The issue of cloud coverage in satellite imagery is not new. Clouds plague observations, yet there are few existing solutions to overcome this huge limitation. So we created the Beyond Cloud solution to change the game.
Everyone has experienced this frustration at least once: Not being able to get crop information from satellites during a critical point within the growing season. This “blindness” can last up to months in certain parts of the world, and destroys the trust that farmers have in satellite crop health monitoring.
And this problem is global. In the image above you can see the average cloud coverage percentage over each part of the world. Most agricultural regions are covered by clouds 50% of the time, thus reducing the amount of available optical satellite images significantly. This reliability issue hinders satellite imagery and is the main reason why farm management solutions see satellite data more as a risk rather than an opportunity.
To turn this situation around, we created Beyond Cloud product. This AI-based solution provides NDVI and LAI, two of the most popular vegetation indices for crop health monitoring and scouting, unaffected by clouds. The number of these images is also higher than traditional optical imagery which means that on average, growers get access up to 5 times more (depending on the location) images per year with 100% reliability ! Let’s discover how we achieve this incredible feat.
We created the Beyond Cloud product by mixing ground field measurements, optical satellite imagery, radar satellite imagery (SAR) and machine learning techniques.
The use of SAR imagery is crucial since it is the core data on which we detect the crop health. The main benefit of SAR imagery in this case is that it is not affected by cloud coverage, which means that we always get data. We did an in-depth article about how SAR imagery can be used in agriculture, that you can read here.
Our main challenge in building this solution was to be able to correlate SAR imagery with ground measurements and optical imagery. We ensure high field level accuracy by validation against ground truth data.
As you can see above, the process had several steps, identical for each crop:
The final result is an image very similar to what you would get using optical satellite imagery, as you can see below. The image on the right is the measure we get using optical imagery, and the image on the left was achieved using Beyond Cloud.
.jpg)
All our models created have an accuracy of at least 85%, which ensures that the solution is reliable. We have tested the solution on most cereals as well as major europeans crops. We can quickly expand it to a new crop should there be some specific request.
This solution is a huge step forward for satellite imagery crop monitoring and scouting . The benefits are numerous:
If you are interested in trying out Beyond Cloud we invite you to reach out to us here. SpaceSense is a leading provider of advanced satellite data. By combining the power of satellite imagery and AI, our solutions can be quickly adapted to each local environment and are deployable via APIs on a global scale. We are now monitoring 4 million acres over 10 countries daily.
Learn more about what we offer on our solutions page.
Want to benefit from geospatial
AI easily?
Check out our upcoming solutions that you can test
Check it out