Welcome to Dyvine As recent crises, such as the tsunami in south-east Asia and Hurricane Katarina in Louisiana have shown, the key element in Environmental Risks Management is to correctly assess the situation, its possible evolutions and implications. Getting the best "visualization" of the situation is therefore essential at all stages of the Crisis. Dynamic Visual Networks can provide such a real-time "visualization" of the situation, helping decision-makers to "feel" the situation as if they were experiencing it. We are in a society that is full of images and videos. If the famous adage, "Images are worth a thousand words" is true then perhaps, videos are worth much more. Advances in telecommunication and communication industry and a rise in demand, have pushed camera costs to a dramatic low. Broadband communication networks will soon be available as well as low- cost high computing power and storage capabilities. There are important advances in the Image and Video Processing. All these components can be leveraged with advances in calibration and compression technologies to build effective Dynamic Visual Networks as proposed in this document.
About DYVINE The DYVINE project focuses on both : * The integration of thousands of video sensors, fixed or mobile, in situ or airborne, and the development of exploitation applications for area monitoring to support the Risk Management cycle and * The associated resilient communication solutions to support the system. Testing will be performed in urban environment with the integration of the city camera networks through Rescue scenarios. DYVINE outputs will include laws/regulations study and standards analyses to propose future orientations for this type of systems. DYVINE objective is to design develop and test a representative version of a surveillance network based on visual sensors (images and video, in situ or airborne) which can be configured as a function of the requirements and events. This network can be used to monitor any kind of area or infrastructure which can be threatened by natural of industrial disasters. The scientific and technological objectives of DYVINE are to: 1. Design a generic architecture encompassing a vast array of various types of visual sensors, fixed and mobile, ground-based and airborne. This architecture will propose system standards and sensors interfaces standards. This architecture will be validated during the trials by demonstrating that it is possible to integrate 1000 cameras, 2. Propose advanced solutions for the (re)configuration of the surveillance network as a function of the events. These solutions will be demonstrated during the trials with the progressive integration of additional fixed and mobile cameras, 3. Study the communication means as well as develop new processing capabilities and networks necessary to integrate the largest possible forest of sensors. The result of the project will be robust wireless communications solutions enabling the dynamic integration of the sensors and compression/optimisation solutions to make the best of the available bandwidths, 4. Study the exploitation applications and fusion/correlation algorithms to provide the operators with the most comprehensive synthetic situation picture. The result of this objective will be advanced software module enabling the fusion of (overlapping) video data, the correlation of heterogeneous information and the tracking of persons or objects in a large area. The resulting test-bed will be demonstrated in the frame of a surveillance/disaster mitigation scenario in an urban environment. It will demonstrate real advances in Surveillance capabilities illustrating how the end-users (Civil Protection, cities, police, etc.) can have a global situation awareness with a large coverage and still detailed view.
Description: DYVINE objective is to design develop and test a representative version of a surveillance network based on visual sensors (images and video, in situ or airborne) which can be configured as a function of the requirements and events. This network can be used to monitor any kind of area or infrastructure which can be threatened by natural of industrial disasters. The scientific and technological objectives of DYVINE are to: 1. Design a generic architecture encompassing a vast array of various types of visual sensors, fixed and mobile, ground-based and airborne. This architecture will propose system standards and sensors interfaces standards. This architecture will be validated during the trials by demonstrating that it is possible to integrate 1000 cameras, 2. Propose advanced solutions for the (re)configuration of the surveillance network as a function of the events. These solutions will be demonstrated during the trials with the progressive integration of additional fixed and mobile cameras, 3. Study the communication means as well as develop new processing capabilities and networks necessary to integrate the largest possible forest of sensors. The result of the project will be robust wireless communications solutions enabling the dynamic integration of the sensors and compression/optimisation solutions to make the best of the available bandwidths, 4. Study the exploitation applications and fusion/correlation algorithms to provide the operators with the most comprehensive synthetic situation picture. The result of this objective will be advanced software module enabling the fusion of (overlapping) video data, the correlation of heterogeneous information and the tracking of persons or objects in a large area. The resulting test-bed will be demonstrated in the frame of a surveillance/disaster mitigation scenario in an urban environment. It will demonstrate real advances in Surveillance capabilities illustrating how the end-users (Civil Protection, cities, police, etc.) can have a global situation awareness with a large coverage and still detailed view.
Technical Approach The DYVINE project focuses on both: * The integration of thousands of video sensors, fixed or mobile, in situ or airborne, and the development of exploitation applications for area monitoring to support the Risk Management cycle and * The associated resilient communication solutions to support the system. Testing will be performed in urban environment with the integration of the city camera networks through Rescue scenarios. DYVINE outputs will include laws/regulations study and standards analyses to propose future orientations for this type of systems.
Project Workpackages The project is split into 6 main work packages: * WP1000 contains all the tasks related to the management of the project and to all the project-level issues; * WP2000 performs all the tasks of the system engineering; * WP3000 is dedicated to the integration of all the types of visual sensors through the communication network(s); * WP4000 deals with different types of required data processing throughout the system; * WP5000 covers all the legal and standards issues; * WP6000 organises and manages the project trials and demonstrations.
Project Roadmap The project will be implemented in 6 major phases: 1. The scenarios and user requirements definition and analysis, from T0 to T0+3, to refine the operational concept of operations of the system and to validate the exploitation interest for the users; 2. From this scenarios and requirements, the second phase will be dedicated to the system design process, from T0+3 to T0+6. The output of this phase will be the generic architecture of DYVINE, describing the system backbone (IT and Communications) and the technical solutions to integrate the various types of video sensors. This output will be validated by the system design review, held at T0+7; 3. The system components will be developed, from T0+3 to T0+19 (with anticipations where possible), in accordance with the system design achieved in the previous phase. These components are the following ones: * gateways to integrate the ground sensors, * gateways to integrate the mobile/airborne sensors, * the wireless communications solutions to link the sensors to the system backbone, * the data storage(s) * the software modules for the exploitation and fusion of the sensors data: image quality-compression module, identification and alarms modules, multi-sensor fusion and tracking module, sensors management module, situation awareness module; 4. The system components will be integrated into the system backbone. This phase has been defined from the beginning as a step by step integration into a factory integration platform. The components of phase 3 will be developed in two steps. A first version of the components, (which will not be finalised as far as the process is concerned but which will present stabilised interface) will be integrated into the platform at T0 +13 to identify and correct where necessary possible technical problems. The objective of this first integration is to mitigate the risks for the final integration by solving at an early stage the major technical issues. The final integration will be performed between T0+16 and T0 + 20; 5. In parallel to these technical works, dedicated tasks will cope with the legal and standardisation issues. These tasks will interact with the technical tasks by 1) defining the constraints and recommendations that the system will have to take into account and 2) giving a feed-back of the technical tasks in terms of system characteristics and performances to analyse if additional laws, specifications or standards should be proposed; 6. Finally the system will be tested in operational-like conditions to validate the technical performances and to validate the exploitability by the operational users. This phase will have two steps: a first step after the initial integration (T0+14) which will consist in a limited factory demonstration using essentially simulated data and a second step which will be real in-the-field trials after the final system integration (T0+21). These trials will take place in Valencia (Spain) and will be supported by the existing video sensors networks of the city.
Project Consortium The consortium is built around one major European industrial group EADS, involving two of its entities in the project: * EADS DS SA (France) and the * Common Research Centre (EADS CRC - Germany). The consortium comprises leading companies in the visual sensors and video processing domains: * REALVIZ (France) and * MARTEC (France). The industrial companies are complemented by research laboratories and universities: * CEA (France), * EPFL (Switzerland), * UPV (Spain), * University of Surrey (UK) and * University of Leuven - KUL (Belgium). The 112 of Valencia (Spain), the cities of Valencia (Spain), Miraflores (Peru) and Segrate (Italy) will form the user group and represent all the possible end-users of DYVINE. Each participant can be considered as a leader in its activity areas and together they make a group that definitely has the capability to achieve the project objectives.
Please notify: The content of this web site is owned by the consortium of the IST-project DYVINE which is partially funded by the European Commission within its 6th Framework Programme. The opinions published in this web site reflects exclusively opinions held by DYVINE and in no case represents an official statement by the European Commission.