Apple orchards are widely expanding in many countries of the world, and one of the major threats of these fruit crops is the attack of dangerous parasites such as the Codling Moth. IoT devices capable of executing machine learning applications in-situ offer nowadays the possibility of featuring immediate data analysis and anomaly detection in the orchard. In this paper, we present an embedded electronic system that automatically detects the Codling Moths from pictures taken by a camera on top of the insects-trap. Image pre-processing, cropping, and classification are done on a low-power platform that can be easily powered by a solar panel energy harvester.
Apple orchards are widely expanding in many countries of the world, and one of the major threats of these fruit crops is the attack of dangerous parasites such as the Codling Moth. IoT devices capable of executing machine learning applications in-situ offer nowadays the possibility of featuring immediate data analysis and anomaly detection in the orchard. In this paper, we present an embedded electronic system that automatically detects the Codling Moths from pictures taken by a camera on top of the insects-trap. Image pre-processing, cropping, and classification are done on a low-power platform that can be easily powered by a solar panel energy harvester.
Albanese, A., D'Acunto, D., Brunelli, D. (2020). Pest Detection for Precision Agriculture Based on IoT Machine Learning. Cham, CH : Springer [10.1007/978-3-030-37277-4_8].
Pest Detection for Precision Agriculture Based on IoT Machine Learning
Brunelli, DavideSupervision
2020
Abstract
Apple orchards are widely expanding in many countries of the world, and one of the major threats of these fruit crops is the attack of dangerous parasites such as the Codling Moth. IoT devices capable of executing machine learning applications in-situ offer nowadays the possibility of featuring immediate data analysis and anomaly detection in the orchard. In this paper, we present an embedded electronic system that automatically detects the Codling Moths from pictures taken by a camera on top of the insects-trap. Image pre-processing, cropping, and classification are done on a low-power platform that can be easily powered by a solar panel energy harvester.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
