Injuries and Illness, Personnel Safety, Technology and Innovation

NSC: Computer Vision May Mitigate Fatality Risks

Computer vision­, or the use of artificial intelligence (AI) to analyze video, may help mitigate the risks of fatal injury in the workplace, according to the new white paper Using Computer Vision as a Risk Mitigation Tool from the National Safety Council (NSC). The publication was released October 27 as part of the NSC’s “Work to Zero” initiative.

The report evaluated findings from academic and professional journals related to computer vision, seeking to provide safety professionals with a greater understanding of computer vision’s ability to identify risks across a variety of environments. The white paper also collected case studies and interview data from reports that may point to the most promising trends and resources that employers leverage to prevent worker injury and death. 

Key findings included:

  • Computer vision is ideal for industries involving heavy machinery and extensive movement, such as construction, logistics, manufacturing, and warehousing, due to its ability to track and log data and instantly deploy information to predict when incidents may occur. 
  • Computer vision is especially effective in determining when personal protective equipment (PPE) is properly worn, such as employee compliance with wearing hard hats and high-visibility vests, as well as enables managers to mitigate unsafe situations.
  • The technology can be used to help monitor fatigue and other impairments during driving.  
  • Computer vision can also be helpful as a tool for employers to identify and prevent workplace violence by detecting anomalies in the workplace, such as unwanted guests, unusual behaviors, and weapons.
  • Computer vision can improve workplace health by monitoring ill employees and identifying others with whom they have interacted and objects they may have come into contact with.
  • Some computer vision systems can be trained to recognize best practices during emergency situations.

The Work to Zero team also uncovered limitations with current computer vision technology, such as image quality on closed-circuit televisions and AI software’s limited ability to operate in unfamiliar settings. The most effective object detection computer vision systems are based on deep learning, according to the report.

The NSC’s Work to Zero initiative is meant to address the fact that workplace fatality rates have remained steady over the past several decades despite government and private sector efforts to reduce serious injury and death on the job.

“Nationwide, 3.4 fatalities occur per every 100,000 full-time equivalent workers,” Paul Vincent, NSC executive vice president of workplace practice, said in a statement.

“While computer vision is being adopted by more organizations, the benefits are not widely understood or used to their full potential. This white paper highlights top research findings to make it easier for employers to leverage this technology to better identify root causes of workplace incidents and keep their employees safe on the job.”

The Work to Zero initiative, funded by the McElhattan Foundation, aims to eliminate workplace fatalities through the use of technology.

Other Work to Zero reports include:

  • “Drones for Working at Height and Confined Space Inspections”
  • “Proximity Sensors to Avoid Equipment Strikes at the Worksite”
  • “Virtual Reality and Augmented Reality for Hazardous Work Training”
  • “Wearables for Fatigue Monitoring”