Wednesday, 21 Feb 2018
You are here:

Subscribe to our Newsletter

How many eyes has a typical person? (ex: 1)

Twitter Feed

EMail Print


Mathieu Einig (UoR) visiting THALES
26.03.2012 - 09.04.2012
Internal Visits


The first aim of the visit to Thales was to gain insights regarding the integration of Video-analytics algorithms which will be developed within VideoSense into the Thales’ video surveillance framework. Their framework enables the creation of large scale networks with plug and play modules, and provides evaluation capabilities and comparison between different algorithms for similar techniques. As a result, some basic insights into how to integrate individual algorithms were gained. However, Thales’ framework is still at the design stage and is therefore likely to change. Therefore, more work will be needed in order to cope with changes in the architecture.
Another aspect of the exchange was to understand the needs and the methodology of the industrial sector in terms of video surveillance and privacy protection. For example, surveillance systems that display a schematic view of the scene are becoming quite popular as they allow a higher level of privacy. On the other hand, traditional Automatic Number Plate Recognition systems are still in high demand even though such systems are already widespread and can cause some high privacy risks.
Finally, the last aspect was to get information regarding the French legal context of video surveillance, which is much more restrictive than in the UK. For example, it is compulsory to blur entrances and windows of buildings that are not part of the surveillance but could be visible. There are also regulations regarding the minimum quality of video footage (image resolution, field of view, etc.) that need to be respected. The details of these regulations were listed in D7.1: Stakeholders Needs and Requirements.
Another visit to Thales will take place in the near future in order to conclude Thales’ needs and requirements for their video surveillance framework. This will allow us to develop the Video-analytics algorithms accordingly, for Thales specific use-cases.