Credit: CC0 Public domain
A team of researchers from Imperial College London, the University of Sydney and Cornel University discovered that it is possible to create a soundscape from noise in the natural environment using machine learning algorithms. In his article published in procedures of the National Academy of SciencesThe group describes how land managers can use such soundscapes to protect natural environments.
In many parts of the world, natural environment areas have been reserved by governments as a way to protect certain ecosystems: forests and national parks are excellent examples. But such areas generally cover large tracts of land, making it difficult to manage. The main concerns are entities that choose to ignore protection laws, such as poachers and those engaged in illegal logging. In this new effort, the researchers looked for a way to detect such activity in a much less labor-intensive way.
The researchers began their work by noting that most natural environments have a particular sound. There are certain birds that live in a certain area, for example, or land animals, each of which produces distinctive sounds. And then there are the sounds that trees make when the wind blows or the sounds of insects or babbling of water in a stream. They also noted that the sounds in a certain area change according to the time of day. They took all of these factors and fed them into a computer that runs machine learning algorithms that let you learn what a certain area sounds like. The researchers describe it as teaching machines to recognize the soundscape of a given area. Then they installed microphones in the same area and fed the sounds that were captured by the same machine; If the machine detected sounds that did not fit in with the normal soundscape for a region, such as a chainsaw or gunshots, it would send an alert.
The researchers found that the system was accurate in their alerts, noting that the system could be used in a wide variety of settings. They further suggest that such systems could be implemented worldwide as a means of helping land managers detect illegal activities in real time.
Noises from human activity can threaten the New England freshwater soundscape
Sarab S. Sethi et al. Characterize soundscapes in various ecosystems using a set of universal acoustic characteristics, procedures of the National Academy of Sciences (2020). DOI: 10.1073 / pnas.2004702117
© 2020 Science X Network
Citation: Use of neural network soundscapes to protect natural environments (2020, July 17) retrieved on July 18, 2020 from https://phys.org/news/2020-07-neural-network-soundscapes-natural-environments .html
This document is subject to copyright. Other than fair dealing for private study or research purposes, no part may be reproduced without written permission. The content is provided for informational purposes only.
