The researchers presented 4685 sounds firmly 1,387 bats in Mexico, the country where most bat species.
These recordings were entered into a computer program that is able to the cries and beeps of different types distinguished with artificial intelligence.
This reports the journal Methods in Ecology and Evolution .
Released
the British and Mexican biologists traveled half years with audio equipment through the jungle of Mexico and captured hundreds of bats to capture their screams. They managed to find 59 of the 130 bat species in Mexico. All captured animals were eventually released.
The computer program analyzed the sounds can be used according to lead researcher Veronica Zamora-Gutierrez to attribute bat sounds in the woods on specific species. In this way it can be determined which bats occur in an area without the animals have to be caught.
algorithm
“Until now it was difficult to do that, since bats make sounds that are very similar, “said Zamora-Gutierrez news on EurekAlert . “This problem we solved by developing an algorithm that bat species automatically recognizes.”
The program can even identify bat species for which no sounds are collected.
The software compares the cries of those animals with other sounds in the database. On the basis of the agreements can be estimated how large the bats that make the unfamiliar sounds and what species they belong
No comments:
Post a Comment