Serban Carata

Resume

Serban Vasile CARATĂ has extensive knowledge and experience in Artificial Intelligence (over 7 years), some of his main education expertise until now being in the applied research field of Advanced Pattern Recognition techniques in Earth Observation Imaging and Advanced Neural Systems. He received his PhD degree in 2017 from the Faculty of Electronics, Telecommunications and Information Technology (ETTI), University Politehnica of Bucharest (UPB). Before that, he studied Intelligent Systems, Computer Vision and Data Mining, at ETTI, UPB, during the Master’s degree which was completed in 2013. 

Education

  • 2017: PhD. degree in Artificial Intelligence obtained at ETTI, UPB.  Principal subjects/skills covered are: advanced pattern recognition techniques in Earth observation imaging, advanced neural systems, data mining;
  • 2013: MSc. degree in Intelligent Systems and Artificial Vision, ETTI, UPB. Principal subjects/skills covered are: neural systems, data mining, artificial intelligence;
  • 2011: Physicist degree, Faculty of Physics, University of Bucharest. Principal subjects/skills covered are: electronics, optical engineering, mechanics;
  • 2010: Mathematician degree, Faculty of Mathematics and Computer Science, University of Bucharest. Principal subjects/skills covered are: differential equations, programming, microprocessor architectures.

Professional experience

  • 2018-present: Associate professor at the ETTI, UPB, Bucharest (Romania). 
  • 2017-2019: Associate researcher at the Institute of Space Science. 
  • 2012 – present: Video Analytics Researcher at UTI Grup (Softrust Vision Analytics). Responsibilities: 

Mr. Serban-Vasile Carata has a complex role, being responsible with planning the direction of research of a variety of initiatives in the fields of, and seeing the plan through. 

He has developed and implemented algorithms for face detection, face recognition, people detection, tracking multiple objects, automatic threat detection, automatic scene change detection algorithms,  LPR, among others, from designinging video analytics solutions, to complex solutions implementation. 

His technical skills are under continuous improvement, as he has extensive experience with: Deep neural network frameworks, OpenCv, Matlab, SQL, and also experience with tools like: Visual Studio, C, C++, SSE. He is also improving his knowledge with: MacOS X, MFC and  embedded Linux systems.

Being a team leader and project manager on most of the initiatives mentioned above, his soft skills have been in a continuous improvement, as he is coordinating different teams in various contexts – his managerial skills being mandatory for the projects’ success – as he is the main responsible in leading, guiding and advising the project teams he manages.

Publications

During his research, Mr. Serban Carata managed to contribute by publishing valuable articles  in established national and international Conferences and Journals, being the author and co-author of 15  research articles.

Some of the most representative papers on which Mr. Serban-Vasile Carata has contributed can be found below in detail. These have been presented at IGARSS, ROLCG and COMM conferences and included in each conferences publication.

  1. S. V. Carata, V. E. Neagoe, „An Innovative Pulse-Coupled Neural Network Approach to Image Segmentation”, 6th European Conference of Computer Science (ECCS ’15) Rome, Italy, pp. 137-141, November 2015, (BDI)
  2. S. V. Carata, V. E. Neagoe , „A Pulse-Coupled Neural Network Approach for Image Segmentation and Its Pattern Recognition Application”, 11th International Conference on Communications (COMM), pp. 61-61, June 2016, (IEEE Xplore, ISI Thomson)
  3. V.E. Neagoe, S.V. Carata, A. D. Ciotec, „Automatic Target Recognition in SAR Imagery Using Pulse-Coupled Neural Network Segmentation Cascaded with Virtual Training Data Generation CSOM-Based Classifier”, International Geoscience and Remote Sensing Conference (IGARSS 2015), Milano, Italy, pp. 3274-3277, July 2015, (IEEE Xplore, ISI Thomson)
  4. V.E. Neagoe, S.V. Carata, A. D. Ciotec, „An Advanced Neural Network-Based Approach for Military Ground Vehicle Recognition in SAR Aerial Imagery ”, Scientific Research and Education in the Air Force-Afases, Brasov, pp.41-47, May  2016, (BDI)
  5. V.E. Neagoe, S.V. Carata, „Subject Independent Drunkenness Detection Using Pulse-Coupled Neural Network Segmentation of Thermal Infrared Facial Imagery”, International Journal of Mathematical and Computational Methods, pp. 305-312, August 2016, (BDI)
  6. V.E. Neagoe, A. D. Ciotec, S.V. Carata, „A new multispectral pixel change detection approach using pulse-coupled neural networks for change vector analysis”, International Geoscience and Remote Sensing Conference (IGARSS 2016), Beijing, China, pp. 3386-3389, July 2016, (IEEE Xplore, ISI Thomson)
  7. C.A. Mitrea, S.V. Carata, M. Ghenescu, SCOUTER: Content-based multiple-instance object retrieval system, IEEE International Conference on Multimedia & Expo Workshops (ICMEW), INSPEC 15330364, DOI 10.1109/ICMEW.2015.7169821, TURIN, 2015.
  8. C.A. Mitrea, S.V. Carata, B. Ionescu, T. Piatrik and M. Ghenescu, Ensemble-based Learning Using Few Training Samples for Video Surveillance Scenarios, fifth International Conference on Image Processing Theory, Tools and Applications IPTA, Orleans, France, 2015
  9. R Mihaescu, SV Carata, V Ghenescu, M Chindea, C Mitrea, Noise Removal Based on Artificial Intelligence Synergy Methods for Surveillance Scenarios, International Conference on Communications (COMM), 501-504, 2018
  10. SV Carata, V Ghenescu, M Ghenescu, M Chindea, R Mihaescu, Salt and Pepper Noise Removal by Combining Genetic Algorithms-Neural Networks and Statistical Methods, International Conference on Communications (COMM), 147-150, 2018
  11. Veta Ghenescu, Roxana Elena Mihaescu, Serban-Vasile Carata, Marian Traian Ghenescu, Eduard Barnoviciu, Mihai Chindea, Face Detection and Recognition Based on General Purpose DNN Object Detector, International Symposium on Electronics and Telecommunications (ISETC), 1-4, 2018
  12. Bogdan Belean, Serban Vasile Carata, Marian Ghenescu, Efficient search algorithms implementation for SAR image analysis, Conference Grid, Cloud & High Performance Computing in Science (ROLCG), 1-4, 2018[13]       Veta Ghenescu, Eduard Barnoviciu, Serban-Vasile Carata, Marian Ghenescu, Roxana Mihaescu, Mihai Chindea, Object Recognition on Long Range Thermal Image Using State of the Art DNN, Conference Grid, Cloud & High Performance Computing in Science (ROLCG), 1-4, 2018
  13. Veta Ghenescu, Eduard Barnoviciu, Serban-Vasile Carata, Marian Ghenescu, Roxana Mihaescu, Mihai Chindea, Object Recognition on Long Range Thermal Image Using State of the Art DNN, Conference Grid, Cloud & High Performance Computing in Science (ROLCG), 1-4, 2018
  14. Serban-Vasile Carata, Mihai-Gabriel Constantin, Veta Ghenescu, Mihai Chindea, Marian Ghenescu:  Innovative Multi Pcnn Based Network for Green Area Monitoring – Identification and Description of Nearly Indistinguishable Areas – In Hyperspectral Satellite Images. IGARSS 2018: 2639-2642
  15. Eduard Barnoviciu, Veta Ghenescu, Serban-Vasile Carata, Marian Ghenescu, Roxana Mihaescu, Mihai Chindea: GDPR compliance in Video Surveillance and Video Processing Application; 2019 International Conference on Speech Technology and Human-Computer Dialogue (SpeD)
  16. Eduard Barnoviciu, Serban-Vasile Carata, Veta Ghenescu, Marian Ghenescu, Roxana Mihaescu, Mihai Chindea: Accurate data identification in low signal-to-noise ratio series, 2019 IEEE International Conference on Advanced Scientific Computing (ICASC)
  17. Serban Carata, Roxana Mihaescu, Eduard Barnoviciu, Mihai Chindea, Marian Ghenescu, Veta Ghenescu:Complete Visualisation, Network Modeling and Training, Web Based Tool, for the Yolo Deep Neural Network Model in the Darknet Framework; 2019 IEEE 15th International Conference on Intelligent Computer Communication and Processing (ICCP)

Other publications, some in the sub-fields of Neural Network Approach to Image Segmentation, Pulse-Coupled Neural Network Segmentation of Thermal Infrared Facial Imagery, Noise Removal by Combining Genetic Algorithms-Neural Networks and Statistical Methods,  have been included in conferences like ICMEW, IJMCM, ECCS.

R & D Projects

Mr. Serban-Vasile Carata has participated in a variety of EU & National Projects in the last 10 years:

National Projects

  •  2013-2016 SCOUTER –  Role in project: coordinate the development and implementation of the video analytics algorithms. (Call: RO- PNII; Competition: IN) ; During the SCOUTER project is designed and developed intelligent video surveillance system for automatically localize the important events. Main contributions are made at the algorithms design and development. 
  • 2014-2017 THzDetect – Role in project: coordinate the development and implementation of the data analytics algorithms. (Call:RO-PNII: , Competition: Parteneriate)
  • 2016-2018 FINSEC –  Role in project: coordinate the implementation of the video analytics algorithms and the data fusion module. (Call: H2020; Competition: BES-2015) 
  • 2016-2019 ACQUIRE –  Role in project: development and implementation of the video analytics algorithms. (Call: CDI; Competition: STAR)
  • 2017-2018 SPOOTER – Role in project: Project Manager from the Economic Partner and coordinate the development and implementation of the video analytics algorithms.(Call: RO-PNIII; competition: PED) 
  • 2017-2020 SPIA-VA –  Role in project: coordinate the development and implementation of the video analytics algorithms. (Call: RO-PNIII; Competition: Solutii) 
  • 2017-2020 UAVINO –  Role in project: coordinate the development and implementation of the video analytics algorithms. (Call: RO-PNIII; Competition: Solutii)
  • 2017-2020 SPERO –  Role in project: coordinate the development and implementation of the video analytics algorithms. (Call: RO-PNIII; Competition: Solutii)

EU Projects

  • 2016-2018 SAFESHORE –  Role in project: coordinate the implementation of the video analytics algorithms and the data fusion module. (Call: H2020; Competition: BES-2015)
  • 2017-2019 EWISA – Role in project: coordinate the development and implementation of the video analytics algorithms. (Call: Proiect FRONTEX);This is a multinational research project that strides to give new life to legacy systems by empowering them with the latest artificial intelligence algorithms.  Main contributions are made at the video analytics algorithms architecture and development.