Every year, South East Asian (SEA) region has suffered from haze pollution significantly caused by forest fires and agricultural-related burning. Mitigating this problem requires a robust system that can monitor haze and air pollution in real time across the region. The SEA-HAZEMON is an IoT platform that consists of low-cost air quality sensors and several cloud services. The platform also consists of a forest fire detection model based on particulate matter (PM2.5) and Carbon Monoxide (CO) concentration. In our trial, the early warning messages were timely sent to navigate the local forest fire authorities via short messaging applications (i.e., Telegram and Line). The notification messages were analyzed together with real forest fire incidents that occurred in the northern part of Thailand during the fire burning season. The results showed that there were 367 fire events detected in April 2022 that achieved an accuracy of 84%. Our study also discovered the effect of humidity which reduced the accuracy of our model by approximately 13%.
View source
