To observe the animal’s specific behavior, scientists often install micro-cameras or other “bio-log” devices on wild animals,media reported. Now, a new technology can help these tools collect more data, as long as they are fully powered up when needed.
Because they are often mounted on small animals, the biological loggers themselves must be very compact. This means that they use relatively small batteries, so he can’t run for too long. Sometimes this limitation can be addressed by programming devices to record only a few predetermined periods of time per day. The problem, however, is that any behavior that occurs between these recording sessions is omitted.
Another option is for the device to record continuously, but the battery will soon run out — as you can imagine, before the target behavior of the study occurs.
In search of a solution, scientists at Osaka University have developed an artificial intelligence-based biometric that includes an accelerometer and a GPS unit. Based on the output of the two sensors, which consume relatively little power, the biometrics are able to determine when an animal begins to perform targeted behavior. It then activates the more power-hungry camera and keeps a record only for the while the behavior lasts.
The technology has been successfully tested on seagulls off the coast of Japan.
Joseph Korpela, lead author of the study, said the new method could detect black-tailed gulls’ foraging behavior 15 times as much as random sampling methods.