Databases and Artificial Intelligence Laboratory

The Database and Artificial Intelligence (DBIA) laboratory is a center of excellence for Artificial Intelligence and Databases research since its founding in 1997. Our current research interests reside in machine learning, neural networks, computer vision, data mining and databases. 

The obtained results of the member of the laboratory were published in prestigious international journals and conferences. We believe that the results of our research will influence the way artificial intelligence and databases will be used to build future intelligent products inside modern enterprise.

Structure

Team

Research  directions:

  • Artificial Intelligence:
    • Neural Networks
    • Machine Learning
    • Computer Vision
    • Fuzzy Systems
    • Data Mining
  • Databases
    • Databases
    • Sensor Networks
    • Massive Distributed Data Processing


Research center vision

  • Human resources
  • Industry-Academia partnership
  • Research projects

Research projects

  • Structural funds
  • Bilateral projects
  • International research projects (FP7, CHIST-ERA)
  • National research projects

Industry partners

  • Everseen
  • Continental Automotive
  • Visma
  • Piconet
  • Krohne

Academic partners

  • University of Manchester (Marea Britanie)
  • University of Innsbruck (Austria)
  • University of New York at Stony Brook (USA)
  • Université Nice Sophia Antipolis (France)
  • Dublin City University (Ireland)

Papers

  • Petre Birtea, Dan Comănescu, Călin-Adrian Popa, Averaging on Manifolds by Embedding Algorithm. Journal of Mathematical Imaging and Vision, Vol. 49, Nr. 2, pag. 454-466, 2014. (ISI Impact Factor 1.603, Q1)
  • Călin-Adrian Popa, Eva Kaslik, Multistability and multiperiodicity in impulsive hybrid quaternion-valued neural networks with mixed delays. Neural Networks, Vol. 99, pag. 1-18, 2018. (ISI Impact Factor 5.785, Q1)
  • Călin-Adrian Popa, Global exponential stability of octonion-valued neural networks with leakage delay and mixed delays. Neural Networks, Vol. 105, pag. 277-293, 2018. (ISI Impact Factor 5.785,Q1)
  • Călin-Adrian Popa, Global exponential stability of neutral-type octonion-valued neural networks with time-varying delays. Neurocomputing, Vol. 309, pag. 117-133, 2018. (ISI Impact Factor 4.072,Q1)
  • Călin-Adrian Popa, Global μ-stability of neutral-type impulsive complex-valued BAM neural networks with leakage delay and unbounded time-varying delays. Neurocomputing, 2019. (ISI Impact Factor 4.072, Q1)
  • M. Baba, V. Gui, C. Cernazanu, D. Pescaru, "A Sensor Network Approach for Violence Detection in Smart Cities Using Deep Learning", Sensors, Vol. 19(7), MDPI Basel, April 2019, pp. 1-17. (ISI Q1)
  • D. Pescaru, D.I. Curiac, "Ensemble based traffic light control for city zones using a reduced number of sensors", Transportation Research Part C: Emerging Technologies, Elsevier, Vol. 46, September 2014, Pages 261-273 (ISI Q1)
  • D. Pescaru, D.I. Curiac, "Anchor Node Localization for Wireless Sensor Networks Using Video and Compass Information Fusion", Sensors, Vol. 14(3), MDPI Basel, March 2014, pp. 4211-4224. (ISI Q1)
  • V. Subramanian, M. Gilberti, A. Doboli, D. Pescaru, "A Goal-Oriented Programming Framework for Grid Sensor Networks with Reconfigurable Embedded Nodes", ACM Transactions on Embedded Computing Systems (TECS), Vol. 11 Iss. 4, ACM New York, December 2012, pp. 79:1-79:30. (ISI-IF = 1.368)
  • C. Istin, D. Pescaru, A. Doboli, "Stochastic Model-based Heuristics for Fast Field of View Loss Recovery in Urban Traffic Management Through Networks of Video Cameras", IEEE Transactions on Intelligent Transportation Systems, Vol. 12 Iss. 3, IEEE Press, September 2011, pp. 895-907. (ISI Q1)

Projects

  • Project PCD-TC-2017-41, Director Călin-Adrian Popa, Universitatea Politehnica Timişoara, Buget 46500 RON (10000 EUR).
  • UEFISCDI grant PN-III-P2-2.1-CI-2017-0427, contr. 51CI/25.07.2017 - "Parking assistance system for TPark", (D. Pescaru, C. Istin, M. Baba) 50.000 lei
  • RC UPT-Everseen Ltd. Ireland, Contract nr. 1/2012 - "Algorithms on Video Analytics for Incident Detection and Classification for Retailer Industry", (D. Pescaru, O. Parvu, D. Seres) 16.000 EUR
  • CNCSIS MEdC nr. 2739 din 19.05.2006  + GR76/377 tema 25 din 23.05.2007 - "Video Surveillance Applications Platform Based on Wireless Sensor Networks", (D. Pescaru, S. Serau, D. Cosma, B. Ciubotariu, P. Mierlutiu) 400.560 lei
  • Mobility Project Socrate-Erasmus for master degree students 2006-2007. Partners: Politehnica University of Timisoara, Romania and Dublin City University, Ireland (coordinators: D. Pescaru (UPT) and G. Muntean(DCU))

Research areas: