How new applications for security cameras could help flatten the curve

Everyone should know by now that social distancing is the key to preventing, or at least slowing down, the transmission of COVID-19 from person to person. According to the World Health Organization (WHO) social distancing buys us time because it slows down the infection rate - it helps flatten the curve. States and governments need time to prepare their health systems for a growing number of severe corona cases.

Source: WHO

Please find further information and advice for public health at www.who.int

In contexts where social distancing can only be realized to a limited extent, such as in hospitals, grocery stores or public transport, the use of protective equipment is crucial to ensure public safety. Unfortunately, both social distancing and the use of protective clothing and masks are not always respected as much as they should be.

As a final consequence, adherence to national regulations such as contact bans and curfews must be paid attention to. Today, we will tell you how IoT technology can make a valuable contribution to achieving this task.

How IoT cameras can help ensure adherence to regulations

Cameras in the IoT, equipped with powerful processors and AI apps, can support safety authorities to better monitor the adherence to safety rules in public and working spaces. Surveillance systems analyze video data in real-time and trigger actions to ensure safety.

The following examples show how smart devices in IoT could help slow down the spread of COVID-19.

1. Social distancing in public and working spaces

COVID-19 transmits rapidly where many people are gathered in confined spaces. For this reason, people in supermarkets, for example, should keep sufficient distance at checkouts. Smart security cameras can measure the density of groups or queues at checkouts and may give an alarm when people are standing too close together. This is also applicable to other areas, public spaces, buildings or working environments.

The following example shows how smart security cameras detect social distance even when people are in motion.

Source: Camio

Unleash live's user interface shows how smart cameras analyze video data in real-time. In addition to live video, density of people is measured and alerted when critical limits are exceeded. Safety executives receive all relevant information in a nutshell.

Source: Unleash live 

2. Ensure the wearing of protective clothing and masks

Where social distancing reaches its limits, protective clothing must prevent the transmission of viruses. This applies especially to healthcare facilities, public transport and grocery stores. Video analysis in security cameras can detect whether people show defined characteristics, such as wearing masks or gloves, or not.

If persons lack protective clothing, for example when entering hospitals, camera systems could automatically alert security personnel or block entrances. By connecting smart cameras with other systems in IoT, safety processes can be fully automated.

Source: Geutebrück

3. Detect people with disease symptoms

Thermal imaging is booming because it is a promising technology to help detect people with typical COVID-19 symptoms such as fever. This technology is not only used in camera systems, but also mobile devices such as smart glasses.

Thermal imaging applications are currently facing the challenge of developing 100% accuracy of real-time video analysis, but are getting closer, as the following example demonstrates in a very impressive way.

Source: Seek Thermal

Technology insight: Object recognition

The fight against COVID-19 is not about using IoT technology to identify people, but about detecting human behavior and characteristics that do not comply with regulations. Do people wear protective masks as required? Do they keep distance to the next person? Do they show symptoms of illness?

All these scenarios involve physical states that can be analyzed by video technology without knowing a person's identity, so privacy is not compromised. The applications we have shown above consider people as objects and, simply put, compare individual characteristics with desired characteristics. This is what object recognition is all about.

AI Video analysis detects protective clothing in hospitals, Source: Martin Peniak

Smart security cameras use machine learning to analyze and interpret video data. Software apps installed on the devices provide the ability to distinguish objects from people. In this way, cameras detect, for example, whether people are wearing protective clothing in hospitals, or not.

Machine learning models are trained to distinguish between people with and without protective clothing in real-time and enable cameras to take on a variety of safety-related tasks.

If you like to know more about how Object Recognition will boost video surveillance in the near future read our article with examples from retail, public transportation and security.

TL;DR

Every crisis offers new opportunities to change things for the better. COVID-19 will help boost technologies that make our lives safer. Companies in all industries will gain experience with new technologies and further digitize their businesses. Camera systems that become smarter through innovative software and AI can help make public life safer and business more efficient.

At Azena, we have built an open platform that brings together manufacturers, integrators and software developers to create the best solutions for every future challenge.