A framework for chainsaw detection using one-class and WSNs: poster abstract

The Amazon Rainforest degradation is a worldwide concern. The forest has been endangered by the illegal wood extraction without control even in the preservation areas. This human behavior endangers the flora and the animals that lose their natural environment. Due to the large geography extension, prevent these crimes with an unmanned aerial vehicle (UAV) is not always possible. The Wireless Acoustics Sensor Network (WASN) technology can alleviate this problem. WASN consists of a set of spatially distributed autonomous devices capable of monitoring the acoustic environmental conditions at different locations [1]. This is not a trivial task because the acoustic sensing requires high sample rates, preventing data transfer to further processing. In previous work, some approaches have been developed based on autocorrelation method [2], [3]. However, these methods are not suitable to recognize the soundscapes of the Amazon, which could include several natural and artificial sounds, such as: animal's calls, weather noises or boat engines.