BrainMap

Camelia Florea

Assistant Professor

Assistant Professor - UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA

Researcher | Teaching staff | Scientific reviewer | Consultant

Web of Science ResearcherID: http://www.researcherid.com/rid/E-4122-2012

Expertise & keywords

Intelligent system for dams surveilance by information fusion Call name: P 2 - SP 2.1 - Proiect de transfer la operatorul economic
PN-III-P2-2.1-PTE-2016-0134 2016 - 2018

Role in this project: Key expert

Coordinating institution: IPA SA

Project partners: IPA SA (RO); UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA (RO)

Affiliation: UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA (RO)

Project website: http://www.automation.ro/damfu/

Abstract:

The project aims to develop a system - complex, modular product destined for dams supervision and other hydrotechnical critical infrastructure, IHC, (dikes, weirs, adductions channel, surge towers), in order to ensure their safety on natural disasters, on technological accidents, and on natural aging processes, for protecting population and environment.

To achieve the proposed goal, it aims to develop and validate under operating conditions similar to real ones TRL6 of the prototype for modular, configurable product-system, destined for IHC supervision, by acquisition techniques, processing and fusion of multisensory information with image information processed by "computer vision" techniques. Will be capitalized obtained results, at the level of laboratory model, and experimental model: on FUZIBAR project (ctr. 705/2006, CEEX), by partners of the project proposal IPA SA (CO coordinator) and Technical University of Cluj-Napoca (P1 partner)

The system will be composed of Information processing and fusion station – STAFI, and the following subsystems specific to IHC supervision: i)Data acquisition and processing station for sensors and transducers incorporated in IHC - SASI; ii)Data acquisition and processing station for optical sensors designed for IHC tracking – SASO; iii)Video Inspection System - IHC- SIVI.

The system will be achieved and validate both on individual components, as well as an ensemble.

Interactive Real-Time Video Surveillance Services for Improved Public Safety in Crowded Urban Areas Call name: Joint Applied Research Projects - PCCA 2013 - call
PN-II-PT-PCCA-2013-4-1791 2014 - 2017

Role in this project: Key expert

Coordinating institution: UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA

Project partners: UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA (RO); IPA SA (RO); ACADEMIA DE POLITIE "ALEXANDRU IOAN CUZA" (RO); AXTEN SRL (RO); E G H IMPORT EXPORT S.R.L. (RO)

Affiliation: UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA (RO)

Project website: http://46.243.115.186/sites/intereviss/

Abstract:

Intelligent distributed video surveillance plays a central role in public safety and security, as they can provide not only the monitoring and detection of anomalous events across wide area public scenes, but also the understanding and semantic representation of these events in real time, to effectively support the authorized personnel decisions. With the large number of existing surveillance cameras, it is feasible to insure the coverage of public areas, but information extraction from a huge amount of video data in real time, at reasonably low false accept rates, is a challenging task. The definition of abnormal events in video requires advanced data mining solutions, but these are often computationally complex, therefore more difficult to implement in a low cost and time effective fashion, to be really helpful to the surveillance personnel. The problem becomes even more challenging in crowded urban scenes, under the constraints of fixed surveillance cameras and subject to the natural occlusions in a typical urban environment. Analysing dense crowd behaviour, tracking objects, crowds and individuals across multiple camera views and identifying abnormal events related to them requires special computer vision algorithms and collaborative video analytics approaches implemented in a computationally effective manner. Another important aspect is to provide an interactive human-machine visual surveillance solution that allows effective false alarm filtering by the operator and user-accepted decision support customizable according to the task and to the trained personnel needs. Considering these challenges, the project Interactive Real-Time Video Surveillance Services for Improved Public Safety in Crowded Urban Areas – INTEREVISS aims to develop a suite of customizable components and various surveillance scenarios for crowded urban environments dedicated to intelligent video feeds analysis in multi-camera networks in real-time for abnormal events detection. Through this, the project INTEREVISS meet the research thematic 8.5.3 of the research direction 8.5 Security systems and infrastructure. The developed software components and services will have a flexible structure that allows their integration in the existing surveillance networks as well on intelligent cameras, by efficient digital algorithms of compressed domain video streams processing, for different standards (MPEG-2,-4,-7, H.26x). The benefits of components and the resulting services will be proven by their integration into an interactive visual surveillance and analytics system, with facilities of communication between the video cameras, a command server and mobile devices on the field, able to provide the operator effective mechanisms for enhanced situational awareness and decision support. The off-line analysis of video sequences for post-events forensics will allow the updating and enrichment of the knowledge base of the decision support system. Geographical information about the events, their impact, operational routes, personnel and resources locations will be integrated in the system. The main goal is to offer reliable real-time solutions capable to improve the personal security and safety in public crowded places, potentially under terrorist attacks, hooliganism actions or accidents which may cause panic in large gatherings. The theoretical and practical solutions developed and implemented in the project will be applicable to different security cases. The experience and interdisciplinarity of the consortium (which includes a technical university, an R&D institute, the Police Academy and two companies with expertise in video surveillance and GIS) guarantees the achievement of the project goals.

Mobile System for Rehabilitative Vocal Assistance of Surgical Aphonia. Call name: Joint Applied Research Projects - PCCA 2013 - call
PN-II-PT-PCCA-2013-4-1660 2014 - 2017

Role in this project:

Coordinating institution: UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA

Project partners: UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA (RO); FORTECH SRL (RO); UNIVERSITATEA DE MEDICINA SI FARMACIE (U.M.F) Cluj-Napoca (RO); UNIVERSITATEA BABES BOLYAI (RO)

Affiliation: UNIVERSITATEA TEHNICA DIN CLUJ - NAPOCA (RO)

Project website: http://speech.utcluj.ro/swara

Abstract:

SWARA (Mobile System for Rehabilitative Vocal Assistance of Surgical Aphonia) aims at helping the people who have unfortunately lost their voice to be able to take part in the daily social life with limited restrictions regarding their communication. Medical studies show that in most cases, people who underwent a total laryngectomy for a larynx or hypo pharynx cancer will either refuse, or be unable to use the existing prosthesis available to them on the specialised market. Furthermore, the patient’s family and close friends have trouble accepting their new voice quality or assistive devices. Speech synthesis systems have long been used as an assistive technology for those who have, by various means, lost their ability to speak. But these systems so far, have a number of drawbacks, out of which the most important, is the fact that the number of available voices for the synthesis system are limited, and that they are hard to personalise for each user. Another disadvantage refers to the fact that the speech synthesis devices are actually expensive mini-computers which cannot be carried around all day long. The third disadvantage, and the one which probably is the most important, is that typing the text to be synthesized causes an unpleasant pause in the conversation, and this results in an uncomfortable interaction between the user and an arbitrary interlocutor. Therefore, the main goal of the project is to develop a speech synthesis technology that will allow the people who have surgically lost their ability to speak to communicate with any other person in a more natural manner, without fear of being rejected or misunderstood. This will be achieved through the following primary objectives:

- a personalised speech synthesis system which will use the patient’s voice or a similar one created from a speech database developed within the project;

- a smart text input method which will reduce the unnecessary delays in the dialogue;

- the access of the system via a Cloud-based service from a mobile device, which corresponds to the functional and portability requirements of such a technology.

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