| Title |
A Cnn-Based Audio Sensor For Rainfall Estimation: Implementation On Embedded Board |
| ID_Doc |
682 |
| Authors |
Russo M.; Puglisi V.F.; Avanzato R.; Beritelli F. |
| Year |
2021 |
| Published |
Proceedings of the 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2021, 2 |
| DOI |
http://dx.doi.org/10.1109/IDAACS53288.2021.9660891 |
| Abstract |
At present, one of the main issues of interest is climate change. It triggers calamitous phenomena (fires, hydrogeological instability and many others) that put crops, vegetation and even human beings at risk. Considering the phenomenon of hydrogeological instability, there is a wide range of smart application contexts (for example, smart city, smart road and smart agriculture), which require a real-time rainfall level monitoring, recognition and classification system. This system is supposed to provide accurate and real-time estimates of rain intensity and alert those in charge in order to reduce the risks associated with the various application contexts. Thus, the present paper proposes an audio sensor enabled real-time audio rainfall classification, based on 1-D convolutional neural network (CNN) which offers 95% classification accuracy in five rainfall intensity classes. © 2021 IEEE. |
| Author Keywords |
Audio classification; Convolutional Neural Networks; Enviroment sound classification; Feature extraction techniques; Multimodal Analysis; Rainfall Estimation |
