Graymatics

A Smarter Look at Surveillance

Last week’s suicide bombing in Manchester, which was by far the worst attack on Britain since 2005’s tube and bus blasts in London.

Following Salman Abedi’s attack on an Ariana Grande concert, which resulted in 22 fatalities, the UK has promised to fast-track legislation that would give British security unprecedented access to private citizens’ internet activity, even encrypted messaging services such as WhatsApp which terrorists can use to communicate.

However, decrypting the messages would require tech companies to build a ‘back door’ into their systems which could create an entirely different problem: hackers could easily exploit this vulnerability to threaten private users’ digital security.

While investigations are ongoing, there is currently no evidence that the Manchester tragedy was the result of encrypted texts – so this may not be the ideal solution. Instead, we need to consider the matter of smarter surveillance.

Machine Learning at Work

A small city-state like Singapore can have up to six security cameras and multiple feeds on a single street corner – buthuman detection, analysis and response ratesmay not be enough to apprehend a culprit in time.

To solve this issue, Graymatics has tapped the rich possibilities of machine learning and big data analytics to develop smarter surveillance systems.

For example, Graymatics’ scalable solution integrates with existing CCTV infrastructure to identify individuals, objects and activities in real time. No matter how large the crowd, the software parses the big data at hand to produce detailed text profiles of everything in an image or video feed – of which the solution can monitor up to a thousand at a time.

Put simply, predictive technology can spotanindividual leaving a bag unattended in a crowded room, and raise the alarm to deploy security. The system goes beyond simple facial or full-body recognition: it can identify emotion, violence or just suspicious behaviour.

Using machine learning, the technology can continually incorporate new information with human guidance, enabling deeper and more wide-ranging visual analysis.

All this can be done 24/7, largely without human intervention – making it the perfect solution for security organisations strapped for manpower.

It doesn’t stop there. Graymatics and other companies at the forefront of this sector continue to explore new possibilities in security, such as thermal scans to detect explosive material in instances where an image feed alone will not suffice.

A World at Risk

For now, the unfortunate reality seems to be that Manchester needed more protection. While the city has roughly half a million inhabitants, most of Britain’s security efforts have been concentrated on London, the UK’s seat of power.

In fact, anycosmopolitan area with high population density is at risk of a terrorist attack. According to the United Nations, two-thirds of the world’s population will live in urban areas by 2050, and 41 cities will house at least 10 million people by 2030.

It is therefore imperative that governments work with the private sector and research institutions to refine their public surveillance infrastructure, keeping citizens safe and possibly preventingfurther tragedy.