• AI-Powered Plant Biosensors
  • 2 million+ Plant-days of field data collected
  • Trusted by growers and industry leaders worldwide
  • Identify crop stress before visible symptoms appear
  • 118.8 B Corp Impact Score
  • Plant-driven cultivation powered by AI
  • Higher yields, lower inputs
CONTACT

Online webinar

Innovating the berry sector with plant biosensors and direct plant feedback

Receive the full webinar recording below!

Did you know you can understand how your berry crops are truly performing, directly from the plant itself? In our most recent webinar, experts from Vivent Biosignals, Kwantlen Polytechnic University, and Andermatt Poland showed how plant biosensors and real-time plant feedback are transforming berry cultivation.

We covered innovations in berry production in Poland, and the latest research on early warning signals for nutrient issues, varietal differences, and supplemental lighting. We also showed how real-time plant data can reveal crop stress, growth dynamics, and overall plant health

The topics

  • Introduction to plant driven data – Andrzej Kurenda (CSO) and Erik Veenman (Sales Manager) – Vivent Biosignals
  • Research on early warning and nutrients issues, varietal differences and supplemental lighting – Deborah Henderson (Director Institute for Sustainable Horticulture) – Kwantlen Polytechnic University
  • Innovating berry growing in Poland – Krystian Rębisz (General Manager) and Bartlomiej Drzazga (Horticultural Crop Manager) – Andermatt Poland
  • Use cases & Q&A

Sign up and watch the full webinar session here!

Webinar FAQ

1. How do you handle data ownership? The sensors gives you data and we have access to online dashboards, but do we have access to the data other then that?
  • Indeed you get access to your data which you can see in your own dashboard.
2. Does each sensor provide data on a variety of factors or can each sensor only provide data on 1 factor?
  • If by "factor" you mean 'stressors', we can detect different types of stressors from the same electrical signal that is recorded by our device on the plants. Imagine our device is a microphone in a stadium full of people. Each stressor will be a different person talking. Our algorithms learn how to listen to specific people to obtain specific information!
3. How do you determine which values classify you Red, yellow, and green zones? How does this get adjusted for different crops?
  • Good to understand first: what we show is how certain our models are that the plant is demanding water/nutrients, not the actual water levels in the plants or substrates. If the model is very certain that the plants are asking for water, the certainty will go up. Based on this principle we've assigned the colours red/orange/green. The models work the same for all plants, and therefore the colours are the same for all crops.
4. For strawberries, should the electrodes be installed to monitor only leaves or is there different signals your model could decode from an electrode placed on a truss ?
  • We install the sensors always in the stem of the leaves of the strawberries to monitor the general overall health of the plant. It is possible to also insert the electrodes into the stem of the flowers and fruits, this will give you more specific data on their development.
5. Also for strawberries, when a leaf gets older is there a point where you should move the electrode to a younger leaf and what impact on the model results does it have if you do, does it confuse it ?
  • Yes indeed, when the leaves get too old and start wilting, you move the electrode. To make sure you have to do this as little as possible, you install in the youngest leaf of the plant when you first install.
6. Does the model require a complete climate dataset to decode the information from the electrodes or is it able to work with only the electrode signals ?
  • It works with only the electrical signals. However, plants are never separate from their environment, knowing about the climate around your plants is therefore very helpful for interpreting the signals and getting a complete picture.
7. How do you keep track of your model’s performance over time?
  • We don't train models 'on the go' so we don't directly track performance over time. However, we regularly retrain our models when we acquire new datasets to increase their generalisability. Does this answer your question?
8. How often each sensor sends electrode data to the gateway.? I assume also the your sampling rate is low...
  • The device records a sample every second yes, but the actual transfer of data from the device to the gateway is spaced out a bit too not overload the system - so the device sends data to the gateway every minute. Hope that makes sense and answers your question!
9. If water uptake is too low, would the reason be indicated eg: low soil water/salinity/water logging etc?
  • Vivent sensors can detect the occurrence of stress responses in plant tissues; however, the underlying causes of stress may vary and are therefore not directly specified by the system. For example, low plant water status may be associated with low soil moisture, but it can also result from root degradation, high medium EC, or high air humidity. Therefore, when the system indicates low water status, we recommend inspecting the plants and growing conditions directly to identify the precise source of stress.

Erik Veenman
Vivent Biosignals

Andrzej Kurenda
Vivent Biosignals

Deborah Henderson
KPU

Krystian Rębisz
Andermatt Poland

Bartłomiej Drzazga
Andermatt Poland