From the cloud to the edge – smart IntegraTion and OPtimisation Technologies for highly efficient Image and VIdeo processing System
The FitOptiVis project was funded by the European Community as part of the Horizon 2020 programme – EU.22.214.171.124 – ECSEL.
ECSEL Joint Undertaking is a public-private partnership, which provides that participants in funded research projects receive contributions from both the European Community and the National Authorities (in Italy, the Ministry of Education, University and Research).
The consortium is made up of 30 partners from five European countries (The Netherlands, Czech Republic, Finland, Spain and Italy) and is coordinated by Philips. The project aims to increase the physical safety of sensitive infrastructures and the role of SAT is that of pilot of the Italian group, which consists of four small and medium companies (Small and Medium Enterprise – SME according to the classification of the EC) and 3 universities.
Using video surveillance equipment and artificial intelligence software to analyse the images collected, people are identified inside sensitive areas and then the behaviour of people who are not recognized, and could therefore be potentially dangerous, is analysed.
The process uses high-potential equipment that can transmit and analyse video streams by working in an innovative manner, considerably reducing the consumption of electricity and bandwidth, while maintaining high image quality.
Started in June 2018, the first year mainly concentrated on the definition of use cases and on the requests of infrastructure operators, as well as on the definition of the perimeters within which the software packages that allow us to achieve the set objectives will be developed.
As already mentioned, Società Acquedotti Tirreni is the pilot for the Italian use case, and has made its well zones available to its partners so that they may install the equipment and software that will be tested. Despite the delays caused by the pandemic, at the beginning of summer 2020 Società Acquedotti Tirreni completed the installation of the cameras in its main well area (Ressora Sud) and is now testing the facial recognition and behaviour analysis software on the images being collected.
The aim is to perfect an automatic, low-cost system (without dedicated personnel) that can generate an alarm when an analysis of the images collected results in a suspected threat to the security of the site under protection. In order to make the system easier to use, an interface is being implemented with the information systems already used by our company, so that part of the IT infrastructure already deployed can be used.