Unusual warning system against forest fires
Bird singing is not only a great sound to help us relax. The acoustic signals generated by the birds also carry information about various events taking place in the area. A research team from Indonesia, Brunei and Malaysia has developed an Indonesian forest fire warning system using this phenomenon. A machine learning algorithm based on convolutional neural networks (CNN) was designed. The system detects changes in the singing of birds caused by the reaction of these animals to a fire. The effectiveness of correct fire identification has been demonstrated at the level of 96%.
Permana, Silvester Dian Handy, et al. “Classification of bird sounds as an early warning method of forest fires using Convolutional Neural Network (CNN) algorithm.” Journal of King Saud University-Computer and Information Sciences (2021).