International Journal of Biometrics (7 papers in press)
- A Secure Palmprint Authentication System using Chaotic Mixing and Watermarking
by Munaga V N K Prasad, Ilaiah Kavati
Abstract: In recent years, there has been a significant increase in the use of biometrics for user authentication. The wide spread use of biometrics in real life applications for user authentication raised the security and privacy concerns of the biometric data. In this paper, a biometric image protection scheme exploiting the properties of chaotic mixing and watermarking technique is proposed. We also proposed an efficient technique for biometric template protection using a random vector. A detailed discussion on the security of the proposed techniques are given. The effectiveness of the proposed method is investigated with application to palmprint authentication. The experimental results evaluated on PolyU and IITD database shows that the proposed system achieves better recognition accuracy and lower error rates while providing protection to the biometric data.
Keywords: Image security; Template security; Palmprint; Chaotic mixing, Watermarking; Random vector
- Person tracking and segmentation for human gait biometric system
by Anup Nandy, Pavan Chakraborty, G.C. Nandi
Abstract: Person tracking and segmentation in an unstructured environment provides an increasing demand to solve human identification problems. This paper addresses mixture of Gaussian (MoG) technique for statistically background modelling and robust human tracking method for deriving an intrinsic gait signature. The front and back leg angles are calculated from the sequence of extracted human motion silhouette frames which are being used as gait features. The training gait database is made with these extracted gait features for ten different training subjects. The principal component analysis (PCA) is applied on derived gait signatures which transforms the input features into a low dimensional feature space. The classification technique is followed by Baye's decision rule coupled with multivariate Gaussian distribution. The results are compared with k-nearest neighbour rule and minimum distance classification (MDC) techniques by accuracy and computational cost metric. The experimental verification has been performed on CASIA standard gait database. The Baye's classifier produces an encouraging classification result with minimum misclassification error rate.
Keywords: human gait; mixture of Gaussian; MoG; segmentation; person tracking; silhouette extraction; front knee angle; back knee angle; Baye's decision rule; k-nearest neighbour; minimum distance classifier; biometrics; human identification; modelling; feature extraction; gait features; principal component analysis; PCA; gait signatures; classification.
- Person recognition using alternative hand geometry
by Asish Bera, Debotosh Bhattacharjee, Mita Nasipuri
Abstract: In this paper, a new approach for user recognition is presented, which is based on the geometric features from either left or right hand images. The hand images are collected at unconstrained pose environment. Image normalisation is applied at the preprocessing stage. Features are extracted from the normalised images, which are mainly comprised of lengths and widths at different positions of the fingers. A simple classification algorithm has been implemented that is primarily dependent on the ratio of modified minimum distance and number of features, which are matched within a distance threshold. Experimental results of identification and verification are quite acceptable, producing 98.8% identification and 99.6% verification (at 0.55% FAR) of 253 standard subjects which are a blend of both left and right hand images.
Keywords: biometrics; hand geometry; human identification; verification; user recognition; geometric features; left hand images; right hand images; image normalisation; normalised images; finger positions; classification; feature extraction.
- On the relevance of using rhythmic metrics and SVM to assess dysarthric severity
by Habiba Dahmani, Sid-Ahmed Selouani, Noureddine Doghmane, Douglas O'Shaughnessy, Mohamed Chetouani
Abstract: Studies of dysarthric speech rhythm have explored the possibility of distinguishing healthy speakers from dysarthric ones. These studies also allowed the detection of different types of dysarthria. The present paper aims at assessing the ability of rhythm metrics to perceive dysarthric severity levels. The study reports on the results of a statistical acoustic investigation using various rhythmic metrics. Among these rhythm features, we propose a new rhythm metric based on an approximation of the speakers' rate of articulation. The investigation was carried out on the speech data of US dysarthric patients recorded on the Nemours corpus. The rhythm features are based on two types of segmentation: vocalic/consonantal and voiced/unvoiced interval durations. Results of different classification experiments show that the rhythm-based measures can be used effectively to characterise the dysarthric severity by classifying speakers into their respective categories. Support vector machine classification method has been successfully used to perform the assessment of the dysarthria severity level.
Keywords: dysarthria; rhythm; pairwise variability index; acoustical analysis; Nemours database; support vector machines; SVM; classification; discriminant analysis; rhythmic metrics; dysarthric severity levels; dysarthric speech rhythms.
- Fingerprint indexing using minutiae-based invariable set of multidimensional features
by Om Prakash Singh, Somnath Dey, Debasis Samanta
Abstract: In fingerprint identification, exhaustive search demands a huge response time for large database and hence impractical in many real-life applications. To alleviate this limitation, researchers advocate indexing technique to narrow down the search space. In this work, we investigate three different indexing techniques (linear, clustered and clustered kd-tree) with invariable set of features for a fingerprint identification system. In our approach, we consider local topology of minutiae using two closest points triangle for index key generation. The features are invariant to rotation and scaling and hence, the approach can deal with fingerprints form different devices and sensors. The proposed approach has been tested on NIST DB4 and FVC 2004 databases. Experimental results substantiate the error rate of 0.35%, 1.5% and 2.45% at penetration rate 15% in NIST DB4 for linear search, clustered search and clustered kd-tree search, respectively. For FVC 2004 databases, we attain 0%, 1.36% and 5.45% for FVC2004 DB1, 0%, 2.73% and 4.09% for FVC2004 DB2, 2.27%, 5.0% and 5.91% for FVC2004 DB3 and 0%, 1.36% and 5.0% for FVC2004 DB4 when penetration rate is 15.45% in linear, cluster and clustered kd-tree searches, respectively. The result is indeed comparable to the existing approaches reported in the recent literature.
Keywords: biometrics; fingerprint identification; fingerprint indexing; biometric data indexing; data clustering; minutiae feature extraction; index key generation; multidimensional features; fingerprints.
- Face recognition in colour JPEG compressed domain
by Alireza Sepas-Moghaddam, Mohammad-Shahram Moin
Abstract: In spite of the positive role of colour features in pixel domain face recognition systems, previous recognition approaches in JPEG compressed domain have been proposed in grey space. In this study, for the first time, we investigated the effects of JPEG compressed colour features on the efficiency of the state-of-the-art approaches in this domain. To achieve this, a comprehensive set of experiments have been conducted on colour FEI facial database. PCA and ICA methods have been used for extracting feature vectors and reducing the dimension of the extracted features. The results in colour space have been compared with those in grey space and showed the benefit of using the colour features in JPEG compressed domain face recognition in terms of recognition performance. It has also been observed an increase in the class separation and discriminatory power using colour compressed domain features over features extracted in grey space.
Keywords: face recognition; JPEG compressed domain; colour features; colour face database; discriminatory power; biometrics; PCA; ICA; principal component analysis; independent component analysis; feature extraction.
- Integration Framework for Assembling Components of a Face Recognition System
by Monpraon Sukroongreung, Pratit Santiprabhob
Abstract: This paper outlines a framework that has been developed to assist in integrating components of a given face recognition system. By applying this framework, performance and restrictions of each component can be explicitly observed. Recognition performance of the resulting face recognition system can be clearly compared with components are replaced by other alternatives. The contribution of our work is a generalized framework for components integration of a face recognition system that facilitates (i) examination of the effects of varying techniques and components to the recognition performance; and (ii) determination of suitable techniques, components and system configurations as per a given problem context. We have applied our framework to assemble different alternatives of a face recognitions system. Such applications reveal the effects when different components are included into a face recognition system and have shown insights into the relationship between dataset quality, techniques selected and recognition performance. This paper also outlines how the proposed framework can be applied to other biometric recognition systems.
Keywords: integration frameworks, component selections, biometrics