• With no vaccine available for COVID-19 yet and a considerable volume of asymptomatic cases, the most effective form of control is to identify potential carriers of the virus and support them through self-isolation. One way to do so is to leverage thermal cameras to collect body temperature readings and analyse those temperatures to determine if there is an elevated temperature, not only a symptom of this disease but many other contagious illnesses as well. As a reference, the WHO advises anyone with a low-grade fever (37.3°C or more) to stay at home.

Following SARS and other infectious disease outbreaks, international standards around temperature screening using thermography were developed. Thermal imaging can provide an adjunctive temperature measurement of body surface temperatures and is acknowledged as far more effective than handheld devices.

The problem with temperature checks, whether done using a thermal camera, a contactless thermometer or a traditional mercury thermometer, is that everyone’s baseline temperature is different. The proper way to classify a fever is to measure a one-degree Celsius increase from one’s baseline. Doing this manually would be costly and impractical, so absolute thresholds are usually used instead.

How can Nirovision add value in this scenario?

• Action temperature results. By building a baseline temperature for each individual, Nirovision can more accurately determine if someone is safe to enter a workplace. Temperature measurements can trigger identity recognition-based actions, such as notifying the individual or the onsite medical officer of unusual readings or revoking access to a restricted area.
  
  
• Augment human capabilities. Thermal screening combined with facial recognition can track more people, more precisely and with far less effort than handheld devices. It also makes fully unmanned setups possible, removing the need to expose staff to potential health threats.
  
  
• Accelerate contact tracing. Contact tracing has historically been a manual process. Facial recognition technology can help identify anyone who might have had close contact with someone during their infectious window.
  
  
• Extend hardware capabilities. Similarly to the way we augment any IP camera with deep learning algorithms, the same could be achieved for bi-spectrum thermal devices, as their API capabilities, accuracy, precision and features vary significantly from manufacturer to manufacturer. Most of these cameras have been designed to adapt and be resilient to ambient temperature, but can’t measure nor manage individual baselines. Some of these cameras have facial recognition capabilities inbuilt, but with minimal functionality.
  
  
• Generate powerful insights. Temperature metadata in Nirovision allows for the analysis of trends over time, by itself but also in combination with any other identity-related metadata. These type of analytics can help improve workplace health and safety measures in the long term.
  
  We consider Nirovision as an identity platform because identifying someone extends beyond names and faces – identities encompass the groups people associate with, preferences, behaviours and any information pertinent to the subject. Following this line of thought, “body temperature” is a type of identity metadata.
  
  So today, we’re announcing the first release of our Nirovision+Body Temperature solution, with the following features:
  
  Multi-person instant measurement
  
  
• Low latency results powered by our on-premise server.
• Temperature metadata overlaid on the live video feed and in our Reception View for live assessment.
• New filter in the Activity view to only display recognition events that exceed a temperature threshold for further investigations.
  
  Smart alerts on body temperature
  
  
• Based on an absolute temperature threshold or individual baselines tracked over time.Send real-time notifications to anyone that needs to be informed.
  
  Contact tracing support
  
  
• Detailed view of other identities seen on-frame with an individual, including temperature information if available.
  
  
• Integrate with access control solutions for a fully touch-less, unmanned system.

Advanced encryption protects data to comply with global privacy laws. Implementing thermography for fever screening is not a trivial process. Still, if done responsibly by following the advice of certified personnel, it can help prevent the spread of infections in the workplace and protect your customers, contractors and employees.

It is essential to clarify that thermal reads only indicate the presence of elevated body temperature, but cannot diagnose an individual as having any medical condition. Procedures need to be put in place to allow a qualified medical practitioner to assess the individual upon an abnormal read.

• Set up a screening zone in an indoor area near the entrance to the site, to minimise the potential interactions of any infected individual. The screening zone should have minimal airflow, be climate controlled and point to a non-reflective background.
  
  
• Screen personnel as they walk through the designated area. If an abnormal read is detected, a medical officer or HSE professional with suitable PPE equipment needs to confirm via thermometer read and define a course of action.

For more information on Nirovision’s identity recognition and temperature analytics technology, please contact us.