This article was taken from the January 2012 issue of Wired magazine. Be the first to read Wired's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online.
Monitoring the movement and population of different species is crucial for wildlife conservation. It helps biologists to identify the creatures most at risk and to chart management programmes. This data has traditionally been gathered manually, with researchers counting visible specimens, nests or dung in a sample area. Video cameras could monitor wildlife remotely, but an eye was still required to trawl through footage and identify animals using a database.
Now researchers at the Fraunhofer Institute in Munich have developed a system that can rapidly process images of great apes in order to recognise them more quickly in the wild.
The program isolates individual faces and captures their biometric data. It supplements this with audio data collected from microphones -- automatically classifying sounds such as gorillas' chest-thumping and grunting. This allows the computer not only to identify the species, but also offers behavioural insights. Other animals could almost have been made for electronic recognition. Biologists at Princeton and the University of Illinois have developed an open-source "barcode scanner" called StripeSpotter, which reads and identifies zebras' unique black-and-white markings.
These developments result largely from the falling price of GPS devices, which allows conservationists to track the movements of our primate and ungulate friends, as well as toucans, cuckoos and bats, with relative ease. As Fraunhofer researcher Alexander Loos says: "These new technologies mean that human resources can be allocated to more complex tasks than filling out checklists."
Explore more: Big Ideas For 2012
This article was originally published by WIRED UK
