More researchers are turning to artificial intelligence (AI) to monitor the environment and support endangered creatures. Unlike traditional methods that can damage the environment or require considerable time and physical work, AI has the ability to quickly and effectively deal with large quantities of data(数据).
Creatures are disappearing at a speed thousands of times faster than that millions of years ago, with one million creatures endangered. The United Nations set a goal in 2020 to safeguard at least 30% of Earth's land and oceans by the end of 2030. AI is"imperfect" but could make important discoveries faster and earlier, says Nicolas Miailhe, founder of The Future Society, an organization that aims to better manage AI. "We very much need skilled humans to design Al models and check and collect data. "
A type of Al model called a convolutional neural network (CNN) has been developed to recognize bird sounds. The CNN is able to pick out 75 kinds of the birds the researchers have. "Our results prove that A1 is ready for recognizing creatures in the wild from sound, " says Maller, the team leader. "All that is needed now is more training data collected by humans. "
Besides monitoring creatures in real time, Al can be used to model the impacts of human activities on the environment and rebuild historical changes. Researchers have used Al to discover how a century's environmental damage in a river has led to creature loss.
Although it is well-known that human activities have resulted in creature loss in rivers and lakes, little is known about which activities have the largest impact. "AI will do soon, " says Luisa Orsini, a researcher from the University of Birmingham.