When people get lost in the woods, it is of utmost importance that efforts be made to find them as quickly as possible because things become more dangerous the longer they say there. Are the tools to organize and carry out search and rescue operations with efficiency and swiftness available to us?
It may soon. Currently, most search and rescue operations in the forest are carried out with helicopters. The officials go around the possible locations of the lost people in the chopper and search for them. Thermal imaging cameras are also used to aid their search.
These cameras highlight the variances in body temperatures which allows the rescue officials tell the people from their environment. The devices sometimes get confused usually during the hot weather and surrounding foliage have increased the temperature.
All these facts were taken into consideration by the researchers who decided to design drones to be specifically good at identifying humans from their surroundings. This was achieved making use of a deep learning application to enhance the images collected by the drones.
The researchers wrote in their study published in Nature Machine Intelligence, ‘In the future, rescuing lost, ill or injured persons will increasingly be carried out by autonomous drones. However, discovering humans in densely forested terrain is challenging because of occlusion, and robust detection mechanisms are required. We show that automated person detection under occlusion conditions can be notably improved by combining multi-perspective images before classification.’
‘Here, we employ image integration by Airborne Optical Sectioning (AOS)— a synthetic aperture imaging technique that uses camera drones to capture unstructured thermal light fields— to achieve this with a precision and recall of 96% and 93% respectively’. This is quite impressive when likened to the 25% attained by traditional thermal imaging.
Best part of this? The research team says the device is ready for use. We know many institutions will be eager to get their hands on this and we are too!
By Marvellous Iwendi.
Source: Interesting Engineering