Contribution of hyperspectral imaging to monitor water content in soilless growing cucumber crop

Drought stress in soilless cultivation plants causes various symptoms.
Until now, in the majority of the greenhouses irrigation control is based either on water content measurements in root zone or on air temperature and relative humidity records both performed at a single point located at the center of the cultivated area.
However, this methodology is not workable in the recent infrastructures, since their size has greatly increased and large water consumption gradient has been performed.
Thus, direct and real-time monitoring systems of plant response in different location within the greenhouse are required.
Hyperspectral machine vision is a non-contact and non-destructive sensing technology that pave the way for the commercialisation of robotic machine vision.
The objective of this work was to map through hyperspectral camera the water content gradient observed in cucumbers cultivated in a greenhouse.
Plants of different irrigation regimes were imaged in different indoor positions.
The gradient of plant physiology response was also studied.
During the measurements, the impact of shadows to the targeted object was eliminated by placing a black surface as background.
The results received within the framework of the current analysis perform a sound way for proceeding in more sustainable irrigation control system.

International Symposium on Advanced Technologies and Management for Innovative Greenhouses: GreenSys2019

A. Elvanidi, J. Zinkernagel, J.F.J. Max, N. Katsoulas

hyperspectral machine vision, remote sensing, greenhouse, substrate water content, irrigation

https://doi.org/10.17660/ActaHortic.2020.1296.133