BrainMap

Mr. Marian Traian Ghenescu

Ph.D

Researcher - INSTITUTUL DE STIINTE SPATIALE-FILIALA INFLPR

Researcher

Expertise & keywords

Identifying the unusual behavior of people in video streams Call name: P 2 - SP 2.1 - Soluţii - 2021
PN-III-P2-2.1-SOL-2021-0024 2021 - 2023

Role in this project:

Coordinating institution: UNIVERSITATEA OVIDIUS

Project partners: UNIVERSITATEA OVIDIUS (RO); SOFTRUST VISION ANALYTICS S.A. (RO); UNIVERSITATEA POLITEHNICA DIN BUCURESTI (RO); Academia Tehnică Militară „FERDINAND I” (RO)

Affiliation:

Project website: https://ideneo.univ-ovidius.ro/

Abstract:

The IDENEO project "Identifying the unusual behavior of people in video streams" aims to raise the current level of knowledge and methods in the field of detecting suspicious behavior of people and crowds and to combat violent or terrorist actions, in certain circumstances. This will allow law enforcement to act preventively and not reactively.

To achieve this goal, the focus will be on three main aspects: (i) the development of state-of-the-art algorithms, such as AI (Artificial Intelligence), which will be able to detect humans positions in the environment, actions and behavior of people, and crowd behavior; (ii) adapting the algorithms to run on both airborne and classic cameras is yet another very important aspect of the project which will allow major flexibility by allowing a processed video stream to be obtained from any location; (iii) the integration of all developed modules in a VMS capable of rendering all relevant information to the operator, including GIS (Geographic Information System) information, in an easy and intuitive way. All the processed information (saved and displayed to the operator) will be in full compliance with GDPR laws; this will be achieved by a dedicated module that will hide sensitive data to the operators without access.

Finally, the proposed solution will allow law enforcement to detect individuals or groups and crowds of suspects, in varied and dynamic scenarios, such as public protests or within monitored areas of interest. This will allow the user to understand the entire operational scenario, by combining information from airborne cameras, classic cameras, image analysis and displaying them on the GIS map.

6. Identify people in video streams, using silhouette biometrics Call name: P 2 - SP 2.1 - Soluţii - 2021
PN-III-P2-2.1-SOL-2021-0114 2021 - 2023

Role in this project:

Coordinating institution: UNIVERSITATEA POLITEHNICA DIN BUCURESTI

Project partners: UNIVERSITATEA POLITEHNICA DIN BUCURESTI (RO); SOFTRUST VISION ANALYTICS S.A. (RO); Ministerul Apararii Nationale prin Agentia de Cercetare pentru Tehnica si Tehnologii Militare (ACTTM) (RO)

Affiliation:

Project website: https://deepvision-romania.aimultimedialab.ro/

Abstract:

The DeepVisionRomania project comes to respond to the call-for-solutions launched for the development, testing and operationalization of software services for automatic identification and re-identification of people in video sequences with GDPR regulations in force, in order to increase operational competitiveness, economy and staff skills via “made-in-Romania” security software solutions. The developed technologies are able, by their degree of novelty, to ensure a strategic security advantage for Romania by supporting operations to prevent and combat terrorism or other categories of threats to the public order and safety, or emergency situations. The project proposes: (i) the development of new state-of-the-art AI algorithms based on deep neural networks capable of ensuring the identification (search in a database of identities) and the re-identification of persons (search for the appearance of a person in multiple sources ) in different video streams using both static and dynamic information; (ii) development of dedicated analysis software services but with a secure architecture and developed in accordance with operational standards / requirements; modular so that any new functionality does not require reprogramming of the system; scalable to multiple existing and new video sources, optimized to ensure all real-time processing; (iii) development of an integrated software and hardware platform that brings together all solutions; (iv) ensuring GDPR compliance through an advanced user management system with access security. The market for the applicability of such systems is particularly wide, from the operational institutions that process this information, critical infrastructures (protection and security, prisons, border, airports, hospitals, banks), to the civil area (transport, shopping centers, surveillance). urban, residential complexes, company headquarters, supermarket).

Enhancing and Improving Customer Experience and Services in Supermarkets via SMART Artificial Intelligence Powered Systems Call name: P 2 - SP 2.1 - Proiect de transfer la operatorul economic
PN-III-P2-2.1-PTE-2019-0055 2020 - 2022

Role in this project:

Coordinating institution: SOFTRUST VISION ANALYTICS S.A.

Project partners: SOFTRUST VISION ANALYTICS S.A. (RO); UNIVERSITATEA POLITEHNICA DIN BUCURESTI (RO)

Affiliation: SOFTRUST VISION ANALYTICS S.A. (RO)

Project website: http://smartretail.aimultimedialab.ro

Abstract:

With the rapidly growing number of supermarkets and the diversification of the available products, their management became more and more complex due to the growing number of tasks that require human expertise (e.g., product monitoring, stock, and events analysis). Therefore, automatization solutions are vital for improving the quality of services and the consumer experience. The current existing national and international systems, require the usage of CCTV technology, whose progress (e.g., unlimited number of video cameras, 4k resolution) led, not to a simplification, but to new technological challenges, i.e., the vast information that requires manual analysis, the impossibility of real-time analysis on large time intervals, the difficulty of retrieving the relevant information from the immense volume of data. The SMARTRetail project responds to the aforementioned challenges through an integrated solution, that can be attached to existing CCTV systems, without the need to modify it, allowing the automatization via Artificial Intelligence techniques for these retail-specific problems. The proposed solutions are based on advanced deep neural network techniques that provided state-of-the-art results in current machine learning problems. SMARTRetail allows the automatic detection of degraded products on shelves that need to be replaced, automatic detection of empty shelves, and real-time monitorization of fraudulent behavior to counteract the possibility of theft and damage to the products. Innovation, integration, and automation is required to reinvent the way retailers manage their resources. Recent surveys show that in 2017, only 3% of retailers implemented AI solutions, while 45% had in their strategy the deployment of such solutions during the 2018-2021 period, being a constantly expanding market. The outcome of the project will be the integration of the solution in the KVision VMS system commercialized by the coordinator and its validation in the supermarket.

Virtual Guardian: Artificial Intelligence Powered Multi-Sensor System for Automatic Securing of Areas of Interest Call name: P 2 - SP 2.1 - Proiect de transfer la operatorul economic
PN-III-P2-2.1-PTE-2019-0570 2020 - 2022

Role in this project:

Coordinating institution: SOFTRUST VISION ANALYTICS S.A.

Project partners: SOFTRUST VISION ANALYTICS S.A. (RO); UNIVERSITATEA POLITEHNICA DIN BUCURESTI (RO)

Affiliation: SOFTRUST VISION ANALYTICS S.A. (RO)

Project website: http://gravi.aimultimedialab.ro

Abstract:

GRAVI (Virtual Guardian: Multisensory Artificial Intelligence System for Automatic Securing of Areas of Interest) proposes an innovative, multi-sensor solution that will ensure the protection of high dispersed geographical areas (surveillance perimetral) with minimum operating cost and installation. GRAVI has the following main objectives: (O1) Development and implementation of an innovative communication protocol between any type of sensor and modern surveillance equipment; (O2) Developing an innovative system of automatic detection and intervention in the field for the analysis and confirmation of the event, using a PanTiltZoom platform; (O3) Development, implementation and transfer to the industrial partner of the state-of-the-art AI algorithms based on advanced technologies such as deep neural networks; (O4) Development and implementation of an advanced Data Fusion system that combines information from different sensors connected to the system; (O5) Development and implementation of a software system for event analysis and management; (O6) Integration of a functional system validated in the field.

The project starts from a technological level TRL 4-5, materialized by software models (deep neural network algorithms) and in order to reach the targeted TRL level 6, the GRAVI system will be integrated into the KVision platform (product registered at ORDA), a highly scalable video management system. This project perfectly merges the experience in the security market as well as in the development of automatic learning algorithms of the two partners, hence guaranteeing real success, both in the finalization of the project and profit in the industrial market after product deployment.

In Romania, there is not yet a similar GRAVI system for detecting, intervention and validating the events of interest in the perimeter areas, so there is an unsatisfied demand, and the resulting technology transferred through this project addresses precisely these needs.

Technologies and innovative video systems for person re-identification and analysis of dissimulated behavior Call name: P 2 - SP 2.1 - SOLUTII - 2 - Tehnologii şi sisteme video/audio inovative
PN-III-P2-2.1-SOL-2016-02-0002 2017 - 2020

Role in this project:

Coordinating institution: UNIVERSITATEA POLITEHNICA DIN BUCURESTI

Project partners: UNIVERSITATEA POLITEHNICA DIN BUCURESTI (RO); UTI GRUP S.A. (RO); Ministerul Apararii Nationale prin Agentia de Cercetare pentru Tehnica si Tehnologii Militare (ACTTM) (RO)

Affiliation: UTI GRUP S.A. (RO)

Project website: http://campus.pub.ro/lab7/spiava/

Abstract:

Within the actual context of terrorist threats, SPIA-VA aims to develop intelligent systems for automatic person re-identification, detection of dissimulated behavior and speech analysis. The proposed system is inherently multi-modal via processing of multiple and variable quality sources, e.g., video, audio, text, infrared, depth, thermal, as well as the fusion of these. The proposed solution pushes forward the state of the art in deep learning and is designed to respond to a large variety of functional and operational scenarios. It is to be validated in real world scenarios, developed in cooperation with the beneficiary military public institutions.

Millimeter Wave (THz) Time Domain Spectroscopy Instrument for Dangerous Substances Identification Call name: Joint Applied Research Projects - PCCA 2013 - call
PN-II-PT-PCCA-2013-4-0967 2014 - 2017

Role in this project:

Coordinating institution: INSTITUTUL DE STIINTE SPATIALE-FILIALA INFLPR

Project partners: INSTITUTUL DE STIINTE SPATIALE-FILIALA INFLPR (RO); INSTITUTUL NATIONAL DE CERCETARE DEZVOLTARE PENTRU FIZICA LASERILOR, PLASMEI SI RADIATIEI - INFLPR RA (RO); UNIVERSITATEA POLITEHNICA DIN BUCURESTI (RO); UTI GRUP S.A. (RO)

Affiliation: INSTITUTUL DE STIINTE SPATIALE-FILIALA INFLPR (RO)

Project website: http://www2.spacescience.ro/projects/project-290/

Abstract:

The security problem of a country it was and still remains an important domain. In this way, the development of devices and systems, in inner production which ensure and improve the security level must be high priority. The research made in this way ensures not just an economical development, but also an access to the top of the technology.

In the last years, the development of the THz technology tends to become gauge devices in complex substances identification. From this, become almost clear the way to use such systems in the security domain, for identification of the dangerous substances, fake drugs, dangerous objects imagistic etc..

THz radiation covers a slightly investigated area until now from the electromagnetic spectrum, between microwaves and infrared. Because of those properties, the molecular structure of a substance could be determinate from the absorption or emission of THz radiation. Having such spectrum (absorption or emission) we can identify a complex substance. With a large data base of THz domain spectrums we can identify more substances.

In the same time, the THz radiation photons have a weak energy, so the radiation is not dangerous for human being (is not an ionizing radiation). THz radiation being a broadband, it is obey to the dispersion, diffraction and divergence phenomena, doing quite difficult the handling of such beam. However, it is possible the quasi optical coupling using the lenses, because some of the dielectrics are transparent at THz radiations. On the other hand, from the diffraction point of view, THz radiation has a behavior specific to microwaves, so we need some bigger optical systems to control the spatial propagation.

The consortium consisting of: Institute of Space Science, National Institute for Laser Plasma and Radiation Physics, Research Center of Optoelectronics, from University Politehnica of Bucharest and an industrial partner UTI Group – have propose to realize an active, portable device to identify dangerous substances in the THz domain for security. The product will be a novelty on the national and international plan, because of it compact structure, response and low cost.

In this project we take account of: design and accomplishment of a molecular spectroscopy setup in the THz domain and the validity of the spectroscopic setup in the national security domain.

SMART surveillance system with automatic identification of the interest events Call name: Dezvoltare Produse - Sisteme - Tehnologii - 2012
PN-II-IN-DPST-2012-1-0034 2013 - 2015

Role in this project:

Coordinating institution: UTI Grup

Project partners: UTI Grup (RO); Universitatea Politehnica din Bucuresti (RO)

Affiliation: UTI Grup (RO)

Project website: http://uti.eu.com/pncd-scouter/index.html

Abstract:

The proposed project represents UTI group’s need to expand the existing range of intelligent video systems with an automatic localization of events of interest (Scout), a new and innovative system with potential for commercial exploitation, in order to obtain profitable results for the company. In this sense, the project aims to improve business capabilities in the development of new solutions to enhance the quality, product diversity and reduced response time to market demands. The project also aims to increase the quality and diversification of innovative products and services of UTI Group, which will lead directly to increasing the group’s strength in the business market.

The SCOUTER project implementation is divided in 5 stages of research and development that will produce: analysis and technical studies on video processing methods, technical feasibility studies, technical specifications, experimental designs, test plans and reports, prototypes, technical and economic documentation, installation manuals, commissioning and use of the system component products, technology transfer in the production department of the UTI group, copyrighting algorithms resulted in the stages of industrial research and patent on technologies results in the experimental stages of development, participation in technical and scientific conferences, article publishing and organizing workshops presenting the results of the project.

The need for implementation of the project is determined by the need to increase the level of innovation in order to increase market competitiveness of products by UTI Group, together with the existence of a real market demand for intelligent video surveillance systems which automatically locate events. The main identified potential customers are the areas of critical infrastructure, local administration (traffic management) of public transport, retail (supermarkets, malls) and banking system.

The project contributes to fulfilling the objective of achieving the National Development Plan 2007 - 2013 (NDP) to "increase economic competitiveness and development of the knowledge economy", as well as the priorities of the National Strategic Reference Framework (NSRF) aimed at "increasing long-term economic competitiveness".

FILE DESCRIPTION	DOCUMENT
List of research grants as project coordinator or partner team leader
Significant R&D projects for enterprises, as project manager
R&D activities in enterprises
Peer-review activity for international programs/projects