New tech leverages edge computing, biometrics and temperature screening to reduce COVID-19 exposure
A safe return to work is requiring a shift in workplace processes and facilities, and a growing number of vendors are offering technology such as thermal and mask detection systems that leverage biometrics and edge computing to give management real time insights into worker health.
Among the new guidance to reduce COVID-19 exposure being provided by the national Association of Manufacturers are setting up a worker checkpoint outside the facility, taking temperatures with a non-touch laser device, and providing for worker privacy while receiving tests by not record temperature readings or employee names.
Setting up technology to enable these new procedures is sometimes a challenge, but companies like Foghorn, Aaeon and Birlasoft are among those aiming to ease the installation of thermal detection systems, and powerful edge computing devices can be part of the solution.
Edge computing can be simply defined as the delivery of computing capabilities close to the source of data in order to improve the performance, operating cost and reliability of applications. Edge-enabled systems are needed because the requirements of worker scanning require that data is kept locally, while the ability to give real-time results means that monitoring systems cannot wait for data to move to a far-away data center for processing.
FogHorn, a developer of edge AI software, is offering an answer to the challenge with pre-integrated software. A new product line, the Lightning Health & Safety Solution (LHSS), takes the company’s real-time analytics capabilities and provides machine learning that are pre-trained for actions such as temperature detection, cough detection, hand washing monitoring, social distancing monitoring, and mask / facial covering detection. Mask detection is typically carried out with video analytics and biometrics. The systems can take feeds from RGB and infrared cameras to determine a person’s temperature to within .25 degrees Celsius, executives said.
The integration of standard models greatly speeds the implementation of monitoring capabilities, but it is also important to be able to easily tailor the systems to specific needs. FogHorn LHSS offers a dashboard that allows customization for different industrial environments, meaning that an industrial company that requires hard hats and safety vests can monitor the use of safety equipment while also adding face mask detection, for example. FogHorn’s software can trigger SMS or email alerts of compliance issues, and management can track compliance trends via the dashboard.
Embedded AI-enabled edge device maker Aaeon has responded to market needs with a device that incorporates NVIDIA Jetson TX2 chips with 256 CUDA cores, enabling support for multiple types of AI frameworks including as TensorFlow, Caffe2 as well as custom AI inference software. The latter ability is important too, as companies are needing to update algorithms to accommodate new requirements around social distancing and mask requirements, for instance.
Aaeon, a company in the ASUS Group known for tech manufacturing, said that its small, fanless device is currently being used in hospitals and is powerful enough to identify and measure the temperature of up to 240 people per minute and can allow for the monitoring of multiple entrances. That translates into fewer staff needed to monitor the workforce. At the same time, facial recognition models can accurately identify people and features, allowing hospital staff to quickly locate and address anyone with an abnormally high body temperature.
Another company, Birlasoft, a provider of consulting, engineering and systems integration services, is aiming to tie edge devices in with analytics and HR systems which are usually done in in centralized cloud services. The company started offering a new thermal scanning offering that allows customers to easily integrate a range of different thermal scanning systems with software that enables companies to comply with local regulations for track and trace by recording data and producing reports. Additionally, data gathered from the “edge” (i.e., thermal scanners) is gathered and analyzed by analytics software to show trends in employee health over time and across different facilities. This is useful because, while edge computing allows for quick local reporting of potential health issues, centralized analysis provides for insights that can help management change practices and policies at other facilities to prevent an outbreak, for example.
Learn more about edge computing at Edge Industry Review.
Aaeon | artificial intelligence | biometrics | biometrics at the edge | edge computing | facial recognition | fever detection | monitoring | thermal | video analytics