The security of a large public infrastructure is currently managed by traditional systems based on security cameras connected to a security con- trol center. The cameras are remotely controlled by the security personnel, and the acquired video streams are saved on a data server, without any processing or analysis. The most important limitation of this approach is the complete absence of real-time automatic processing of the video stream, which could provide important information directly when the monitored events are occurring. Real-time video analysis performed by the security personnel is not reliable, and the reliability level drops proportionally with the number of video streams: humans are not “designed” for long-term focus and attention. The human brain is very efficient when reacting to stimula- tion, but it is not efficient in keeping constant control over long periods of time, especially when nothing interesting is happening. The attention of a professional operator drops on average by 60% only after 30 min of video streaming surveillance. Attention is further lost when an event occurs and the security operators are focused on that single event, with the risk of miss- ing the identification of other important events. Existing solutions allow the analysis of the video stream after the events occurred, only during investigations, and rely entirely on the security per- sonnel expertise. Very frequently, security companies incur high expenses just to provide unreliable surveillance. The idea of adopting biometric security technologies that are natively compliant with the SHIELD framework introduces a new secure paradigm in surveillance that is based on the concept of prevention, which is imple- mented using a proactive video stream analysis that allows near-real-time event identification.

Biometric security domain

Antonio Abramo;
2018-01-01

Abstract

The security of a large public infrastructure is currently managed by traditional systems based on security cameras connected to a security con- trol center. The cameras are remotely controlled by the security personnel, and the acquired video streams are saved on a data server, without any processing or analysis. The most important limitation of this approach is the complete absence of real-time automatic processing of the video stream, which could provide important information directly when the monitored events are occurring. Real-time video analysis performed by the security personnel is not reliable, and the reliability level drops proportionally with the number of video streams: humans are not “designed” for long-term focus and attention. The human brain is very efficient when reacting to stimula- tion, but it is not efficient in keeping constant control over long periods of time, especially when nothing interesting is happening. The attention of a professional operator drops on average by 60% only after 30 min of video streaming surveillance. Attention is further lost when an event occurs and the security operators are focused on that single event, with the risk of miss- ing the identification of other important events. Existing solutions allow the analysis of the video stream after the events occurred, only during investigations, and rely entirely on the security per- sonnel expertise. Very frequently, security companies incur high expenses just to provide unreliable surveillance. The idea of adopting biometric security technologies that are natively compliant with the SHIELD framework introduces a new secure paradigm in surveillance that is based on the concept of prevention, which is imple- mented using a proactive video stream analysis that allows near-real-time event identification.
2018
978-1-138-04275-9
File in questo prodotto:
File Dimensione Formato  
BSD.pdf

non disponibili

Descrizione: Lavoro finale
Tipologia: Versione Editoriale (PDF)
Licenza: Non pubblico
Dimensione 3.72 MB
Formato Adobe PDF
3.72 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11390/1194661
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact