Plant health monitoring using multispectral imaging
At Polariks we are currently building a multispectral camera that will continuously monitor plant activity to alert farmers at the earliest stages of diseases. We wanted to tell you more about this technology and how it can convert multispectral data into meaningful insights on plant health status.
Did you know that everybody has its own high tech yet limited spectral sensor?
The human eye is composed of millions photoreceptors but we can only see three different colors: blue, green and red. What if we could see other colors? For example here is how bees see flowers that look yellow to us:
Bees are able to see light beyond the visible spectrum which helps them to detect nectar.
A multispectral camera works in a similar way: it can see visible light (blue-green-red) in combination with non-visible light (infrared). This can be used to obtain more information on plant activity.
Our stress monitoring technology is based on the observation that different objects reflect and absorb light in different ways. Stressed and healthy plants have different photosynthetic activity and interact with light in very specific ways.
By looking beyond the visible spectrum, the camera can measure light reflectance variations and identify specific fingerprints of plant stress before they are visible to the human eye.
By combining the light reflectance measurements it is possible to build NDVI images (Normalized Difference Vegetation Index).
NDVI provides clear and easy to interpret information on plant vigour. Stressed plants have a lower NDVI value and are represented in yellow/orange shades.
With NDVI, it is possible to quickly identity in-field variability at a given moment. It can also be used to compare images over time to identify significant changes in plant health status.
With these information, farmers can provide an early, local and adapted response. Which will allow them to use less resources, lower their costs, save time and eventually prevent yield losses.
Interested to find out what multispectral imagery can do for your production? Please contact us at firstname.lastname@example.org