International Journal of Biometrics (4 papers in press)
- A Non-Invertible Cancellable Fingerprint Construct based on Compact Minutiae Patterns
by Vedrana Krivokuca, Waleed Abdulla, Akshya Swain
Abstract: This paper proposes a new, privacy-preserving fingerprint construct, which consists of a single pattern created from a small subset of minutiae from the corresponding minutiae template. The sparsity of the resulting feature vector ensures that it cannot be used to reconstruct the underlying fingerprint image, while simultaneously allowing for the cancellability of compromised patterns. The proposed construct is, therefore, inherently a non-invertible cancellable fingerprint template protection scheme. Experimental results on the recognition accuracy of patterns consisting of three, four, and five minutiae were very encouraging: an FRR of 0.4% or less was achieved for all three pattern sizes, while the FAR ranged from 2.1% for 5-minutia patterns to 6.5% for 3-minutia patterns. It took about 0.1 seconds to verify a single person, regardless of the pattern size. These results suggest that the proposed fingerprint construct is suitable for deployment in fingerprint-based civilian authentication applications.
Keywords: biometrics; fingerprints; minutiae; minutiae template; minutiae patterns; minutiae structures; fingerprint representation; fingerprint template protection; fingerprint image reconstruction; cancellable biometrics; non invertible; fingerprint matching; verification; authentication; recognition.
- Fingerprint Quality Assessment Based On Wave atoms Transform
by Leila Boutella, Amina Serir
Abstract: Fingerprint image quality is of great importance for an automatic fingerprint identification system (AFIS), and affects its performance. In this paper, a new fingerprint image quality measure based on Wave atoms transform is proposed. Indeed, fingerprint features are extracted by exploiting Wave atoms multiscale and multi-directional properties. Hence, for good quality block, the singularities are concentrated in a same directional subbands. In the opposite, for poor quality, the singularities are scattered over several directional subbands. In order to evaluate the performance of the algorithm, FVC2002 databases have been considered. The results show that this method has a serious potential in fingerprint quality evaluation and will improve the performance and effectiveness of AFIS.
Keywords: Fingerprint quality assessment, Wave atoms transform, Multi-directional analysis.
- Functional Data Analysis in the Use of Eyebrow Shape as a Biometric Indicator in Face Recognition
by Midori Albert, Cuixian Chen, Yishi Wang, Yaw Chang
Abstract: This paper reports the use of eyebrow shape as a point feature for face recognition in the identity sciences. An approach to quantifying eyebrow shape and results of an experiment to test how human perceptions of eyebrow shape (qualitative analyses) compare with quantitative analyses (i.e., computer generated algorithms) of eyebrow shape are discussed. The aim is to develop a method for face identification using a point feature, such as the eyebrow, inasmuch as face images used in forensic face image comparisons may be obscured due to sunglasses or head coverings, or indiscernible due to pose or lighting issues. Results showed that functional data analysis was successful in interpreting eyebrow shape from digitized face images, and that computer-classified (i.e., quantitative analyses) eyebrow shapes were more reliable than human perception (i.e., qualitative analyses) as a relatively high level of human subjectivity was evident from findings of a two-trial experiment on eyebrow classification.
Keywords: Functional Data Analysis; Face Recognition; Eyebrow Shape; Biometric Indicator; Biometric; Identity Sciences; Point Feature
- Personal Identification using Local Gaussian Quadrature Filter Pair Phase Quantization of Hand Vein Images
by Anantha kumar T, Premalatha K
Abstract: Hand vein pattern is a biometric feature in which the actual pattern is the shape of vein network and its characteristics are the vein features. This paper proposes a new approach uses local phase quantization with Gaussian quadrature filter pair for hand dorsal vein identification. The proposed work extracts the phase information computed locally in a window for every pixel position by employing Gaussian quadrature filter pair. The phases of six frequency coefficients are quantized and it is used to form a descriptor code for the local region. Whitening transformation is used to decorrelate these local descriptors and a histogram is generated for every pixel which describes the local pattern. The proposed work is experimented with minimum distance classifiers and results are examined for recognition rate, False Acceptance Rate (FAR), False Rejection Rate (FRR) and Equal Error Rate (EER). Experiment results show that the proposed work outperforms the existing methods.
Keywords: Dorsal hand vein; Gaussian filter; Phase quantization; Decorrelation; Whitening transformation.