Matteo Capelli: "We Can Integrate Sensors with AI"
Commercial Director, Matteo Capelli, at Cias talks about video analytics and why it has limitations in external perimeter detection, as well as real time communication between CCTV and PIDS. GIT SECURITY has asked leaders in Perimeter Protection to give a concrete outlook.
What do you think was the greatest game-changing technology for the security industry in the last 20 years if you had to choose one?
Matteo Capelli: In my opinion, video analysis represents the game-changing technology of the last twenty years. The reasons for this lie in the evolution of CCTV, increasing its fields of application, such as traffic control, license plate recognition, industrial automation, indoor and outdoor applications or for intelligence on a large scale as in China. Video analysis boosted the expansion of CCTV and will also represent its future as long as it integrates more technologies. In fact, I believe that it has objective limitations of reliability in external perimeter detection, for example, in bad weather conditions such as heavy rain, fog, snow, sandstorm, etc. Only the synergy between different technologies enables the most efficient intrusion detection performance.
“The next ten years will be dominated by AI applications.” Why would you agree or disagree?
Matteo Capelli: AI will dominate the coming years. In our sector, it will represent the element of active integration of multiple technologies, as well as a self-improvement driver for each sensor or process. The situation is like the IT revolution of the 80s, the only difference is the speed factor, since AI will have a much faster impact and will affect many more areas, including cultural and social ones. Of course, as history teaches, every invention or innovation can incorporate threats. It's good that a strong debate on possible risks and limitations in application is already underway. As security experts, we took this chance to innovate our technology and in 2022 we developed Synapses, a new “super-sensor” system with embedded AI. This self-adaptive system with video analytics and deep learning criteria enables the real time communication between CCTV and PIDS: intrusion analysis is dynamically adapted to the real site conditions, distinguishing different targets in the video flow. It maximizes detection performance outdoor.
Your company will have a great impact on the industry in the next ten years because…?
Matteo Capelli: Our company will play its role in the security industry in the coming years thanks to its long-term expertise and constant drive for R&D. The risk in our sector is that new players advance in the software field forgetting or presuming that they can bypass the laws of physics. One example is video analytics: if applied in indoor environments it can achieve a great performance, but in outdoor environments the camera is like an eye that can be easily blinded by natural or artificial events. In that moment of need, we will play a fundamental role knowing how to reconcile the laws of physics with the most modern and advanced analysis software. We can integrate sensors with AI solving the challenge to provide a very high security degree.
