Thermography for facial tracking


Workload as a concept in the field of human factors is important because it helps us gain a better understanding of workplaces and how they can be designed to take into account human abilities.

The idea of workload seems intuitive. Most of us have experienced working at different levels. Despite this, quantifying it poses challenges and it remains an active area of research. Subjective measures, such as asking a person to estimate their level of workload after performing a task, rely on memory – and workload can be remembered incorrectly. Another popular technique is to assess task performance. The main disadvantage of this approach is that performance may not be sensitive to changes in workload.

Our research developed a third technique of workload assessment – physiological measures – and was aimed at tasks that have a dominant cognitive nature, such as piloting an aircraft.

This approach assumes that as more demands are placed on an operator, more physiological resources will be used. One advantage is that most physiological parameters are not under conscious control and can be recorded continuously in a non-intrusive manner. We chose to monitor face temperature using thermography, heart rate, breathing rate and pupil diameter, because these measures interfered the least with the task.

Facial thermography measures variations in surface temperature on the face using a thermal camera and uses facial landmark tracking algorithms to capture the temperature variation on a person’s face without restricting head movement.

Our research initially explored the human physiological response to workload in a laboratory setting. The study involved participants playing a computer game with varying levels of difficulty while their physiological signals were monitored. This approach allowed us to more accurately control the level of demand imposed on the participant while minimizing external influences.

The results demonstrated that facial thermography and pupil diameter can be used for non-invasive real-time estimation of workload. The most noticeable changes were observed in the nose area, which showed significant decreases in its temperature.

The research was later extended to a six degrees of freedom, high-fidelity helicopter simulator study, with commercial pilots as subjects, to test if highly trained individuals would have a similar physiological response to the general public. This study showed that it is feasible to deploy physiological monitoring such as facial thermography in an aircraft cockpit, and that pilots do show similar physiological changes to the general public. However, a similar study with a larger number of participants is needed to confirm these findings.

Future studies will look at applying these measures to different domains, collecting data from a larger population, exploring other measures such as functional near-infrared spectroscopy, which measures blood oxygenation levels in the brain, and making use of the latest developments in deep learning to analyze facial expressions. The research will increase our understanding of how humans interact with work environments, such as cockpits, in addition to informing the design of such workplaces. There is also high potential to apply these technologies in other sectors.

Share this story:

About Author


Ben has worked as a journalist and editor, covering technology, engineering and industry for the last 20 years. Initially writing about subjects from nuclear submarines to autonomous cars to future design and manufacturing technologies, he was editor of a leading UK-based engineering magazine before becoming editor of Aerospace Testing in 2017.

Comments are closed.