OneSoil — a Copernicus-enabled start-up from Belarus

OneSoil
OneSoil News
Published in
7 min readOct 29, 2019

--

⚡ The original article was published on Copernicus Observer, an official e — magazine of the European Commission. Copernicus is a satellite-based Earth Observation monitoring program launched by the European Commission.

OneSoil has been dubbed “a potential disrupter” of the approach to agricultural and environmental management by Hervé Pillaud, a European influencer in the field of AgriTech and one of the promoters of La Ferme Digitale, the digital farm, a French AgriTech start-up support structure. OneSoil is a platform developed by a Belarusian start-up which uses Copernicus Sentinel data and Artificial Intelligence (AI) to support the actors of agriculture 4.0 — from farmers to businesses and governments. So, what’s the deal?

It is not too surprising that a “farming by satellite” tool such as OneSoil originated in Belarus as agriculture is an important economic sector in the country. OneSoil is a precision farming platform which monitors fields according to criteria crucial for efficient and fruitful agricultural activities. It can detect and mark field boundaries, detect crop types, determine required amounts of fertilisers, show weather forecasts, etc. As a source of satellite imagery, it relies exclusively on Copernicus Sentinel-1 and Sentinel-2 data which it processes with sophisticated AI algorithms for user-friendly results.

More specifically, OneSoil uses Copernicus Sentinel-2 multispectral images to automatically determine the crop type that grows in any given field. Then, to reduce uncertainty, it uses Sentinel-1 radar satellite images. This enables OneSoil to spot 20 types of crop with, according to its estimates, high accuracy. The OneSoil technologies can potentially determine the sowing date and the stage of plant development. However, to improve its accuracy, the model needs in situ data sets with high quality data, which are collected with help of the platform users. For instance, this feature would help farmers choose the best time for applying fertilizers and pesticides.

Satellite monitoring of the Earth. Credit: ESA

OneSoil offers all the platform functions free of charge all over the world, and it will stay free-to-use for farmers. The service currently covers Europe and the USA, but the company already envisages global coverage in 2020!

OneSoil says:

We’re not using Sentinel-1 data in the global pipeline as it represents a large amount of data and thus requires more complex pre-processing operations. However, it is very useful for removing uncertainties in some crop type identification processes.

The platform that could well disrupt agriculture and the environmental approach is Belarusian!

OneSoil services are enriched with interactive functionalities such as attaching notes with relevant information about fields: spotted some nasty parasites in your wheat field? — leave a note! Lucky to harvest particularly juicy grapes in a specific field area? — leave a note!

In addition, considering that the platform maps the fields automatically, users can contribute to the precision of the service by refining the automatically detected borders or identified crop types manually.

The platform applies Machine Learning, which constantly improves the service as more data and feedback is collected. Considering that a mind-boggling 376,835,301 hectares of fields across Europe and the USA have already been analyzed and catalogued, the system has reached a remarkable level of maturity.

OneSoil says:

To detect fields all over the world, we need to solve two main Machine Learning problems. The first one includes training a model in an efficient way. The second problem is more complex — it entails developing a cloud pipeline prediction (inference) of the model for large areas. Training a model in the cloud (x2 Graphics Processing Units (GPU) k80 with Tensorflow/Keras and multi-GPU training, for those who are in the know) takes around one day. It takes us approximately 1−2 days to predict fields in Europe and the USA for a 3 year-period.

OneSoil is focused on building a product that “brings real value to end users”. Currently, the team is planning to delineate all the fields in the world as well as to complete the development of a yield prediction function for each field type. Additionally, in the future, the team plans to launch OneSoil analytics to create and provide reports for businesses such as banks, traders, insurance companies, advisers, machinery manufacturers, food processing companies as well as seed, fertilisers and crop protection dealers. The data from OneSoil could also be of use to state and non-profit international agencies.

Users of OneSoil

OneSoil has several types of users: farm owners, farm employees, agronomists, researchers, technical specialists/machinery operators and consultants/advisers.

For instance, the platform assists farmers in complying with European and national regulations. A Belgian farmer, agronomist by education, who is currently running a family farm, has been using OneSoil for 5 months. “I was looking for a way to see the vegetation index in a clear way” — he told the team. Right now, he is mostly interested in an efficient fertiliser solution because there are area-specific regulations that apply to his farm. Thus, he has chosen the OneSoil variable rate fertiliser application to control his inputs.

OneSoil heavily relies on user feedback, which ultimately helps the team to define how the service will develop and to improve the accuracy of the AI crop type engine.

OneSoil users also use the platform to keep track of their field’s health. A farmer from Minnesota (USA) uses the OneSoil Scouting Android app to check the Normalised Difference Vegetation Index (NDVI), an information which is derived from Copermicus Sentinel-2 free and open imagery, of his fields and he uses in-app notes to mark problems such as parasites, weeds and crop diseases.

The story of creating OneSoil

OneSoil founders Usevalad Henin and Slava Mazai met in Minsk, the capital of Belarus, at the Belagro exhibition, an annual agricultural event for professionals, in 2014. Both were looking for a new way to apply their skills. Slava, a programmer, had been fond of aerial drone photography for several years. Usevalad, a GIS expert, worked in the agricultural sector. As a result of this meeting, Usevalad and Slava started working for farmers — Slava would take pictures from a drone and Usevalad would make maps for the differentiated application of fertilisers based on the analysis. The demand for the service was growing, and it became clear that processing this information required automation, and accordingly, extension of their know-how.

In 2016, Sasha Yakovlev, a designer and an old friend of Slava, who lived in Argentina at that time, came to Belarus for his summer holidays. The friends took a kayaking trip, during which Slava explained an idea they had worked on with Usevalad, and Sasha was so impressed that he began to help the guys remotely, becoming the third founder of OneSoil. Soon enough, he packed up everything and came back to Belarus together with his family. Now, Sasha is Head of Product at OneSoil.

OneSoil team in 2016

In fact, the team worked without a salary during 2015 and 2016. This thankfully changed after the next crucial meeting, which happened at another event, an AI Hackathon. There, Slava and Usevalad met Yury Melnichek, the co-founder of the investment company Bulba Ventures. The OneSoil team presented the technology that enabled weed identification using drone images. Yury saw great potential in the product and wanted to invest in the start-up, but proposed they switch to the analysis of satellite images and to build a free web-platform for precision farming.

Today, the OneSoil team consists of 15 people (it’s 28 already — OneSoil), and it is the only start-up in Belarus that is involved in both Space and AgriTech industries!

Working with Copernicus

Considering the service is free for farmers all over the world, for OneSoil it was of the utmost importance to have access to open public data to predict field borders, crops and other information. Landsat, Modis and Copernicus were identified as the three candidates of data sources for OneSoil. Copernicus won this race mainly thanks to three decisive advantages. Firstly, Copernicus offers a significantly higher spatial resolution: 10 m vs. 30 m for Landsat, thus enabling analysis at field level. Secondly, use of its data is completely free, and finally Copernicus offers more frequent data updates by having two identical satellites working on opposite sides of the orbit. As such the Sentinel-1 mission achieves its global coverage in just six days while Sentinel-2 has a revisit time of 2–5 days depending on the latitude.

OneSoil says:

To build the map, we used images acquired by the Sentinel-2 satellites of the European Union’s Copernicus programme. A total of 250 Tb of data was processed, covering Europe and the USA. We pre-processed the images by removing the clouds, shadows and snow. We then compressed the data down to 50 Tb. In a quick calculation, approximately 140Tb of raw data (we used float32 data at the inference stage) was passed through the GPU. Following this, we searched for field boundaries and classified cultures with our Machine Learning algorithms. This produced approximately 250Gb of vector maps with field geometries and cultures.

--

--

OneSoil News
OneSoil News

Published in OneSoil News

News about OneSoil team, agtech, precision farming and even more

OneSoil
OneSoil

Written by OneSoil

Beating guesswork in agriculture! Free apps for farmers around the world.