How artificial intelligence video analytics can help businesses comply with COVID-19 restrictions

Published: Thursday, 19 November 2020 09:35

As restrictions to protect the public from the spread of COVID-19 continue to adapt and change, it is vital to review and assess how social distancing measures can be effectively enforced within business environments. Chris Bishop looks at how AIVA (artificial intelligence video analytics) is proving effective in this area.

AIVA can be applied to existing CCTV networks to autonomously identify instances of non-compliance towards wearing face coverings as well as people grouping rules. Such technology uses sophisticated algorithms to determine breaches by recognising when a particular scene does not comply with set criteria (e.g. when two people are within a one-metre proximity).

The data gathered is analysed in real-time and is flagged to the relevant authorities or individuals when detected, meaning that more informed decisions can be made, and suitable courses of action can be taken. The potential cost-saving for businesses is significant too. Firstly, this type of technology makes use of existing CCTV camera networks; and secondly an algorithm crunches all of the data instead of it having to be done manually.

Consider the ubiquity of cameras in offices and shops and then tailoring the feeds to detect instances of social distancing breaches. With recent research showing that London has 627,727 cameras for 9.3 million residents (the equivalent of 67.5 cameras per 1,000 people), it is clear to see how existing infrastructure can play a key role if used to its full capability.

Flexible solutions for unprecedented times

As restrictions are constantly changing, the adoption of AI technologies is becoming a go-to solution for many businesses. They can be quickly adapted and tailored to identify different circumstances from the available data, meaning costs do not increase if the guidelines change.

The insights gained from applying AIVA solutions can also be applied practically based on the current restrictions in place. For example, detecting groups of people and automatically recording any instance when this occurs so businesses can monitor behaviour and tailor communications and guidance accordingly, as well as identifying where one-way systems, protective screens and hand sanitisation stations should be placed.

Perhaps most importantly, such systems can be used to identify individuals who have tested positive for COVID-19 in a group setting. People can be identified through multiple camera feeds by identifying features such as their clothes, height, and movements, thereby ensuring the privacy of the individual is withheld. This provides a much more efficient and quicker way to identify the individuals that the infected person has interacted with, instead of manually sifting through hours of footage.

As businesses struggle with the uncertainties of today, they should take some comfort from the availability of technology to support them as they adjust. AI has long been touted as the technology set to revolutionise life as we know it, and now it has the chance to unlock its full potential and protect people in a world significantly impacted by COVID-19.

The author

Chris Bishop is Sales Director APAC & Marketing Director, Ipsotek