Preventing Deforestation using IoT

Abstract

Forests play a critical role in preserving the environment and combatting global warming, but they are continuously being depleted by human actions such as deforestation and wildfires. The proposed system introduces and assesses a framework that uses audio event classification to automatically detect illegal tree-cutting activity in forests. The proposed solution involves the use of ultra-low-power small devices that integrate edge-computing microcontrollers and longrange wireless communication to cover vast forest areas. To minimize energy usage and resource consumption while achieving widespread detection of illegal tree-cutting, the system recommends an efficient and precise audio classification technique that utilizes convolutional neural networks, specifically designed for resource-constrained wireless edge devices. Keywords: Machine learning, Edge computing, Audio classification, LoRa, Illegal tree cutting, Deforestation, Internet of things, Wireless sensor network, Microcontroller