Mohammad Imran

Multimodal Biometrics: Analysis of Handvein & Palmprint Combination Used for Person Verification

There is a global concern to implement accurate person verification in various facets of social and professional life. These include banking, travel and secure access to social security services. While biometrics have been deployed with various choices as face, finger print, etc the importance to higher levels of security have influenced two things. One is of finding newer and more universal biometrics and other of multimodal options. Recently, hand vein based person verification has attracted increased attention. The reason seems to be that hand vein patterns are unique, universal and invariant over time and extremely non intrusive. In this paper, we analyze hand vein biometric in unimodal status and also in combination with palm print in multimodal situation. One of the key aspects of this is extracting hand vein features. It is here that the standard edge detection masks yield poor result. We then propose using non standard edge mask in schemes to accurately extract the hand vein pattern which in turn is classified using Kernel Direct Discriminant Analysis (KDDA) to make the decision about accept/reject. The performance of the proposed non-standard edge masks are compared with conventional edge detection masks and statistical validation of the results are presented with 90% confidence interval. Robustness of such scheme is analyzed by evaluating these algorithms and schemes on data corrupted by noise. The final results show the efficacy of our schemes.

Multibiometric systems: A comparative study of multi-algorithmic and multimodal approaches

In this paper, we present a comparative analysis of Multi-algorithmic and Multimodal approaches. We have used palmprint and face as biometric traits and other popular subspace algorithms (PCA, FLD, and ICA). Subsequently, the different combinations of algorithms are also evaluated in our experiment. The multi-algorithmic approach achieves incremental results where as multimodal approach yields far improved results. Hence Complimentary information available through multimodal approach always performs better than multi-algorithmic approach which mainly builds on supplementary information.

Multimodal biometric fusion of face and palmprint at various levels

Recent years have witnessed researchers paying enormous attention to design efficient multi-modal biometric systems because of their ability to withstand spoof attacks. Single biometric sometimes fails to extract adequate information for verifying the identity of a person [7]. On the other hand, by combining multiple modalities, enhanced performance reliability could be achieved. In this paper, we have fused face and palmprint modalities at all levels of fusion viz sensor level, feature level, decision level and score level. For this purpose, we have selected modality specific feature extraction algorithms for face and palmprint such as LDA and LPQ respectively. Popular databases AR (for face) and PolyU (for Palmprint) were considered for evaluation purposes. Rigorous experiments were conducted both under clean and noisy conditions to ascertain robust level of fusion and impact of fusion strategies at various levels of fusion for these two modalities. Results are substantiated with appropriate analysis.

A New Hybrid Approach for Information Fusion in Multibiometric Systems

This paper proposes a new hybrid approach to verification aspect of a multibiometric system. This also gives a comparative analysis with traditional approaches such as multialgorithmic and multimodal versions of the same. For evaluating the performance we have considered different level of fusion with different fusion strategies for all the approaches, we have also calculated average EER to compare. Results from the experiments allow us to claim that the proposed hybrid approach performs better than traditional approaches and also consumes optimal number of modalities.

Multi-Algorithm Decision-Level Fusion Using Finger-Knuckle-Print Biometric

This paper proposed the use of multi-algorithm feature level fusion as a means to improve the performance of Finger Knuckle Print (FKP) verification. LG, LPQ, PCA, and LPP have been used to extract the FKP features. Experiments are performed using the FKP database, which consists of 7,920 images. Results indicate that the multi-algorithm verification approach outperforms higher performance than using any single algorithm. The biometric performance using feature level fusions under different normalization technique as well have been demonstrated in this paper.

Feature Extraction of Currency Notes: An Approach Based on Wavelet Transform

In this paper, we propose to extract the texture features of currency note images. To extract the features, first the Discrete Wavelet Transform (DWT) in particular Daubechies 1 (DB1) is utilized on a currency note and the approximate coefficient matrix of the transformed image is obtained. A set of coefficient statistical moments are then extracted from the approximate efficient matrix. The extracted features are stored in a feature vector. The extracted features can be used for recognition, classification and retrieval of currency notes.

Efficient multi-algorithmic approaches for face recognition using subspace methods

Face recognition is an active research in the field of biometrics due to its potential benefits to various security based applications. To make the results of face recognition unsusceptible to different kinds of variations in the image and to enhance the accuracy, robustness of two or more methods can be fused in a single framework. The fusion can be achieved at various levels. Objective of this paper is to suggest optimal fusion of subspace methods to achieve robust results for various test conditions. This is achieved by performing feature level fusion of popular subspace methods namely PCA, LDA, LPP and ICA1. The Fusion is performed by considering different combinations of set of two, three and four subspace methods. Experiments are conducted by using two different databases: ORL and Yale. Experimental results suggest that by the fusion of these subspace approaches; there is a significant improvement in the accuracy compared to performance of an individual subspace method. This work helped us to determine the optimal combination of subspace methods to achieve robust results for specific test conditions.

Multi-modal biometric system on various levels of fusion using LPQ features

Abstract In the proposed multimodal biometric verification system, the system is implemented on all levels of fusion strategies (i) fusion prior to matching (sensor level and feature level) and (ii) fusion post matching (score level and decision level), a binding required and a deserving step that outputs a reliable and robust biometric identification and verification systems. We have chosen benchmark databases for our experimentation and considered physiological modalities such as face, palmprint, finger knuckle print, handvein. The performance measures considered here are FAR (False Acceptance Rate) and FRR (False Rejection Rate). Extracting texture features from a well-known texture operator-LPQ (local phase quantization), we have performed sensor level fusion adopting HAAR wavelets, feature level fusion using Z-Score normalization, score level fusion employing simple sum rule and decision level fusion with AND rule. For the implemented biometric recognition system, score level fusion strategy outer performs than the other fusion techniques in terms of EER, yielding good verification rate on all benchmark threshold values (0.01%, 0.1%, 1%), with the GAR=100% at 1% FAR. The tabulated results of the experiments are visualized by BAR chart

Robust multimodal biometric verification system based on face and fingerprint

In this paper, we have proposed a multimodal biometric verification system adopting the physiological traits-face and fingerprint. The system has been evaluated for robustness analysis imposing the Additive White Gaussian noise (AWGN) on clean data of the employed databases-AR facial database, PolyU High-resolution fingerprint database. The unimodal and multimodal (pre and post matching fusion strategies) verification system is investigated extracting the log-gabor features. Considering the performance measures such as GAR (Genuine Acceptance Rate) and FAR (False Acceptance Rate) at the benchmark threshold values — 0.01%, 0.1%, 1%. The accuracy of the system for clean and noisy data can be visualized by ROC (Receiver Operating Characteristics) curve, area under it expresses the performance of the implemented system.

Combination of physiological and behavioral modalities based on score level fusion

The impulse of ubiquitous biometrics may refer to the unique identification of an individual by analyzing psychological and behavioral traits. We have employed psychological biometric modalities such as AR database for face trait, poly-U database for palm trait and behavioral biometric modalities such as MCYT-100 database for signature trait, TIMIT database for speech trait. Desirable features extraction algorithms are employed, on both psychological and behavioral traits and fused at score level. The each of scores from the matcher from respective modalities fused into a single score by adopting sum rule. The proposed system evaluated using the performance measures of standard biometric verification GAR (Genuine Acceptance Rate), FAR (False Acceptance Rate), and the obtained results analyzed graphically using ROC (Receiver Operating Characteristics). We have motivated in conducting the experiments on fusion of both psychological and behavioral traits and observed how the verification rate increases with the increase in fusion of more than one modalities.