI knee images is ∼93% for a dataset of 100 individuals, and ∼45% for the entire dataset of 2,686 persons. Since MRI is used for the purpose of imaging internal parts of the body, this approach of biometric identification can potentially offer high resistance to deception. MRI-based knee image for personal identification
Lior Shamir
International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 113 - 125
While external body parts such as the face, fingerprints, or retina are often used for biometric identification, it can also be reasonably assumed that internal organs imaged with biomedical imaging devices can also allow biometric identification. Here we studied the use of MRI images for the purpose of biometric identification, and show that the accuracy of person identification using knee MRIs is significantly higher than random. Knee MRI images of 2,686 different patients were used in the experiment, and analysed using the wndchrm image classification scheme. Experimental results show that the rank-10 identification accuracy using the MRI knee images is ∼93% for a dataset of 100 individuals, and ∼45% for the entire dataset of 2,686 persons. Since MRI is used for the purpose of imaging internal parts of the body, this approach of biometric identification can potentially offer high resistance to deception.

]]> 10.1504/IJBM.2013.052943 International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 113 - 125 Lior Shamir Department of Computer Science, Lawrence Technological University, 21000 W Ten Mile Rd., Southfield, MI 48075, USA MRI images magnetic resonance imaging image analysis internal biometrics knee images medical imaging personal identification biometric identification 2013-03-31T23:20:50-05:00 © 2013 Inderscience Enterprises Ltd. 5 2 113 125 2013-03-31T23:20:50-05:00 http://www.inderscience.com/link.php?id=52963 The purpose of this study was to study the latent structure of the cephalofacial variables and to explain the clustering of these measurements. Eleven cephalofacial measurements were done on 561 male entities aged 18-35 years. The collected data are analysed through descriptive parameters, correlation analysis, and factorial analyses. Correlation analysis confirms three sets of variables that realise the higher correlations with each other. According to the principal component analysis of the factor analysis three latent factors have been extracted that explain 53.3% of the total variance: the neurocranial volume factor; the transversal factor of the face; and the longitudinal factor of the face. The reduction of multiple measured manifest variables in a small number of latent factors may facilitate the work of anthropologists in studying and explaining of the genetic basis of development of bones, cartilages and soft tissues of the head and their morphological construction. Clustering of the cephalofacial measurements
Agron M. Rexhepi; Behlul Brestovci
International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 126 - 136
The purpose of this study was to study the latent structure of the cephalofacial variables and to explain the clustering of these measurements. Eleven cephalofacial measurements were done on 561 male entities aged 18-35 years. The collected data are analysed through descriptive parameters, correlation analysis, and factorial analyses. Correlation analysis confirms three sets of variables that realise the higher correlations with each other. According to the principal component analysis of the factor analysis three latent factors have been extracted that explain 53.3% of the total variance: the neurocranial volume factor; the transversal factor of the face; and the longitudinal factor of the face. The reduction of multiple measured manifest variables in a small number of latent factors may facilitate the work of anthropologists in studying and explaining of the genetic basis of development of bones, cartilages and soft tissues of the head and their morphological construction.

]]> 10.1504/IJBM.2013.052963 International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 126 - 136 Agron M. Rexhepi; Behlul Brestovci Sport Center for Increasing of Morpho-functional Abilities, Fitness and Nutrition 'Corpore Sano', Str. 'Sali Butka' Nr. 31/D, 10020 Prishtina, Kosovo; University College for Sport Education 'Universi', 10000 Prishtina, Kosovo; Institute of Sports Anthropology, Str. 'Sali Butka' Nr. 31/D, 10020 Prishtina, Kosovo ' Faculty of Special Education and Rehabilitation, University of Zagreb, Kušlanova 59a., 10000 Zagreb, Croatia anthropometry cephalofacial measurements correlation analysis factor analysis neurocranium clustering principal component analysis PCA medical anthropology cephalofacial variables human face human head cephalometry human skull genetic development 2013-03-31T23:20:50-05:00 © 2013 Inderscience Enterprises Ltd. 5 2 126 136 2013-03-31T23:20:50-05:00 http://www.inderscience.com/link.php?id=52964 The interest in biometric technology is received much attention in the recent years. However, the security issue still persists the main challenge for the reliable functioning of biometric-based authentication systems. Much has been reported on the vulnerabilities of biometric systems that breach the security and user privacy. We present a high-level classification of biometric system vulnerabilities and discuss the defence techniques of these vulnerabilities. We present a multidimensional threat environment of the biometric systems that includes faults, failures and security attacks. A framework of biometric security attacks on man-machine model is presented and the system vulnerabilities are represented using Ishikawa's diagram. The provable defence techniques such as biometric vitality detection and biometric template protection are critically evaluated, in particular, a classification of current state-of-the-art of vitality detection techniques of commonly used biometrics is proposed. Our main contributions include: (1) propose a taxonomy of biometric system vulnerabilities; (2) present a framework of biometric security attacks using man-machine model; (3) representation of vulnerabilities using Ishikawa's diagram; (4) an evaluation of defence techniques of these vulnerabilities. A taxonomy of biometric system vulnerabilities and defences
Yogendra Narain Singh; Sanjay Kumar Singh
International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 137 - 159
The interest in biometric technology is received much attention in the recent years. However, the security issue still persists the main challenge for the reliable functioning of biometric-based authentication systems. Much has been reported on the vulnerabilities of biometric systems that breach the security and user privacy. We present a high-level classification of biometric system vulnerabilities and discuss the defence techniques of these vulnerabilities. We present a multidimensional threat environment of the biometric systems that includes faults, failures and security attacks. A framework of biometric security attacks on man-machine model is presented and the system vulnerabilities are represented using Ishikawa's diagram. The provable defence techniques such as biometric vitality detection and biometric template protection are critically evaluated, in particular, a classification of current state-of-the-art of vitality detection techniques of commonly used biometrics is proposed. Our main contributions include: (1) propose a taxonomy of biometric system vulnerabilities; (2) present a framework of biometric security attacks using man-machine model; (3) representation of vulnerabilities using Ishikawa's diagram; (4) an evaluation of defence techniques of these vulnerabilities.

]]> 10.1504/IJBM.2013.052964 International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 137 - 159 Yogendra Narain Singh; Sanjay Kumar Singh Department of Computer Science and Engineering, Institute of Engineering and Technology, Gautam Buddh Technical University, Lucknow 226 021, India ' Department of Computer Engineering, Indian Institute of Technology (BHU), Varanasi 221 005, India biometrics authentication vulnerabilities vitality measures template protection security defences 2013-03-31T23:20:50-05:00 © 2013 Inderscience Enterprises Ltd. 5 2 137 159 2013-03-31T23:20:50-05:00 http://www.inderscience.com/link.php?id=52965 This article looks into pros and cons of the conventional global and local feature matching techniques for iris. The review of related research works on matching techniques leads to the observation that local features like scale invariant feature transform (SIFT) gives satisfactory recognition accuracy for good quality images. However the performance degrades when the images are occluded or taken non-cooperatively. As SIFT matches keypoints on the basis of 128-D local descriptors, hence it sometimes falsely pairs two keypoints which are from different portions of two iris images. Subsequently the need for filtering or pruning of faulty SIFT pairs is felt. The paper proposes two methods of filtering impairments (faulty pairs) based on the knowledge of spatial information of the keypoints. The two proposed pruning algorithms (angular filtering and scale filtering) are applied separately and applied in union to have a complete comparative analysis of the result of matching. The pruning approaches has given better recognition accuracy than conventional SIFT when experimented on two publicly available BATH and CASIAv3 iris databases. Postmatch pruning of SIFT pairs for iris recognition
Sambit Bakshi; Hunny Mehrotra; Banshidhar Majhi
International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 160 - 180
This article looks into pros and cons of the conventional global and local feature matching techniques for iris. The review of related research works on matching techniques leads to the observation that local features like scale invariant feature transform (SIFT) gives satisfactory recognition accuracy for good quality images. However the performance degrades when the images are occluded or taken non-cooperatively. As SIFT matches keypoints on the basis of 128-D local descriptors, hence it sometimes falsely pairs two keypoints which are from different portions of two iris images. Subsequently the need for filtering or pruning of faulty SIFT pairs is felt. The paper proposes two methods of filtering impairments (faulty pairs) based on the knowledge of spatial information of the keypoints. The two proposed pruning algorithms (angular filtering and scale filtering) are applied separately and applied in union to have a complete comparative analysis of the result of matching. The pruning approaches has given better recognition accuracy than conventional SIFT when experimented on two publicly available BATH and CASIAv3 iris databases.

]]> 10.1504/IJBM.2013.052965 International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 160 - 180 Sambit Bakshi; Hunny Mehrotra; Banshidhar Majhi Department of Computer Science and Engineering, National Institute of Technology Rourkela, Rourkela 769008, Odisha, India ' Department of Computer Science and Engineering, National Institute of Technology Rourkela, Rourkela 769008, Odisha, India ' Department of Computer Science and Engineering, National Institute of Technology Rourkela, Rourkela 769008, Odisha, India scale invariant feature transform SIFT local feature matching iris matching pruning impairments biometrics iris recognition iris images image matching image analysis filtering impairments recognition accuracy 2013-03-31T23:20:50-05:00 © 2013 Inderscience Enterprises Ltd. 5 2 160 180 2013-03-31T23:20:50-05:00 http://www.inderscience.com/link.php?id=52972 Poor quality images can affect sclera recognition accuracy. An image quality measure can help improve the recognition system performance. In this paper, we proposed a comprehensive approach for sclera image quality measure, which includes quality filter and quantitative quality assessment unit, segmentation evaluation unit, feature evaluation unit, and score fusion unit. The experimental results show that the combination score is highly correlated with the sclera recognition accuracy and can be used to improve and predict the performance of sclera recognition systems. A comprehensive approach for sclera image quality measure
Zhi Zhou; Eliza Y. Du; N. Luke Thomas; Edward J. Delp
International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 181 - 198
Poor quality images can affect sclera recognition accuracy. An image quality measure can help improve the recognition system performance. In this paper, we proposed a comprehensive approach for sclera image quality measure, which includes quality filter and quantitative quality assessment unit, segmentation evaluation unit, feature evaluation unit, and score fusion unit. The experimental results show that the combination score is highly correlated with the sclera recognition accuracy and can be used to improve and predict the performance of sclera recognition systems.

]]> 10.1504/IJBM.2013.052972 International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 181 - 198 Zhi Zhou; Eliza Y. Du; N. Luke Thomas; Edward J. Delp Department of Electrical and Computer Engineering, Purdue School of Engineering and Technology, Indiana University-Purdue University Indianapolis, Indianapolis, IN 46214, USA ' Department of Electrical and Computer Engineering, Purdue School of Engineering and Technology, Indiana University-Purdue University Indianapolis, Indianapolis, IN 46214, USA ' Department of Electrical and Computer Engineering, Purdue School of Engineering and Technology, Indiana University-Purdue University Indianapolis, Indianapolis, IN 46214, USA ' School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907, USA sclera quality measures sclera recognition feature information segmentation evaluation score fusion image quality biometric recognition biometrics blood vessels white of the eye 2013-03-31T23:20:50-05:00 © 2013 Inderscience Enterprises Ltd. 5 2 181 198 2013-03-31T23:20:50-05:00 http://www.inderscience.com/link.php?id=52977 The aim of this study was to research the relationships between set of body variables and set of cephalofacial variables, using statistical method of canonical analysis. The measurements were done on 288 adult male entities, aged 18-35 years old. Correlation analysis indicates mediocre inter-correlation among two sets of variables. Canonical analysis has enabled the extraction of two significant canonical roots for each set of variables. While left canonical roots explain 48.9% of total variance computed for the set of body variables, the right canonical roots explain 52.8% of the set of cephalofacial variables. Reduction of the measured variables in a small number of canonical roots may facilitate the work of the physical anthropologists in studying and explaining the genetic development of the bones, cartilages and soft tissues of humans, as well as their morphological construction. The relations between cephalofacial and body measurements measured on adult males of Kosovo Albanian population
Agron M. Rexhepi; Behlul Brestovci; Atif K.F. Idris
International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 199 - 210
The aim of this study was to research the relationships between set of body variables and set of cephalofacial variables, using statistical method of canonical analysis. The measurements were done on 288 adult male entities, aged 18-35 years old. Correlation analysis indicates mediocre inter-correlation among two sets of variables. Canonical analysis has enabled the extraction of two significant canonical roots for each set of variables. While left canonical roots explain 48.9% of total variance computed for the set of body variables, the right canonical roots explain 52.8% of the set of cephalofacial variables. Reduction of the measured variables in a small number of canonical roots may facilitate the work of the physical anthropologists in studying and explaining the genetic development of the bones, cartilages and soft tissues of humans, as well as their morphological construction.

]]> 10.1504/IJBM.2013.052977 International Journal of Biometrics, Vol. 5, No. 2 (2013) pp. 199 - 210 Agron M. Rexhepi; Behlul Brestovci; Atif K.F. Idris Sport Center for Increasing of Morpho-functional Abilities, Fitness and Nutrition 'Corpore Sano', 10000 Pristina, Kosovo, Serbia; University College for Sport Education 'Universi', 10000 Prishtina, Kosovo, Serbia; Institute of Sports Anthropology, 10000 Prishtina, Kosovo, Serbia ' Faculty of Special Education and Rehabilitation, University of Zagreb, 10000 Zagreb, Croatia; Institute of Sports Anthropology, 10000 Prishtina, Kosovo, Serbia ' Institute of Sports Anthropology, 10000 Prishtina, Kosovo, Serbia morphometry cephalofacial variables human body correlation analysis Albania cephalofacial measurements body measurements adult males Kosovo Albanians anthropometry neurocranium clustering medical anthropology human face human head cephalometry human skull genetic development 2013-03-31T23:20:50-05:00 © 2013 Inderscience Enterprises Ltd. 5 2 199 210 2013-03-31T23:20:50-05:00