Time filter

Source Type

Chalkída, Greece

Nikitakis A.,Technical University of Crete | Papaefstathiou I.,Synelixis Solutions Ltd | Makantasis K.,Institute of Communication and Computer Systems | Doulamis A.,National Technical University of Athens
Proceedings of the 2016 Design, Automation and Test in Europe Conference and Exhibition, DATE 2016 | Year: 2016

Real-time segmentation of moving regions in image sequences is a very important task in numerous surveillance and monitoring applications. A common approach for such tasks is the background subtraction which tries to extract regions of interest from the image background for further processing or action; as a result its accuracy as well as its real-time performance is of great significance. In this work we utilize a novel scheme, designed and optimized for FPGA-based implementations, which models the intensities of each pixel as a mixture of Gaussian components; following a Bayesian approach, our method automatically estimates the number of Gaussian components as well as their parameters. Our novel system is based on an efficient and highly accurate on-line updating mechanism, which permits our system to be automatically adapted to dynamically changing operation conditions, while it avoids over/under fitting. We also present two reference implementations of our Background Subtraction Parallel System (BSPS) in Reconfigurable Hardware achieving both high performance as well as low power consumption; the presented FPGA-based systems significantly outperform a multi-core ARM and two multi-core low power Intel CPUs in terms of energy consumed per processed pixel as well as frames per second. Moreover, our low-cost, low-power devices allow for the implementation, for the first time, of a highly distributed surveillance system which will alleviate the main problems of the existing centralized approaches. © 2016 EDAA.

Voulkidis A.C.,Synelixis Solutions Ltd | Cottis P.G.,National Technical University of Athens
IEEE Transactions on Wireless Communications | Year: 2016

A distributed game theoretic framework based on correlated strategies is proposed to maximize the lifetime of dense homogeneous multiservice wireless sensor networks (MS-WSNs), that support multiple services continually and ubiquitously over the WSN deployment. The MS-WSN operation is dealt with as a game played by the multimode nodes. A correlated strategies approach is proposed to lead the MS-WSN close to its theoretical optimal state with respect to the network lifetime. The computationally efficient correlated strategy proposed to implement service selection by the multimode WSN nodes is distributed, based solely on local information exchange. Indicative simulation results concerning the application of the proposed scheme on top of k-hop clustering reveal that the proposed correlated strategies based framework leads the MS-WSN operation close to its theoretical optimal at no significant exchange of overhead messages. © 2016 IEEE.

Nikitakis A.,Technical University of Crete | Paganos T.,Technical University of Crete | Papaefstathiou I.,Synelixis Solutions Ltd
Proceedings -Design, Automation and Test in Europe, DATE | Year: 2014

One of the most important challenges in the field of Computer Vision is the implementation of low-power embedded systems that will execute very accurate, yet real-time, algorithms. In the visual tracking sector one of the most promising approaches is the recently introduced OpenTLD algorithm which uses a random forest classification method. While it is very robust, it cannot be efficiently parallelized in its native form as its memory access pattern has certain characteristics that make it hard to take advantage of the conventional memory hierarchies. In this paper, we present a novel embedded system implementing this algorithm. We accelerate the bottleneck of the algorithm by designing and implementing a high bandwidth distributed memory sub-system which is independent of the various software parameters. We demonstrate the applicability and efficiency of this novel approach by implementing our scheme in a modern FPGA. © 2014 EDAA.

Brokalakis A.,Synelixis Solutions Ltd | Papaefstathiou I.,Technical University of Crete
IEEE Wireless Communications and Networking Conference, WCNC | Year: 2012

Providing an energy-efficient communication scheme is highly desirable in Wireless Sensor Networks (WSNs), however this is often constrained by the processing and energy limitations of the wireless nodes. In this paper, we propose the use of a Turbo Code scheme to increase the robustness and energy efficiency of the communication between end nodes and base stations in single-hop topologies. Using several real-world energy consumption measurements from a widely used WSN platform, we demonstrate the operational environment in which the end-user can take full advantage of the proposed scheme. We propose, for the first time, the use of a reconfigurable hardware device that executes the encoding scheme; this approach can reduce the overall energy consumption of a node by more than 40%, when compared with a Turbo Code scheme implemented in software, as well as by more than 70% when compared with the traditional WSN transmission schemes that do not support any kind of Forward Error Correction. © 2012 IEEE.

Papaefstathiou I.,Synelixis Solutions Ltd | Chrysos G.,Technical University of Crete | Sarakis L.,TEI of Sterea Ellada
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015

This paper present a recently started multi-national project which will provide an innovative comprehensible, ultra-Fast, security-Aware CPS Simulator (called COSSIM) that will seamlessly simulate both the networking and the processing parts of the Cyber Physical Systems (CPS) significantly faster and more accurate than any existing solution. In order to achieve that COSSIM will develop a novel simulator framework based on a processing simulation subsystem (i.e. a “full-system simulator”) which will be integrated with a novel network simulator. Furthermore, innovative power consumption and security measurement models will be developed and incorporated to the end framework. On top of that, COSSIM will also address another critical aspect of an accurate CPS simulation environment: the performance as measured in required simulation time. COSSIM will create a framework that is orders of magnitude faster, while also being more accurate and reporting more CPS aspects, than existing solutions, by applying hardware acceleration through the use of field programmable gate arrays (FPGAs), which have been proven extremely efficient in relevant tasks. This paper presents the high-level architecture of the end-system as well the most innovative aspects of it that will allow it to be the fastest and most accurate such simulator, while it does not cover the security aspects of this novel system. © Springer International Publishing Switzerland 2015.

Discover hidden collaborations