Plataforma embarcada para monitoramento fenológico da cultura do morangueiro

Abstract

Agriculture is an area that allows for different technological innovations, like the use of computer visionin an embedded solution. In order to contribute with phytosanitary management techniques, and toassist researchers with data from sensors, this work presents the development of an embedded visionsystem for the strawberry crop. Computer vision allows researchers to implement high precision phenotyping processes. With this in mind, it is possible to use image manipulation techniques to determine a leaf area of the crop, as well as collect data from meteorological sensors. In order to create the solution, there was an integration between embedded platform, Raspberry PI 3, sensing peripherals anda software to operate the system from a graphical user interface. To validate the equipment, a greenhouse was used in the horticulture sector at the University of Passo Fundo. Results suggested ourcost-effective system that could be used in practice byresearchers and producers, allowing an effectivemonitoring of the crop. Data collections were performed during the 21 days, and the data obtained were statistically analyzed. A comparison was executed between the manual method of estimating leaf area of Albion culture, through prediction equations, and the proposed method of image processing,showing that data measured by the platform does not exceed 10 % variation. Pearson’s correlation coefficient showed a strength significance (0,96) between leaf area and accumulated temperature duringthe period.