Search and rescue study

Preliminary results summary

The Human Factors group at the NRC Flight Research Lab wishes to thank all those who participated in our online experiment as part of our Search and Rescue Study. The team gathered data from almost a thousand participants between mid-November 2015 and the end of January 2016. We have begun analyzing the data to see how well individuals can spot a downed aircraft from nighttime infrared sensor images. This provides valuable information needed to support and improve search and rescue operations and to reduce search times. The data collected will help shape the future of search and rescue missions in Canada, particularly in the future when sensor imagery can be reviewed by individuals working remotely.

In this experiment, participants viewed 100 images captured from an infrared sensor camera and were asked to judge if there was an aircraft present in each one. Of these images, 80 contained an aircraft where the degree of contrast with its background had been altered (in terms of its relative brightness); the remaining 20 images did not contain an aircraft.

Results reveal that on average, participants detected approximately 57 aircraft or 70.91 % of the presented aircraft. The two figures below represent the percentage of aircraft detected (Figure 1) and the average reaction time of the participants (Figure 2) as a function of the aircraft's contrast level. As predicted, participants had a harder time detecting low contrast aircraft, than they did with high contrast aircraft (Figure 1). Additionally, it took participants longer to respond when the contrast level between aircraft and terrain was lower (Figure 2).

Bar plot representing the percentage of correctly identified aircraft when present (hits)  as a function of contrast levels.
Figure 1. Bar plot representing the percentage of correctly identified aircraft when present (hits) as a function of contrast levels.
Bar plot representing participant response time in milliseconds as a function of contrast levels
Figure 2. Bar plot representing participant response time in milliseconds as a function of contrast levels.

This study has been successful in gaining information on the ability to use new technologies in airborne search and rescue and has benefited from the growing popularity of Citizen Science activities facilitated by the Web. Citizen Science projects such as this rely on the participation of interested individuals to record observations or do experiments, contributing data and benefiting both participants and researchers. We would like to thank all of those who participated in this study once more as it has enabled us to collect data from a much larger sample than would be possible during traditional laboratory experiments.

Although official data collection is now complete, the experiment is still available online for anyone wishing to try it.

If you would like to learn more about the involved collaborators in this research please visit the Working and Travelling on Aircraft Program website.

Date modified: