Belgian IoT firm deploys Nokia machine learning tool for security streaming
Belgian smart-city security provider Room40 has announced a deal with Nokia to use the Finnish vendor’s machine learning tool for video analytics to discern anomalies in stream streams of video, audio and sensor information, and flag incidents in real-time. It is using the Nokia analytics solution to increase security and safety in various sites across Belgium.
Highway service stations are often the site of vandalism, theft and coordinated drug and human trafficking, noted Nokia. The cost of 24/7 video and audio monitoring, either by personnel or software, is prohibitively expensive. Room40 and VIAS Institute, the Belgian knowledge centre for road safety and security, are jointly tackling this issue on Belgian highways and other public spaces.
They are also to offer their security analytics solution, with the Nokia component, to construction companies to monitor productivity and safety on site.
Typically, up to 99 per cent of video, audio and sensor data is normal behaviour. Nokia’s Scene Analytics software uses machine-learning to establish what is ‘normal’, enabling it to track and record only irregular behaviour that might be connected to safety or security incidents.
This reduces the amount of data to process and more quickly and efficiently alerts personnel to incidents. It is also capable of extracting information from older, low-resolution CCTV cameras, as well as audio and other IoT sensors, which protects investment and lowers costs.
Harald Oymans, chief executive and co-founder at Room40, said: “In a world that changes so drastically almost every minute, it is of the utmost importance to be able to see what’s really there. Scenario based patterns only provide insight in what was, not in what is. The self-learning capability of the Nokia software enhances true insight in what’s really there, therefore enhancing the existing knowledge base with new information based patterns, which in turn enable the collection of the really needed data.”
Josee Loudiadis, vice president for enterprise analytics and IoT, said: “Real-life situations require an intelligent platform that can proactively detect, select, and track only relevant video streams for a variety of surveillance tasks. An important element of our Future X for industries strategy, the Nokia Scene Analytics software is a game-changing solution to the problem of security monitoring and will have wide-ranging applications in many other smart city and industrial applications. We are very proud to work with Room40 and VIAS to develop this security solution and help make Belgian highway service stations safer.”
The core of the technology is a patented (in the US and Israel) method for integrating statistical algorithms with ontological based input, to imitate processes of human-reading of texts, says Room40. The combined knowledge of seasoned subject matter experts regarding texts of the same domain used in training are applied to the processed texts through algorithms.
“This method is called ‘artificial intuition’,” it says. “A human reaches “intuitive” conclusions – even by superficial reading – regarding the authorship and intent of a given text, subconsciously deriving them from previous experiences with similar texts or from linguistic knowledge relevant to the text. Then as the human accumulates more information through other features (statements, spelling, and references) in the text, he either strengthens his confidence in the initial interpretation or changes it.”