Biometrics
Faculty: Mayank Vatsa, Richa Singh
Biometric research group has worked on several interesting and challenging issues of biometrics including face recognition with disguise and aging, fingerprint recognition using level-3 features, iris recognition using non-ideal images, and adaptive multimodal biometric fusion. The group now plans to work on biometrics for uncontrolled environment with non-cooperative users. Some of the research projects we are currently working on include recognizing scanned images with age difference and matching sketches with a database of face images. Information collected from multiple biometric modalities can provide uncertain, fuzzy and conflicting information which lead to reduction in the performance of fusion algorithms. We are designing adaptive and computationally efficient fusion algorithms for improving the performance in such instances. The group is also working on addressing challenges of the forensic community such as recognizing latent fingerprints using extended fingerprint feature set and video authentication.
Selected Recent Publications:
• R. Singh, M. Vatsa, and A. Noore, Face Recognition with Disguise and Single Gallery Images, Image and Vision Computing - Special Issue on Multimodal Biometrics, Vol. 27, No. 3, pp. 245-257, 2009.
• M. Vatsa, R. Singh, and A. Noore, Feature based RDWT Watermarking for Multimodal Biometric System, Image and Vision Computing - Special Issue on Multimodal Biometrics, Vol. 27, No. 3, pp. 293-304, 2009.
• M. Vatsa, R. Singh, and A. Noore, Unification of Evidence Theoretic Fusion Algorithms: A Case Study in Level-2 and Level-3 Fingerprint Features, IEEE Transaction on Systems, Man and Cybernetics-A, Vol. 39, No. 1, pp. 47-56, 2009.
• R. Singh, M. Vatsa, and A. Noore, Integrated Multilevel Image Fusion and Match Score Fusion of Visible and Infrared Face Images for Robust Face Recognition, Pattern Recognition - Special Issue on Multimodal Biometrics, Vol. 41, No. 3, pp. 880-893, 2008.
• M. Vatsa, R. Singh, and A. Noore, Improving Iris Recognition Performance using Segmentation, Quality Enhancement, Match Score Fusion and Indexing, IEEE Transaction on Systems, Man and Cybernetics-B, Vol. 38, No. 4, pp. 1021-1035, 2008.
Data Mining and Management
Faculty: Anirban Mondal, Vikram Goyal, Ashish Sureka
The main research areas of the data management research group are data management for P2P and emerging mobile-P2P applications, data mining and analytics, multi-dimensional and large-scale data indexing, data privacy and data-centric business intelligence applications. In the mobile-P2P domain, our focus is on performing effective data management in Mobile-P2P application scenarios and creating mobile-P2P-style applications such as a futuristic mobile eBay trading platform. In data mining and analytics, we work on predictive analytics, software engineering data mining and patent data mining. In the realm of multi-dimensional and large-scale data indexing, we are researching new indexing techniques for complex multi-dimensional database queries. In the data privacy area, we will work on designing efficient algorithms to determine probable attackers in post-violation scenarios, the design of audit query languages as well as the design of audit trail repositories. In the business intelligence area, we will focus on developing new techniques for effective information integration with the aim of facilitating timely and effective business decision-making.
Selected Recent Publications
• A. Mondal, S.K. Madria and M. Kitsuregawa “An Economic Incentive Model for encouraging Peer Collaboration in Mobile-P2P networks with support for constraint queries” International Journal of Peer-to-Peer Networking and Applications (PPNA), 2009 (To appear).
• D. Zhang, Y.M. Chee, A. Mondal, A. Tung and M. Kitsuregawa “Keyword search in spatial databases” Proceedings of the International Conference on Data Engineering (ICDE), 2009 (To appear).
• A. Sureka and P.P. Mirajkar “An Empirical Study on the Effect of Different Similarity Measures on User-Based Collaborative Filtering Algorithms” Proceedings of the Pacific Rim International Conference on Artificial Intelligence (PRICAI), 2008.
• A. Sureka and K.I. Varma “Using Genetic Algorithms for Parameter Optimization in Building Predictive Data Mining Models” Proceedings of the International Conference on Advanced Data Mining and Applications (ADMA), 2008.
• V.Goyal, S.K. Gupta and A. Gupta “A Unified Audit Expression Model for Auditing SQL Queries” Proceedings of the 22nd Annual IFIP WG 11.3 Working Conference on Data and Applications Security (DAS), 2008.
• S.K. Gupta, V. Goyal and A. Gupta “Precomputation of Privacy Policy Parameters for Auditing SQL Queries” Proceedings of the International conference on Ubiquitous information management and communication (ICUIMC), 2008.
Mobile and Ubiquitous Computing
Faculty: Pushpendra Singh, Anirban Mondal
The group focuses on two complementary aspects, namely (a) system design aspects of mobile systems and (b) the design of economic incentive models for enticing user participation in mobile-P2P networks. In case of mobile applications design, our research is directed towards reducing the complexity of developing applications and services for dynamic pervasive systems with emphasis on overcoming constraints associated with device heterogeneity and underlying network characteristics. We also work on designing middleware solutions, which can provide seamless access to networked contents over hybrid networks, while accounting for the resource constraints of wireless environments. We will also develop tools to prototype and evaluate pervasive computing environments to assess their suitability early-on. The design of economic incentive models for Mobile-P2P networks is motivated by the fact that mobile-P2P data availability is typically low due to frequent network partitioning and rampant free-riding. Hence, we also focus on economic incentive models for improving data availability and user participation in Mobile-P2P environments. In particular, we focus on economic models such as a bid-based brokerage model, a lease model and a peer-group-based model. We also consider mobile replication techniques in conjunction with our economic models to further improve data availability.
Selected Recent Publications:
• A.W. Guo, P. Blythe, P.L. Olivier, P. Singh, and H.N. Ha “Using Immersive Video to Evaluate Future Traveller Information System” IET journal of Intelligent Transport Systems, Vol. 2, 2008.
• A.Mondal, S.K. Madria and M. Kitsuregawa “An Economic Incentive Model for encouraging Peer Collaboration in Mobile-P2P networks with Support for Constraint Queries” International Journal of Peer-to-Peer Networking and Applications (PPNA), 2009 (To appear).
• A. Bennaceur, P. Singh, P.-G. Raverdy, and V. Issarny “The iBICOOP Middleware: Enablers and Services for Emerging Pervasive Computing Environments” Proceedings of the IEEE PerWare’09 Middleware Support for Pervasive Computing Workshop at IEEE PerCom, 2009.
• S.K. Madria and A. Mondal “Economic-based Incentive Schemes for Dynamic Data Management in Mobile P2P Computing” Proceedings of the International Conference on Mobile Data Management (MDM), 2008.
• P. Singh, H. N. Ha, P.L. Olivier, C. Kray, P. Blythe, and P. James “Rapid Prototyping and Evaluation of Intelligent Environments using Immersive Video” Proceedings of the MODIE: Modeling and Designing User Assistance in Intelligent Environment, in Conjunction with Mobile HCI, 2006.
Software Engineering
Faculty: Pankaj Jalote, Ashish Sureka
The group has traditionally worked on issues of software quality, software process, quantitative techniques like focused metrics, empirical and experimental methods, and statistical process control. More recently, the group is also working on a range of issues in the rapidly emerging areas of service oriented computing and web services. Some of the areas on which work is going on include performance prediction, negotiation, and issues related to composition. Software as a service is a related area in which the group is exploring configurability and multi-tenancy. Mining software process data for extracting knowledge for improving the software or software process, an area in which the group has past experience, is now being enhanced with extracting information from unstructured data using text mining so better knowledge can be obtained.
Selected Recent Publications:
• D Mukherjee, P Jalote, and M Gowri Nanda. Determining QoS of WS-BPEL Compositions, Proceedings of the Sixth International Conference on Service Oriented Computing (ICSOC), Sydney, Australia, 2008.
• P. Jalote, B. Murphy, V. S. Sharma, Post-Release Reliability Growth in Software Products, ACM Transactions on Software Engineering and Methodology, 2008.
• A. Gupta, P. Jalote, “An Approach for Experimentally Evaluating Effectiveness and Efficiency of Coverage Criteria for Software Testing”, Journal of Software Tools for Technology Transfer (STTT), 2008.
|
