The plants, as living organisms, possess a defense mechanism that helps them react to the alterations of the environmental conditions. The intrinsic electrical signaling network is the common mean for transmitting information related to such alterations. In fact, the adaptation of a plant to a perturbation triggers changes in its underlying physiological process which are expressed as electrical signal variations. Hence, the plant electrophysiological signal could reveal the health status of a plant.

The miniaturization of electronics and integrated signal processing coupled with huge data storage and fast data transmission are enabling the creation of a wide range of innovative sensors valuable for precision agriculture. However, current agricultural devices are mostly focused on environmental monitoring without including a direct survey of a plant physiology.

The aim of this project is to design a multi-channel electrophysiological sensor that is able to acquire the electrical signal from plants growing in typical production condition, without a Faraday cage, and then, by applying advanced data analysis techniques to assess the potential of the information portrayed by this signal for identifying the status of a plant.

To assure continuous and stable recordings in different environments, the developments of this project will address the challenges of both designing a high-impendence sensor and achieving a good signal-to-noise ratio. On the other hand, as it is a relatively new field of research, to characterize the complex plant electrical response to different stimuli, a wide range of signal features will be explored, which would further lead to new insights to plant behavior, signaling and physiology.

In the framework of this project, the HEIG-VD team will employ its expertise in signal processing and intelligent data analysis to build classification models that would be able to distinguish the stressed state of the plant caused by drought, deficit of nutrients or insect infestation.

Project funded by Innosuisse - 27661.1 PFLS-LS.

Conferences

• Early detection of Tetranychus urticae in tomato soilless culture using electrophysiology and machine learning
  Poster at 4th International Symposium on Horticulture in Europe (08-11.03.2021, Virtual symposium)
• Insights of plant electrophysiology : using signal processing techniques and machine learning algorithms to associate tomatoes reaction to external stimuli
  Oral presentation at 31st Conference of the International Biometric Society of the Austro-Swiss Region (9-12.09.2019, Lausanne, Switzerland)

Publications

• Classification of plant electrophysiology signals for detection of spider mites infestation in tomatoes
  Applied Sciences 2021, vol. 11(4), no. 1414
• Electrophysiological assessment of plant status outside a Faraday cage using supervised machine learning
  Scientific Reports 2019, vol. 9, no. 17073