Wildfire Detection Using Sound Spectrum Analysis Based on the Internet of Things

Integrating sensory ecology and predator-prey theory to understand animal responses to fire

Fire regimes are changing dramatically worldwide due to climate change, habitat conversion, and the suppression of Indigenous landscape management. Although there has been extensive work on plant responses to fire, including their adaptations to withstand fire and long-term effects of fire on plant communities, less is known about animal responses to fire. Ecologists lack a conceptual framework for understanding behavioural responses to fire, which can hinder wildlife conservation and management. Here, we integrate cue-response sensory ecology and predator-prey theory to predict and explain variation in if, when and how animals react to approaching fire. Inspired by the literature on prey responses to predation risk, this framework considers both fire-naïve and fire-adapted animals and follows three key steps: vigilance, cue detection and response. We draw from theory on vigilance tradeoffs, signal detection, speed-accuracy tradeoffs, fear generalization, neophobia and adaptive dispersal. We discuss how evolutionary history with fire, but also other selective pressures, such as predation risk, should influence animal behavioural responses to fire. We conclude by providing guidance for empiricists and outlining potential conservation applications.

Fight Fire with Fire: Detecting Forest Fires with Embedded Machine Learning Models Dealing with Audio and Images on Low Power IoT Devices

Forest fires are the main cause of desertification, and they have a disastrous impact on agricultural and forest ecosystems. Modern fire detection and warning systems rely on several techniques: satellite monitoring, sensor networks, image processing, data fusion, etc. Recently, Artificial Intelligence (AI) algorithms have been applied to fire recognition systems, enhancing their efficiency and reliability. However, these devices usually need constant data transmission along with a proper amount of computing power, entailing high costs and energy consumption. This paper presents the prototype of a Video Surveillance Unit (VSU) for recognising and signalling the presence of forest fires by exploiting two embedded Machine Learning (ML) algorithms running on a low power device. The ML models take audio samples and images as their respective inputs, allowing for timely fire detection. The main result is that while the performances of the two models are comparable when they work independently, their joint usage according to the proposed methodology provides a higher accuracy, precision, recall and F1 score (96.15%, 92.30%, 100.00%, and 96.00%, respectively). Eventually, each event is remotely signalled by making use of the Long Range Wide Area Network (LoRaWAN) protocol to ensure that the personnel in charge are able to operate promptly.

Audio-Based Wildfire Detection on Embedded Systems

The occurrence of wildfires often results in significant fatalities. As wildfires are notorious for their high speed of spread, the ability to identify wildfire at its early stage is essential in quickly obtaining control of the fire and in reducing property loss and preventing loss of life. This work presents a machine learning wildfire detecting data pipeline that can be deployed on embedded systems in remote locations. The proposed data pipeline consists of three main steps: audio preprocessing, feature engineering, and classification. Experiments show that the proposed data pipeline is capable of detecting wildfire effectively with high precision and is capable of detecting wildfire sound over the forest’s background soundscape. When being deployed on a Raspberry Pi 4, the proposed data pipeline takes 66 milliseconds to process a 1 s sound clip. To the knowledge of the author, this is the first edge-computing implementation of an audio-based wildfire detection system.

Light-Weight Student LSTM for Real-Time Wildfire Smoke Detection

As the need for wildfire detection increases, research on wildfire smoke detection combining low-cost cameras and deep learning technology is increasing. Camera-based wildfire smoke detection is inexpensive, allowing for a quick detection, and allows a smoke to be checked by the naked eye. However, because a surveillance system must rely only on visual characteristics, it often erroneously detects fog and clouds as smoke. In this study, a combination of a You-Only-Look-Once detector and a long short-term memory (LSTM) classifier is applied to improve the performance of wildfire smoke detection by reflecting on the spatial and temporal characteristics of wildfire smoke. However, because it is necessary to lighten the heavy LSTM model for real-time smoke detection, in this paper, we propose a new method for applying the teacher–student framework to deep LSTM. Through this method, a shallow student LSTM is designed to reduce the number of layers and cells constituting the LSTM model while maintaining the original deep LSTM performance. As the experimental results indicate, our proposed method achieves up to an 8.4-fold decrease in the number of parameters and a faster processing time than the teacher LSTM while maintaining a similar detection performance as deep LSTM using several state-of-the-art methods on a wildfire benchmark dataset.

A Framework for Agricultural Pest and Disease Monitoring Based on Internet-of-Things and Unmanned Aerial Vehicles

With the development of information technology, Internet-of-Things (IoT) and low-altitude remote-sensing technology represented by Unmanned Aerial Vehicles (UAVs) are widely used in environmental monitoring fields. In agricultural modernization, IoT and UAV can monitor the incidence of crop diseases and pests from the ground micro and air macro perspectives, respectively. IoT technology can collect real-time weather parameters of the crop growth by means of numerous inexpensive sensor nodes. While depending on spectral camera technology, UAVs can capture the images of farmland, and these images can be utilize for analyzing the occurrence of pests and diseases of crops. In this work, we attempt to design an agriculture framework for providing profound insights into the specific relationship between the occurrence of pests/diseases and weather parameters. Firstly, considering that most farms are usually located in remote areas and far away from infrastructure, making it hard to deploy agricultural IoT devices due to limited energy supplement, a sun tracker device is designed to adjust the angle automatically between the solar panel and the sunlight for improving the energy-harvesting rate. Secondly, for resolving the problem of short flight time of UAV, a flight mode is introduced to ensure the maximum utilization of wind force and prolong the fight time. Thirdly, the images captured by UAV are transmitted to the cloud data center for analyzing the degree of damage of pests and diseases based on spectrum analysis technology. Finally, the agriculture framework is deployed in the Yangtze River Zone of China and the results demonstrate that wheat is susceptible to disease when the temperature is between 14 °C and 16 °C, and high rainfall decreases the spread of wheat powdery mildew.