Somnath Dey

Multimodal biometrics: state of the art in fusion techniques

The concept of multimodal biometrics is gaining importance to realise reliable, accurate and robust authentication systems. In multimodal biometrics, information fusion can be done at various levels: sensor level, feature level, match-score level, rank level and decision level. This paper surveys the issues of fusion at different levels and recent researches addressing these issues. Furthermore, there are several fusion rules available to combine biometric traits at different levels of fusion. We have reviewed the recent approaches of fusion rules. A comparison of the state-of-the-art fusion techniques has been made to comprehend their merits and demerits.

Iris Data Indexing Method Using Gabor Energy Features

Biometric features are extracted from a complex pattern and stored as high dimensional data. These data do not follow traditional sorting order like numerical and alphabetical data. Hence, a linear search method makes the identification process extremely slow as well as increases the false acceptance rate beyond an acceptable range. To address this problem, we propose an efficient indexing mechanism to retrieve iris biometric templates using Gabor energy features. The Gabor energy features are calculated from the preprocessed iris texture in different scales and orientations to generate a 12-dimensional index key for an iris template. An index space is created based on the values of index keys of all individuals. A candidate set is retrieved from the index space based on the values of query index key. Next, we rank the retrieved candidates according to their occurrences. If the identity of the query template is matched, then it is a hit, otherwise a miss. We have experimented our approach with Bath, CASIA-V3-Interval, CASIA-V4-Thousand, MMU2, and WVU iris databases. Our proposed approach gives 11.3%, 14.5%, 16.3%, 13.5%, and 10.3% penetration rates and 98.2%, 91.1%, 90.7%, 85.2%, and 96% hit rates for Bath, CASIA-V3-Interval, CASIA-V4-Thousand, MMU2, and WVU iris database, respectively, when we consider the retrieving templates up to the fifth rank. Experiments substantiate that our approach is capable of retrieving biometric data with a higher hit rate and lower penetration rate compared to the existing approaches. Application of Gabor energy features to index iris data proves to be effective for fast and accurate retrieval. With our proposed approach, it is possible to retrieve a set of iris templates similar to the query template in the order of milliseconds and is independent of sizes of databases.

Parameters effecting the predictive virtual keyboard

Text prediction is one of the most widely used technique to enhance the text entry rate in message composition system with virtual keyboard. Several text prediction techniques have been reported in recent literature. Most of them concentrate on prediction method and keystroke savings. It is observed that other than these issues, many factors influence the performance of text prediction. This paper discusses these different parameters. In addition to prediction method, we also examine organizational structure and fast retrieval of suggestion, factor such as User Interface and Human factors, such as orientation of prediction list, number of suggestions to be displayed, proper placing of prediction window, training and testing etc. Issues such as keyboard-adjacency and motor-coordination slip are also illustrated. In this paper we elaborate the proper placing of displaying prediction list for maximum benefit and its effect on searching the word in the predictive text entry system.

Fast and accurate personal identification based on iris biometric

In this paper, we propose an efficient and accurate technique to process iris biometric. Our approach handles degraded quality of eye images, occlusion of eyelids and eyelashes. Our approach considers non-circular pupil and iris boundary. With our approach it is possible to store an iris template with 304 bits only unlike 1024 bits used in majority of the existing approaches. Further, we assign different weights to different iris regions enabling us to compare two iris templates with a lesser number of features and thus achieving better accuracy. Experimental results on several iris databases establish the efficacy of our approach.

A privacy-preserving cancelable iris template generation scheme using decimal encoding and look-up table mapping

Biometric-based recognition systems have overcome passive issues of traditional human authentication systems. However, security theft and privacy invasion are two passive issues that still persist in the effective deployment of biometric-based authentication systems. Compromise of biometric data can potentially lead to serious security violation as the users biometric trait cannot be changed. In order to prevent the invasion of biometric templates, it is desired to morph the original biometric template through non-invertible or irreversible transformation function. This transformed template is referred to as cancelable template and can be replaced or reissued in case of compromise. In this paper, we propose a novel cancelable iris template generation technique based on randomized look-up table mapping. The technique utilizes a decimal vector generated from a rotation-invariant feature vector. The feature vector is generated using 1-D Log Gabor filter applied to the iris image. Experiments carried out on various iris databases confirm the efficacy of the proposed approach. After applying the template protection mechanism, we have achieved Equal Error Rate (EER) of 0.37%, 0.43% and 0.79% for CASIA-V 1.0, CASIA-V3-Interval and ICE 2005 iris databases, respectively. Moreover, the transformation preserves the irreversibility, revocability and diversity properties of the concealable iris templates.

An efficient approach to iris detection for iris biometric processing

Detection of iris in an eye image poses a number of challenges, such as inferior image quality, occlusion of eyelids, eyelashes etc. Owing to these problems, it is not possible to achieve 100% accuracy in any iris-based biometric authentication system. In this paper, we have proposed an approach to detect the iris boundary in an efficient and accurate way. Experimental results show that our approach is approximately 75% faster than the existing approaches. With our approach it is possible to detect the iris part 98% accurately as substantiated by our experiments on Bath, MMU and UBIRIS iris image databases.

Cancelable iris template generation using look-up table mapping

One of the potential passiveness in a biometric system is the invasion of stored biometric template, which may lead to serious security and privacy thefts. The emerging need of biometric approaches is evolved from privacy invasion and irrevocable issues of identity thefts as passwords and tokens can be easily compromised. To address these issues, the notion of cancelable biometrics is introduced to signify biometric templates that can be canceled and replaced with the inclusion of another independent authentication factor. Cancelable biometric generation technique based on randomized look-up table mapping has been proposed in this paper. The technique uses a decimal vector, which is evaluated from a row vector. Row vector is generated by applying 1-D Log Gabor filter on the raw iris template. The decimal vector is then mapped to look-up table based on the position of bits. The proposed approach enables an operation of cancelable iris biometric systems at a high security level. Experiments which are carried out on the CASIA V3 Interval iris database confirm the effectiveness of the proposed approach. Although, the accuracy of the proposed approach is confirmed as 94.26% but the approach is efficient if we consider the security and non-revocable perspectives.

An Efficient Approach for Pupil Detection in Iris Images

Iris-based biometric system is gaining its importance in several applications. However, existing methods of detecting boundary between pupil and iris, which is the first task in any iris-based biometric identification methods are computationally expensive. Further, existing methods are not able to detect pupil boundary accurately and hence leading to errors in identification process. In this paper, we address these two problems and propose a technique to detect pupil boundary efficiently and accurately. We propose scaling and power transform followed by edge detection and circle finding. Scaling reduces the search space significantly and power transform is helpful for image thresholding. Experiments on CASIA iris database reveal that with the proposed approach, we able to detect pupil almost 100% accurately.

Collaborative Approach for Data Integrity Verification in Cloud Computing

High security is one of leading restriction for shining up bright eras and vision of Cloud computing. In latest trend of Cloud, all the sensitive applications and data are moved towards cloud infrastructure and data center which run on virtual computing resources in the form of virtual machine. The large scale usage of virtualization to achieve cloud infrastructure brings additional security burden for tenants of a public cloud service. In this paper, we primarily aim to achieve better data integrity verification technique and help users to utilize Data as a Service (Daas) in Cloud computing. The experimental results are included in order to show the effectiveness of the proposed method for data integrity verification.

Development of iconic interface to retrieve information from Internet

With the fast growth of computer and technology, information are spread out all over the world. Information are gathered and maintained in internet in a fashion that anyone can get answer or solution to a question within a faction of second. But this facility is only limited among the privileged people those who are digitally literate. The main approach of this paper is to provide an interface for those digitally illiterate people, effective and efficient enough to seek information from the Internet.