Tools & data

Background 489 views

New cloud-based crop monitoring platform from EOS

Cloud-based (monitoring) systems developed by start-ups are emerging faster and faster nowadays. Californian start-up EOS introduced its Crop Monitoring platform aimed at input suppliers, farmers, commodity traders, crop insurance companies, and all members of the Agriculture supply chain.

The company says EOS Crop Monitoring is the result of 3 years of extensive collaboration between EOS’ team of data scientists and software engineers and industry partners.

Deep learning technology

EOS Crop Monitoring allows its partners to extract insights from remote sensing data, for example crop types classification maps, crop yield forecasts, field boundaries, vegetation indices, crop conditions, soil moisture and weather data on a field, regional or country scale. To do so, EOS developed a multi-level deep learning architecture that targets land cover and classifies crop types from multi-temporal, multi-source satellite imagery.

EOS Crop Monitoring uses radar imagery and convolutional neural networks to eliminate the coverage of clouds and shadows which is a common issue in satellite imagery analytics. Photo: EOS
EOS Crop Monitoring uses radar imagery and convolutional neural networks to eliminate the coverage of clouds and shadows which is a common issue in satellite imagery analytics. Photo: EOS

The solution is empowered by the Platform-as-a-Service (PaaS) engine from EOS, called EOS Engine. It is said to be able to support many types of earth observation data sets and capable of on-the-fly analytics processing system. EOS Engine can automatically remove cloud cover and shadows as well as extract valuable information on a different scale with the processing of large area data. This allows quick identification of a field’s performance throughout the growing season as well as high-risk areas affected by droughts, floods or hail.

Some unique features

According to Olga Denisenko, content marketer at EOS, EOS Crop Monitoring stands out from similar solutions by its unique algorithms for achieving the highest accuracy level, 90% and above. “It uses radar imagery and convolutional neural networks to eliminate the coverage of clouds and shadows which is a common issue in satellite imagery analytics. It also includes the full range of crop monitoring services, for example plant health, soil moisture and weather data to create a one-stop solution.”

EOS (Earth Observing System) was founded by Max Polyakov in 2015 and is based in Menlo Park, California (USA). The company currently employs 160 people and it says the EOS Crop Monitoring solution is scalable to any region. As of today, it has delivered solutions to North America, Europe, South America, and currently expanding towards Asia and Africa. The pricing depends on the project scope.

Or register to be able to comment.