We would like to keep you informed about our progress. For this reason, we invite you to subscribe to our newsletter. We will not burden you with unnecessary information. Only the interesting stuff. The previous newsletters can be read below the subscription form.

Subscribe

* indicates required









previous newsletters:

 Kappazeta NewsLetter, May 2024
 Kappazeta NewsLetter, January 2024
 Kappazeta NewsLetter, December 2023
 Kappazeta NewsLetter, November 2023
 Kappazeta NewsLetter, September 2023
 Kappazeta NewsLetter, August 2023
 Kappazeta NewsLetter, May 2023
 Kappazeta NewsLetter, April 2023
 Kappazeta NewsLetter, March 2023
 Kappazeta NewsLetter, February 2023
 Kappazeta NewsLetter, January 2023
 Kappazeta NewsLetter, December 2022
 Kappazeta NewsLetter, November 2022
 Kappazeta NewsLetter, October 2022
 Kappazeta NewsLetter, September 2022
 Kappazeta NewsLetter, August 2022
 Kappazeta NewsLetter, July 2022
 Kappazeta NewsLetter, June 2022
 Kappazeta NewsLetter, May 2022
 Kappazeta NewsLetter, April 2022


See older newsletters here.
 
 

Newsletter, May 2024

1. KappaZeta grows agricultural expertise
We’re excited to announce that agronomist Teet Liiv has joined KappaZeta as Agriculture Domain Lead. Over his career, Teet has gained extensive insight into the agriculture sector as an adviser, product specialist and sales consultant. He is most interested in figuring out how financially sustainable agriculture and ecological agriculture fit together, while also preserving a diverse natural environment.

Agriculture has been at the heart of KappaZeta since the establishment of the company in 2015, inspired by the need to monitor mowing activities for agricultural subsidy checks. Since then, KappaZeta has provided agricultural monitoring services for agricultural agencies and has completed several R&D projects for detecting agricultural practices and events from satellite images.

With advice and guidance from the new domain lead, we have a comprehensive understanding of agricultural processes, the complex relationship between all players in the field of plant production, and improved knowledge about user needs. We’re looking forward to steering our agricultural services in the right direction with Teet’s leadership!

Portrait of Teet


2. Detecting clear-cutting in Estonian forests
KappaZeta completed a pilot project for the Estonian Environment Agency to monitor forest clear-cutting. As a result of a fast-paced pilot project commissioned by the agency, our machine learning model achieved 90% accuracy in identifying clear-cutting events on the plots that are at least 0.5 hectares large. The Environment Agency was pleased with the results and plans to prototype a service for forest clear-cut detection in near future.

Just as the Agricultural Registers and Information Board has been pioneering technological innovations for agriculture, the Estonian Environmental Agency has taken a significant step forward for forestry in Estonia. We are thrilled to contribute to the agency's effort in enhancing the reliability, quality, accessibility, and transparency of forest data.

Read a more detailed press release by the Estonian Environment Agency here (in Estonian).

Presentation event

Presentation of the results from the clear-cut pilot project to the Estonian Environment Agency and other stakeholders. Source: Estonian Environment Agency.


Newsletter, January 2024

1. Five satellite analytics tools for agriculture
During the past two years, the KappaZeta team has focused on developing services for agricultural management and analysis. Prototypes for the services were developed within the project “Satellite monitoring-based services for the insurance sector – CropCop”, supported by the European Regional Development Fund and Enterprise Estonia.

The suite of tools for agriculture includes:
  • Crop Type Detection, enabling precise identification of various crop types,
  • Parcel Delineation for accurate land mapping,
  • Seedling Emergence Detection to monitor early crop growth,
  • Farmland Damaged Area Delineation for assessing areas affected by adverse events,
  • Ploughing and Harvesting Events Detection to track critical farming activities.
Learn more from our latest blog post.

Crop type detection

Image: Results of crop type detection.

2. New pilot for the Environment Agency
KappaZeta has launched a pilot project for the the Estonian Environment Agency to monitor forest clear-cutting. Automatic and timely monitoring of felling activities would free land managers from the obligation of reporting and significantly reduce the workload for inspectors.

The detection of clear-cutting will be based on radar Sentinel-1 imagery, which sees through clouds, and on optical Sentinel-2 imagery, whose spatial resolution exceeds the radar’s. The image below displays an example of what clear-cutting looks like in optical vs. radar imagery.

In two temporally consecutive optical images captured by Sentinel-2 (top row), it is visible that as of August 5, a fresh clear-cut existed in the right side of the scene. By September 11, a second clear-cut had appeared nearby. Backscatter (s0) and coherence (coh) radar images are averaged over a 2-week period to reduce noise (bottom row). In addition to that, an existing smaller clear-cut can be identified in both Sentinel-1 products, which confirms that the second clear-cut visible in the optical image from September 11 had not yet appeared by August 14.

Clear-cut

Image: Detection of forest clear-cutting using optical and SAR images.

3. KappaZeta at the Winter Satellite Workshop
KappaZeta’s delegation kicked off the new conference season at the Winter Satellite Workshop in Espoo, Finland, where we had three busy days of presentations and discussions with good people.

Tauri Tampuu demonstrated the benefits of analysis-ready SAR data for various applications, highlighting key events of the Russo-Ukraine war as seen in satellite images.

Martin Jüssi presented the concept and status of 3D-SAR, our planned bistatic SAR mission for monitoring vegetation volume from space.

Looking forward to meeting you at the next space, agtech and defence events soon!

Tauri Tampuu presenting

Image: Tauri Tampuu presenting at the Winter Satellite Workshop.

Martin Jssi presenting

Image: Martin Jüssi presenting at the Winter Satellite Workshop.


Newsletter, December 2023

Thank you for following KappaZeta's story! We really appreciate it. 2023 is about to end and it's time to sum up the year - please see our timeline below.
 
KZ timeline


🎄 🎄 🎄  KappaZeta team wishes you a beautiful holiday season! And may your dreams come true!


Newsletter, November 2023

1. Martin Jüssi joins KappaZeta
Martin Jüssi, previously the Space and Earth observation director of CGI Estonia, has joined KappaZeta to lead the 3D-SAR mission development. Martin has a wide network in the space domain and a multidisciplinary background with good experience in GIS, space and technology management. Martin was also part of the Tartu Observatory Remote Sensing research group in 2012, which eventually led to the formation of KappaZeta. “In a way, it feels like rejoining the team after a number of years rather than being a new joiner”, comments Martin “I am glad to be leading this ambitious endeavor, with the best possible team I can imagine, in our lovely hometown of Tartu.”

Martin and Kaupo


2. 3D-SAR mission development in full force
With Martin at the wheel, we kicked off the concept exploration phase (Phase-0) of the 3D-SAR mission. The high-level objective of the mission is to map global forest biomass with unprecedented accuracy and revisit frequency using bi-static SAR data, therefore enabling direct measurements of carbon stock, biomass change and density from space. In the concept exploration phase, we will define high-level mission requirements and explore various concepts for fulfilling these in the best possible way, ensuring a balance of cost and data quality based on the user requirements.

3. From estimation to precision: Your path to reliable carbon assessment!
As we believe the 3D-SAR data products can have a massive positive impact on carbon stock awareness and data-based climate decisions, we will participate in COP28 in Dubai from December 2 to 10 to increase awareness and gather support for our mission.

COP28


4. Winter Satellite Workshop in Espoo, Finland
KappaZeta will take part in Winter Satellite Workshop, 17-19 January 2024 in Espoo, Finland (https://spaceworkshop.fi/). Join us to discuss 3D-SAR constellation and the future use cases of its data, Sentinel-1 analysis ready data on KappaOne platform (https://kappaone.eu/) and other interesting topics with curious, knowledgeable and interesting people.

Winter Satellite Workshop 2024




Newsletter, September 2023

1. KappaZeta is hiring!
After the recent successful investment round (you can catch up with the details here), KappaZeta is gearing up for the ambitious endeavors of the near future. To make our plans a reality, we’re on the lookout for talented individuals to join our team.

Are you or someone you know a perfect fit for one of these roles at KappaZeta?

🌱 Agriculture Domain Lead: If you're passionate about agriculture and ready to lead the way in this domain, we want to hear from you!

🧠 Senior Machine Learning Engineer: Are you a seasoned machine learning expert looking for your next challenge? Join us and help shape the future with cutting-edge technology.

📊 Data Quality Specialist: Precision and attention to detail are your superpowers. Join our team and ensure the highest data quality standards.

We're inviting you to be a part of our journey as we drive innovation and make a real impact. If you or someone you know is ready to embark on this exciting adventure with us, please don't hesitate to get in touch.

2. AI-based tillage detection for smarter agricultural and climate policies with the European Space Agency
Among the myriad of causes of climate change, have you ever stopped to wonder how tilling your land may be contributing to this issue? Conventional tillage methods and their long-standing application comprise techniques that have a negative environmental impact on the soil, such as erosion and carbon loss.  

With the European Commission's Common Agricultural Policy (CAP), the EU encourages member states to apply sustainable agricultural practices, e.g., conservative tillage methods. Conservative tillage techniques reduce erosion and carbon loss from the soil as they involve minimal soil disturbance or retention of crop residues on the soil surface. The main aim of the policies in the agriculture sector is to ensure that we produce our food sustainably.

Tillage

Agricultural paying agencies in EU member states are mandated to monitor farmers' agricultural practices and ensure they comply with the policies put in place. KappaZeta has embarked on a new project funded by the European Space Agency (ESA) to support the vital work undertaken by the paying agencies. The development of an AI-based tillage detection model is a project that leverages Earth Observation as an integral aspect of the area monitoring system (AMS) and is set to run from September 2023 to September 2024. The AI-based tillage detection model aims to promote sustainable agriculture according to the European Green Deal that rewards farmers for low-impact practices that contribute to carbon sequestration. Encouraging sustainable agriculture and leveraging the appropriate technologies will lead to a smarter and more resilient future.

3. Taking a deeper dive into the world of SNDVI
The full version of summer intern Karl Hendrik Tamkivi’s adventures in the realm of SNDVI is now available on the KappaZeta Medium blog.

sNDVI comparison



Newsletter, August 2023

1. First investment round successful
KappaZeta received €480 000 investment from business angels to develop our existing services and kick-off the development of an exciting new satellite mission to create an accurate, global, and frequently updated data set about forest biomass and carbon stock.

Business angels Ahti Heinla, Prof. Adam Beaumont, Markus Tamme, Mart Noorma and Anu Noorma, Priit Salumaa, Priit Vellak, Toomas Kams and Eva Kams, Tõnu Runnel and Estonian Business School Group invested in our growth. With the investment we will develop further our agricultural and defence products and create a service for forest carbon stock assessment. We will kick-off the development of an exciting new satellite mission making Sentinel-1 data three-dimensional. The mission will create an accurate, global, and frequently updated data set about forest biomass and carbon stock.

Kaupo Voormansik, CEO
 
3D-SAR concept

Artist’s impression of the planned satellite mission (© Taavi Torim and European Space Agency).

 
2. Adventures in the realm of Synthetic NDVI
This summer, we had the pleasure of hosting a machine learning intern Karl Hendrik Tamkivi who explored ways to improve Synthetic NDVI measurement. Over two months, he carried out experiments with the aim of achieving a lower error rate, better visuals, and higher resiliency. Karl dissected the problem from three perspectives: model architecture, data manipulation, and performance evaluation.

As a result of several experiments, Karl managed to improve SNDVI model accuracy and efficiency, and identified several areas which deserve further attention (and what is equally important – which approaches do not).

SNDVI is a promising solution which can identify crop failure early, assist in tracking different growth stages, and provide automated ways to monitor agricultural activities.

Thank you, Karl Hendrik, for being a great addition to our team this summer!

sNDVI comparison

Example of experiment results. From left to right:
Original Image: The initial, untouched image.
Initial Model Output: Notice regular pattern distortions and lack of spatial detail.
Enhanced Model Output: Result from an improved model with adjusted kernel size and updated interpolation techniques. Notice the clearer and more spatially accurate representation.

3. Meet us!
Estonian companies at DALO

Estonian companies at DALO Industry Days in Denmark. KappaZeta (Tanel Kobrusepp on the far right) introduced situational awareness solutions for the defence sector.

Next events:
11.-15.09 FRINGE 2023. Posters: Sentinel-1 3D: Constellation of Bistatic Passive Receiver Satellites Formation Flying with Sentinel-1 for Operational Applications, Analysis Ready SAR Backscatter and Interferometric Coherence Data for Professional and Non-Professional Users (Leeds, UK)
12.-14.09 6th Sentinel-2 Validation Team Meeting. Poster KappaMask v2: Going Global.
(ESA ESRIN, Frascati, Italy).
12.-15.09 DSEI 2023 (London)

Newsletter, May 2023

1. KappaMask over Europe

KappaMask banner

Since the beginning of the year, KappaZeta has been using EOSC Future cloud computing resources with the aim to provide KappaMask-based cloud and cloud segmentation masks for every Sentinel-2 product over Europe.

By generating KappaMask-based cloud and cloud shadow segmentation masks for every Sentinel-2 product available in the European region and hosting the masks on CreoDIAS with a free licence, we provide ample opportunities for testing KappaMask performance for all interested parties.

Learn more from Tetiana’s presentation at the “Use cases from the EOSC community” webinar:  https://youtu.be/Yuvd3y0JEXk?t=3239

2. German - Estonian Defence Industry Seminar in Tallinn
KappaZeta took part in the German – Estonian Defence Industry Seminar in Tallinn held by the Estonian Defence and Aerospace Industry Association. We were represented by Tanel Kobrusepp and Jürgen Lina who had a great opportunity to connect with the German defence industry leaders and the public sector.

KappaZeta is building near real-time Earth Observation solutions that are crucial for security and defence capabilities.

We are looking forward to the upcoming events to set our foot even further in the defence industry.

In the seminar

 

3. A look at the Estonian forest together with Levila
Levila journalists are tackling one of Estonia’s biggest mysteries – the lack of reliable information on the condition of Estonian forests.

With so much monitoring equipment (as seen in the illustration below), shouldn’t we be able to know just about everything about the forests? Levila contacted KappaZeta to ask how satellites can help get an overview of Estonian forests.

The first Levila blog posts are available in Estonian here: https://www.levila.ee/tekstid/kuhu-kaob-mets

Monitoring equipment


Newsletter, April 2023


1. SERA presentation day in Tallinn
On Thursday, 20 April 2023, eleven Estonian companies from the defence and aerospace sector, KappaZeta Ltd among them, could promote their unique innovative products and services to 80 top European defence and security executives. There was a presentation day (https://www.facebook.com/ESTDefenceIndustry/) where the Estonian companies met the attendees of an international course SERA (Session Européennee des Responsables d’Armement) for the high rank European defence and security personnel held in Estonia and organised jointly by the Estonian Defence and Aerospace Industry Association and the Estonian Ministry of Defence. We joined the Estonian Defence and Aerospace Industry Association recently. The heightened international security situation has made us want to use our expertise on radar and AI for the benefit of the security of Estonia and the defence capabilities of ours and our allies. While KappaZeta does not currently offer any significant ready-made services for the defence sector, we have the capacity to develop such solutions, and our ideas were well received at the event. Now, we are working to take these ideas forward.

KappaZeta demo area
Image: Tauri Tampuu presenting KappaZeta at the SERA presentation day.

 
Tauri Tampuu, research and development manager

2. Towards Sentinel-1 3D: interesting stories and criticism
KappaZeta is planning a new satellite mission with a goal to make Sentinel-1 data 3-dimensional and to provide an excellent data source to measure forest biomass at global level with unprecedented accuracy and frequency. The following paragraphs are an exerpt from our CEO Kaupo Voormansik’s  blog post ”A story of starting a satellite mission”.

The mission concept
Image: The mission concept  constellation of passive receiver satellites formation flying with Sentinel-1.


One of the most interesting critical remarks came from a good colleague: „Your undertaking has too much of a community feel to it.“ I would say that’s completely intentional. KappaZeta is a do-good company with a mission and that’s no secret. We’re convinced that an open approach will be successful, as opposed to being afraid, trying to hide, building walls wherever possible, keeping our work and plans a secret. The Copernicus programme, which the new mission would complement, is a success largely thanks to its open philosophy.

All interested parties need to know that the door for discussing cooperation is open, be that for benefitting from the planned data or for collaborating on satellite development. Experience from the first Estonian satellite project ESTCube-1 taught me many lessons that led to this conclusion. I’m glad that I never started criticising the new „aliens“ who had just joined the team and didn’t seem very reliable at first sight. Many of them later surprised the entire team, proving to be thorough and careful engineers that could always be relied upon. The second lesson was that each person plays their own role and not every person has to be very profoundly involved in development. For example, one ESTCube-1 member never stood out with great engineering results but brought a friend to the team who became one of the core developers. All of this demonstrates the power of community, which is enormous. Everyone needs to feel welcome to join and contribute. This is especially important in the beginning, when the entire competence to complete a mission doesn’t already exist within the team.

More criticism on the same note: „It doesn’t matter that you have a lot of friends!“ It matters tremendously! One person alone, however smart, and capable, could not get a lot done. Looking at the lively and supportive cooperation between Estonian startups, it seems that most people have understood this already long ago. From our experiences, we can thank the SuperAngel accelerator, ESA BIC Estonia, the Estonian Founders Society, and the Estonian Space Office.

To sum all of this up — there’s still a lot of work ahead, but there aren’t any fundamental obstacles preventing the mission from becoming a reality. The only thing that might make us reconsider our plans, would be if a better alternative technology were to appear. As long as our planned mission (bistatic SAR interferometry using accompanying satellites) provides the best combination of high accuracy, global coverage, and a reasonable price for measuring the global forest biomass, we will be working onwards.

Today, we have verbal agreements to raise 2,35 million euros in investments. Hopefully, we can publish official news about it soon. I extend my sincere gratitude to all our team members, advisors, the future investors, and the constructive critics.
 
Kaupo Voormansik, CEO


Newsletter, March 2023

1. A story of starting a satellite mission by Kaupo Voormansik
It all started in November 2021 when TalTech space programme manager Rauno Gordon invited all the interested parties to discuss the idea of a new „Estonian satellite“. We and the other enthusiasts met on a several occasions to discuss ideas – the common denominator being the desire to develop a mission useful for Estonia. Soon enough I was able to convince everyone that building a new satellite to observe just Estonia is not the way to go. Instead, we should think about how to use satellite remote sensing to benefit the world. Most of the other participants at the meeting seemed to agree.

At KappaZeta, we had been thinking of the idea of enriching the Sentinel-1 dataset for a long time. At first, we hoped ESA would do it by developing accompanying missions to Sentinel-1 that would add the height dimension and make the next generation fully polarimetric. Discussions on this topic showed that ESA was not planning to do it themselves but would be willing to support third parties developing the mission. OK, great – I pitched this idea at the next „Estonian satellite“ meeting and it was received quite well. Adding a small and relatively inexpensive accompanying satellite to enhance the data of a large and important mission seemed both feasible and meaningful.

After one or two more meetings, I realized that in order to make it actually happen, I have to take the lead. I needed to stand up and say that we (i.e., KappaZeta) will start organising it, and others would then join us. We had what was necessary to get the snowball rolling – systematic knowledge about SAR, personal experience from developing ESTCube-1, an understanding of the Earth observation market and about the importance of the dataset to be developed. As the last and very important component, we also had a decent network within the Estonian and other European countries’ space technology and remote sensing communities. Today’s KappaZeta could not do it alone, but with help from friends – absolutely!

At the beginning of the story, I referred to the „Estonian satellite“ using quotation marks, but despite all our national pride, we shouldn’t emphasize its nationality. The more countries that consider the Sentinel-1 accompanying satellite constellation to be their own, the better. Satellite remote sensing is inherently global and international cooperation its key to success, helping to avoid unnecessary fragmentation and duplication, and helping to progress further together. A satellite that is suitable for gathering data about Estonia is just as capable of doing the same for all other countries of the world.
  • If you’re interested in the technical content of the planned mission, please see the abstract from the Finnish Satellite Workshop Remote Sensing seminar: https://drive.google.com/file/d/1rhppDdSWLjU7CXeanmLWo1Vl4xLE3h8_/view
  • The story will be continued on our Medium blog next week, where Kaupo will share his experience receiving feedback to the mission concept and will explain how far along the mission preparations are.
 
Kaupo Voormansik, CEO

2. Collaboration with 1MTN
We are delighted to work together with 1MTN (https://www.1mtn.com/) on nature-based carbon credit projects. KappaZeta’s role is to assist 1MTN with satellite remote sensing expertise and solutions. Currently we are collaborating on a bamboo planting project in Africa. It has provided us with valuable insight into the remote sensing and GIS needs of the voluntary carbon market. We see that correct adoption of remote sensing solutions facilitates transparency of the voluntary carbon market and reduces costs of the carbon project developers.

We wish Anete Garoza and Kristjan Raude a lot of energy for the development of carbon projects!

Logos of 1MTN and KappaZeta

 
Jürgen Lina, business development manager


Newsletter, February 2023

A new satellite mission: Sentinel-1 data to 3D
We are excited to share that KappaZeta is planning a new satellite mission with a goal to make Sentinel-1 data 3-dimensional and to provide an excellent data source to measure forest biomass at global level with unprecedented accuracy and frequency.

The quiet homework in cooperation with Estonian and Finnish Universities and with the local space community has been ongoing for about a year now. The first public presentation about our mission concept was given at the Finnish Satellite Workshop (https://spaceworkshop.fi/) in January 2023. Currently we are preparing the user requirements mapping campaign and partner search for the necessary space hardware and system integration. We are also in the process of acquiring the first major investment. More news will be announced in the coming months.

Why is it important? Why does it matter?

The mission applies a technology called bi-static across-track synthetic aperture radar interferometry which (especially when implemented with additional passive receivers) is one of the best trade-offs for global forest biomass and carbon stock mapping in terms of accuracy versus frequent global coverage and cost. It is vital for global carbon trading to add transparency and avoid greenwashing.  We could significantly increase the accuracy of global forest carbon stock estimation while keeping the costs reasonable.  Today’s 30-40% error could be reduced to about 10% error.

Making Sentinel-1 data 3-dimensional would also improve almost any existing agricultural and forestry application that currently relies on Sentinel-1 data. Having more linearly independent input parameters for AI-models results in higher accuracy and ultimately happier customers.

The image below illustrates the planned data for accurate forest biomass estimation at parcel level.

Forest volume

 
Kaupo Voormansik, CEO


Newsletter, January 2023


CropCop services for crop insurance
The process of on-site loss adjustment for crop insurance companies is often lengthy and inefficient. Using specialized inspectors or local drone companies can have geographical limitations and obtaining uniform historical data can be difficult. In a perfect world, field information such as vegetation health and farming activities could be easily accessed from anywhere in the world, quickly and either on a computer or a portable device. This is the solution that CropCop offers. Our goal is to streamline loss adjustment as much as possible.

CropCop on a tablet


As more and more crop insurance companies are aiming to make satellite data work for them, we decided to be a part of this digitalization movement. The CropCop team is currently partnering up with the first crop insurance companies to be the leading data provider for the upcoming 2023 season. If you're interested in a chance of becoming a part of the CropCop Customer Advisory Board (CAB) program, please contact Tanel Kobrusepp HERE to have a say in future CropCop developments and gain a competitive advantage over other insurance companies.

Some of the AI models that are in development:
  • Crop Type Detection
  • Agricultural parcel delineation
  • Seedling Emergence detection
  • Detection of Cultivation Events
  • Delineation of Damaged Areas
  • Yield Estimation
  • Risk Assessment
If you want to learn more, then visit our freshly completed website: https://www.cropcop.io/
 
Tanel Kobrusepp, CropCop product manager


Newsletter, December 2022

2022 highlights
This year for KappaZeta has been a time of growth and many exciting new developments. Here are some of the main highlights we’d like to share with you.

KappaMask logo
KappaMask went global! The most accurate free and open Sentinel-2 cloud mask KappaMask became available globally in all seasons. We went through rainforests and deserts, mountains and oceans, cities, and farmlands to create KappaSet, the largest cloud and cloud shadow masks dataset. Not only are different surface types included in it but also different cloud types and seasons from all over the globe. Another improvement is the faster processing time
.

Start using KappaMask: https://github.com/kappazeta/km_predict
Download KappaSet: https://zenodo.org/record/7100327


KappaOne logo
KappaOne launch. The KappaOne service is designed to make Sentinel-1 SAR data easily accessible and ready to be used and analysed. There are six Sentinel-1 ARD products available, which are presented in human-friendly as well as machine-readable form. Our raster processing got a major upgrade that significantly improves the quality of our backscatter and coherence rasters. This means end-users get calibrated and noise corrected imagery products with the highest possible spatial resolution using advanced speckle suppression methods. Significant focus has been on operationalizing the sNDVI model as a layer of KappaOne.

While KappaOne can be used for a variety of applications, it has already proven extremely helpful for subsidy checks under the Common Agricultural Policy (CAP). Sentinel-1 images are well suited for assessment of mowing, harvesting, ploughing and other markers.

Explore the KappaOne demo: https://demodev2.kappazeta.ee/ard_demo/  
Learn more: https://kappazeta.ee/blog/why-do-we-need-sentinel-1-data-service
Watch CAP-related webinar: https://youtu.be/TyH1B25g2gI


sNDVI images

We also have great memories from the productive summer months thanks to our six hard-working interns. At one point in summer, the KappaZeta employee count reached a total of 21 people! At the end of the year, there are 16 team members.

While we have strongly reinforced our non-technical team this year, science remains at our core. We’re now very proud to have six PhD holders on the team. Our software developer Heido Trofimov defended his PhD thesis “Polluted clouds at air pollution hot spots help to better understand anthropogenic impacts on Earth’s climate” and Tauri Tampuu defended his PhD thesis “Synthetic Aperture Radar Interferometry as a tool for monitoring the dynamics of peatland”.

Having more people on board has also given us the opportunity to generate more ideas, write more proposals, focus more on sales and participate in more tenders. We’re excited to start and continue new collaborations and keep working toward our goal – making satellite data valuable for everyone. Thank you to everyone who is supporting us on this journey.

The figure above illustrates a ploughing example on the Synthetic NDVI layer, which is computed from Sentinel-1 raster with usage of historical Sentinel-2 data. It is visible that the vegetation index before ploughing on the parcels is significantly higher than after ploughing, as the color changes rapidly from yellow to dark orange.

New crop insurance tools
A new project to develop tools for crop insurance began with support from the program for applied research funded by Enterprise Estonia and the European Regional Development Fund
.
Contrary to the United States, crop insurance is not currently common in most European countries. The main reasons producers do not insure their fields are price, lack of flexibility for custom terms from the insurance companies and lack of insurance products to choose from.

Tools based on satellite data provide opportunities for faster underwriting, better risk assessment, and reducing costs on time-consuming and expensive claims handling. Incorporating our tools into insurance companies’ work will help increase the amount of available tailor-made insurance policies in a cost-effective manner. We hope these new insurance policies will be attractive for farmers and activate the crop insurance market in Europe.

We’re always looking for accurate field data (i.e., from smart harvesters) to improve the accuracy of how we interpret satellite data. Please contact mariana.rohtsalu@kappazeta.ee if you would like to become a contributor. To find out more about our crop insurance tools, contact tanel.kobrusepp@kappazeta.ee.

Growing the team
This year KappaZeta welcomed many new and talented members:

  • KappaOne product owner Andres Luhamaa;
  • Chief operating officer Karoli Kahn;
  • Product manager of crop insurance oriented services Tanel Kobrusepp;
  • Spatial data manager Mariana Rohtsalu;
  • Research & development manager Tauri Tampuu (PhD).
We also have great memories from the productive summer months thanks to our six hard-working interns. At one point in summer, the KappaZeta employee count reached a total of 21 people! At the end of the year, there are 16 team members.

While we have strongly reinforced our non-technical team this year, science remains at our core. We’re now very proud to have six PhD holders on the team. Our software developer Heido Trofimov defended his PhD thesis “Polluted clouds at air pollution hot spots help to better understand anthropogenic impacts on Earth’s climate” and Tauri Tampuu defended his PhD thesis “Synthetic Aperture Radar Interferometry as a tool for monitoring the dynamics of peatland”.

Having more people on board has also given us the opportunity to generate more ideas, write more proposals, focus more on sales and participate in more tenders. We’re excited to start and continue new collaborations and keep working toward our goal – making satellite data valuable for everyone. Thank you to everyone who is supporting us on this journey.

KappaZeta team
Wishing a successful and happy new year from KappaZeta!
 
Karoli Kahn, chief operating officer


Newsletter, November 2022

1. To our dear readers
It is more than two years since our first monthly newsletter. Thank you for showing interest in our work. This means a lot to us. In order to serve our readers more efficiently we have a request for you. Please write to jurgen.lina@kappazeta.ee and let us know what would make our newsletter more useful and interesting for you. Your feedback is very much appreciated.

Jürgen Lina, business development manager
 
2. Global KappaMask
Anyone who has ever worked with optical satellite imagery knows that clouds can be quite unpleasant, significantly decreasing the usability of the data. Previously, KappaZeta addressed the problem by developing KappaMask, an AI-based cloud and cloud shadow masking processor, for terrestrial North European summer conditions. Its accurate performance motivated us to extend it to global conditions.
 
A year later, we are happy to announce that KappaMask v2 is finally out! Now, KappaMask v2 can detect clouds all over the globe in any season. It was trained on KappaSet, a large and diverse cloud and cloud shadow dataset developed with the help of active learning. We believe KappaSet will benefit the research community and boost cloud masking development. KappaSet is available here: https://zenodo.org/record/7100327 
 
KappaMask v2 is free and can be found here: https://github.com/kappazeta/km_predict
Feel free to try it yourself and let us know about your experience!
KappaMask v2

Tetiana Shtym, machine learning engineer



Newsletter, October 2022

1. KappaZeta at the Agroinsurance International Conference 2022
KappaZeta had a great opportunity to visit the AgroInsConf 2022 in beautiful Tbilisi in Georgia, which was designed to bring together agricultural insurance companies, reinsurance companies, brokers and insurtech companies. A lot of topics were covered during the conference from the overview of agricultural insurance markets to problems that insurance and reinsurance companies are facing and how technological applications of satellite imagery can help in solving them. The conference ended with a field day, where we had a chance to test out existing technological solutions for loss adjustment.

It was a pleasure to see that insurance companies understand the value that EO solutions can provide them and it’s a matter of time when satellite data will be used by most of the insurance companies in the world. KappaZeta’s expertise is in using SAR (radar) satellite data, incorporating it with optical satellite data and providing some of the most accurate AI models on the market. Radar data is different than most other satellite data because it is not dependent on weather or daylight, and thus is able to provide data on days where other satellites cannot because of clouds. Having data is extremely important for insurance companies no matter what the weather is like. That is the main reason why a lot of the insurance companies are looking forward to cooperating with us and vice versa.
Participants

Figure 1. The participants of the Agroinsurance International Conference 2022, image by www.agroinsurance.com.


Tanel Kobrusepp, product manager of crop insurance oriented services
 

2. Spring Barley Growth Monitoring with SNDVI
To limit the effect of clouds in optical remote sensing for farm monitoring, we developed a generative model to synthesize NDVI images from radar and historical optical data, which we name Synthetic NDVI (SNDVI).
 
In the past we have modeled and reported SNDVI on the level of image sub-tiles (512 x 512 px.). However, NDVI values are as important as the parcels they’re measured on hence recent evaluations have focused on in-field changes. This newsletter article discusses one such, exploring spring barley growth with SNDVI images.
 
NDVI increases with crop growth for spring barley till after the flowering stage (where it peaks) and decreases as plant dries up in preparation for harvest. By categorizing some of these growth stages, we have discovered the model accuracies for monitoring early growth, especially between sowing, seed emergence and early tillering could be significant, if not practical for monitoring fields in the absence of cloud-free images.
 
The parcels in Figure 2 & Figure 3 compare historical NDVI (HNDVI) inputs, predicted SNDVI and target NDVI images as the fields move from bare soil to what could be an early tillering stage in spring barley. The synthesized images can identify zones or direction of growth (as growth is not even across the field).

Comparison of NDVIs

Figure 2: Detecting early crop growth with SNDVI (HNDVI 18 days from NDVI). Target date: May 22, 2018

Comparison of NDVIs
Figure 3: Detecting early crop growth with SNDVI (HNDVI 18 days from NDVI). Target date: May 28, 2019

 
Hudson Taylor Lekunze, data analyst


Newsletter, September 2022

Please take 5 minutes and see what has kept us busy. In point three we invite you to vote.

1. KappaZeta at the International Astronautical Congress
KappaZeta was present in the IAC expo area in Estonia’s Software Defined Space booth from September 18-22. Over the five days in Paris, our COO Karoli Kahn had the pleasure of introducing KappaZeta to entrepreneurs, public sector representatives and students from all over the world. We had the opportunity to finally meet face to face with good partners and plant the seeds for new initiatives. Meeting professionals from all over the world provided good insight into new trends of agricultural services.

The expo with around 250 exhibitors also made it easy and interesting to gain a better perspective of the many public actors and companies all over the world that are operating Earth Observation satellites and providing the data for free or at a competitive price point. It was clear that EO activities are booming both upstream and downstream. However, very good SAR expertise is still rare – a problem KappaZeta can help with. It will be exciting to share all the new inspiring ideas with the KappaZeta team and set up follow-up meetings to discuss cooperation opportunities with the newly made contacts. Thank you to everyone who was interested to hear about KappaZeta and to Enterprise Estonia for organizing participation at the IAC with a joint Estonian booth.

Karoli at IAC

Figure 1. KappaZeta at the IAC with Estonia's Software Defined Space booth.

The next big conference where you can meet KappaZeta is Agroinsurance International Conference on 26-28 September in Tbilisi, Georgia, where product manager Tanel Kobrusepp will be exploring in-depth the ecosystem and opportunities of agricultural insurance.
 
Karoli Kahn, chief operating officer
 
2. KappaMask v2 - going global
The second phase of the KappaMask project is coming to an end. For the last year, we worked on extending KappaMask from Northern European terrestrial summer conditions to a global scale. We went through rainforests and deserts, mountains and oceans, cities and farmlands to create the one and only, KappaSet, the largest cloud and cloud shadow masks dataset. Not only are different surface types included in it but also different cloud types and seasons from all over the globe. We believe that KappaSet would benefit the research community and promote the further development of cloud masking methods.

KappaMask v2 was trained on KappaSet and hence naturally adapted for different geographical locations and seasons. Another improvement over the previous version is the faster processing time. It takes only around 3 minutes to get a classification mask. We compared KappaMask v2 with other cloud masking processors including Sen2Cor, Fmask, MAJA, IdePix and S2Cloudless on the challenging and diverse test set. KappaMask v2 demonstrated exceptional performance reaching the highest accuracy and outperforming all the above-mentioned methods.

KappaMask v2 is available to everyone here: https://github.com/kappazeta/km_predict

Feel free to check it out and share your experience with us!
Comparison of cloud masks
Figure 2. Results of KappaMask compared with other cloud mask and the corresponding true color image.
 
Tetiana Shtym, machine learning engineer

3. KappaZeta is in the finals
Tartu Postimees is organizing a nice contest to find out the product of the 2022 of the City of Tartu and Tartu county. If you agree that KappaZeta deserves the award please give us your vote at https://tartu.postimees.ee/7610731/haaletus-aita-valida-tartu-ja-tartumaa-aasta-toodet 


Newsletter, August 2022

Summer in Estonia is about to end and our team is getting more and more busy. These are the latest news from our team.

1. Important in August
Vacations are mainly over, and August has been a busy month in and around KappaZeta.

Firstly, I’m glad to welcome Tanel Kobrusepp to our team as the product manager of crop insurance oriented services from 1st of August. Tanel has important mission to lead our business and product development in the growing agri-insurance services market. KappaZeta has been traditionally strong in science and engineering and not so on the business side. With Tanel Kobrusepp joining this is about to improve. Coming months will tell. I wish Tanel a lot of courage to take the leadership, make clients happy and things happen.

While making deals and developing the business, it is equally important to be deep and have content. In August two important PhD thesis were defended in the University of Tartu. On 23rd of August KappaZeta’s friend and an unofficial team member Tauri Tampuu defended his PhD thesis “Synthetic Aperture Radar Interferometry as a tool for monitoring the dynamics of peatland”. After mine, Karlis Zalite’s, Aire Olesk’s and Tanel Tamm’s doctoral thesis this was the 5th doctoral thesis in the University of Tartu on synthetic aperture radar data. Welcome to the club, Tauri!

On 25th of August KappaZeta’s software developer Heido Trofimov defended his PhD thesis “Polluted clouds at air pollution hot spots help to better understand anthropogenic impacts on Earth’s climate”. An important piece from the critical puzzle to understand how we humans alter the climate on Earth. Heido is now the 5th member in KappaZeta team with doctor’s degree. Congratulations!

It has also been busy days with improving KappaMask, our free and open cloud mask for Sentinel-2, even if the results are not yet so visible. By August more than 5000 Sentinel-2 sub-tiles have been carefully manually labelled to give our AI model a high-quality global reference set, which is a pre-requisite for a globally accurate cloud mask – your AI model is just as good as your reference data. Great job, our interns Abdullah, Catherine, Joosep, Mariana and Nikita for doing the hard work and Olga & Tanya for managing the development and improving the model. I’m sure the results will be satisfying once we measure the accuracy and KappaMask will be an important tool for the world in taking full advantage of Sentinel-2 data.

Figure 1. Kaupo and Karoli in sTARTUp day 2022 business festival. A very useful networking event to establish new contacts and refresh the old ones.

 
Kaupo Voormansik, CEO of KappaZeta
 
2. Crop insurance oriented services
People and companies insure everything from cars, homes to travel expenses. Crop insurance does not seem to be as popular, at least in the European countries. For example, only 10.8% of farmers insured their fields in Poland, even though insurance was made compulsory in 2008 to receive full aid in case of a loss. Otherwise, the payment is reduced by 50%.
 
On the other hand, ~87% of crop fields in the U.S. are insured, with more than 1.1 million policies issued in 2021.
 
There are a lot of blockers on both, the producers and insurance side, that are causing the underuse of crop insurance. The main blockers for producers to insure their fields are price, lack of flexibility for custom terms from the insurance companies and lack of insurance products to choose from.
 
New technologies and data need to be incorporated into insurance companies’ work to increase the amount of tailor-made insurance policies in a cost-effective manner, thus increasing the interest in crop insurance products.
 
One option is to use Earth Observation (EO) data, which has endless opportunities from faster underwriting and better risk assessment to reducing costs on time-consuming and expensive claims handling. Imagine a loss adjuster going to a 3m high maize field, where a loss has occurred. Using satellite data, the loss adjuster can automatically see on a portable device, where exactly the damages happened, so he/she does not have to spend hours looking for the damaged areas. It is also very difficult to assess the extent of the damage with bare eyes, but it can definitely be done by AI, which could automatically determine how many acres were damaged, using satellite imagery.
 
Our main goal right now is to determine the biggest problems that the agricultural insurance providers face and prioritize our development of solutions according to their needs.
 
I am very excited to be working on these solutions with my amazing team of AI, Machine Learning and Earth Observation specialists. If this is a topic that you feel related to, don’t hesitate to get in touch with me at tanel.kobrusepp@kappazeta.ee.

Tanel Kobrusepp, product manager of crop insurance oriented services


Newsletter, July 2022

1. ESA Latvia Workshop 2022

The 12th ESA Training Course on Earth Observation (EO) took place in Riga Technical University from June 27 to July 1. Targeting researchers and young professionals from in and around the Baltic area in EO, we participated (remotely and in-person) in theory and practical sessions covering (Interferometric) SAR, and applications of satellite data in Forestry, Agriculture and Marine observations.


Figure 1. ESA Workshop in Latvia.


It was heart-warming to see our Mowing and Grazing projects featured in different presentations, and exciting to learn more about the implementation (and challenges) of agricultural projects similar to our current portfolio, directly from ESA experts. The sessions were a blend of interactive theory discussions and hands-on data processing or analyses of radar and optical satellite data with old and new GIS tools. These new tools (including ForestryTEP, OpenEO Cloud) are mostly focused on aggregating satellite data sources and easing access to, and processing on this data.

Special thanks to the ESA, Ministry of Education and Science (Latvia), the Institute of Environmental Solutions and Riga Technical University for putting this workshop together. Details on the course outline, and materials used are available here.
 
 Hudson Taylor Lekunze, data analyst
 
2. Space Festival in Tallinn
On 12th of July KappaZeta took part in the Space Festival "100 km from space" event in Tallinn. Our EO Analyst Jelizaveta Vabištševitš & Geospatial Data Quality Specialist Olga Wold with KappaZeta intern Abdullah Toqeer were talking about the importance of satellites and Copernicus programme in agricultural sector.


Figure 2. KappaZeta at the Space Festival.

Many great meetings with young bright space enthusiasts, who learned a lot about the space sector in Estonia and how could the open-source satellite data be used. Many Estonian companies working in the field of space presented their challenging work and impressive research.

Figure 3. KappaZeta at the Space Festival.

Also, the main guest of the festival was the European Space Agency astronaut Matthias Maurer! He spent 177 days at the International Space Station and returned from space in May this year. His inspiring presentation was about the beauty of our planet and the importance of taking care of it, as it is our only home (for now).

Olga Wold, geospatial data quality specialist


Newsletter, June 2022

1. Webinar: “KappaOne – Sentinel-1 data layers for the subsidy checks under CAP”



Sentinel-1 is a radar satellite that works both day and night and can see through the clouds and is, therefore, an excellent source of data for monitoring changes in agricultural land. The KappaOne service is designed to make Sentinel-1 SAR data easily accessible and ready to be used and analysed. It is extremely helpful for the subsidy checks under the Common Agricultural Policy (CAP). For example, it is valuable for planning field visits to the most problematic areas and getting an initial overview of the field before leaving the office. Moreover, it is applicable for checking the parcels manually, where the machine learning model gives an uncertain or borderline result. Sentinel-1 images are also well suited for assessment of mowing, harvesting, ploughing and other markers.
 
Join the KappaOne webinar to learn more about the service and its use cases for the subsidy check under CAP. The webinar will be held on the 30th of June from 15:00 until 16:00 EEST. You can register here: https://forms.gle/dzYjorhjrPNxRYym7 
 
 Jelizaveta Vabištševitš, EO analyst
 
2. KappaMask in cloud masks comparison
KappaZeta is taking part in the Cloud Mask Intercomparison Exercise (CMIX) meeting in ESA/ESRIN, 20-21 June. It is a framework to compare alternative cloud masks for Sentinel-2, Landsat-8/-9 and other widely used optical missions. Cloud mask processor developers and Earth Observation experts discuss and (hopefully) agree on a common way how to label the reference data and measure the accuracy of cloud masks. Taken the hard work our team has done I’m sure KappaMask will stand out here as one of the most accurate cloud masks for Sentinel-2.

Kaupo Voormansik, CEO, SAR expert

3. NISAR is coming
There are several new and interesting upcoming SAR missions including Copernicus ROSE-L, Sentinel-1 NG, ALOS-4 and NISAR. Perhaps the most interesting in the near future is NISAR jointly developed by NASA and ISRO (Indian Space Research Organisation). The satellite is almost ready with the testing and qualification procedure yet to be concluded before the launch, which is expected to early 2024. NISAR is very interesting mission because of three reasons, firstly it is a dual-band mission acquiring SAR data simultaneously in S- and L-band. It will be a Sentinel-1 like data factory imaging the whole planet several times per week and its data policy is planned to be free and open.


NISAR, image by NASA.

The last two points are very important for an operational service development point of view. KappaZeta will keep an eye on NISAR developments and help the world to make the most of this beautiful up-and-coming SAR data factory. If everything goes well people can access and use NISAR data in a simple way via our KappaOne service starting from 2024.
 
Kaupo Voormansik, CEO, SAR expert


Newsletter, May 2022


1. KappaZeta in the Living Planet Symposium
It is 23rd of May 2022 and we are glad to share that we are present at European Space Agency Living Planet Symposium in Bonn!



On Friday 27th of May, our CEO Kaupo Voormansik will give an oral presentation on “High quality food for AI - Sentinel-1 analysis-ready data (ARD) with interferometric coherence”. The presentation will focus on freshly launched KappaOne service. The SAR expertise of KZ helps to take full advantage of the interferometric and polarimetric data content of Sentinel-1. And for the end-user, it means calibrated and noise corrected imagery products with the highest possible spatial resolution using advanced speckle suppression methods.

On Tuesday our EO analyst Jelizaveta Vabistsevits will give a session on “KappaOne fresh Sentinel-1 data layers’ use case examples for the subsidy checks in Common Agricultural Policy”. She will tell you about KappaOne and its use case for the CAP subsidy checks through a hands-on experience!

On Friday there will also be a poster presentation on “AI-based Cloud Mask Processor for Sentinel-2” by Olga Wold, our Geospatial Data Quality Specialist. This presentation will describe the KappaMask model details and its progress on going global!
See you at the Living Planet Symposium 2022!
 
Olga Wold, geospatial data quality specialist
 
2. Crop Damage Detection
In KappaZeta we are among other things investigating the possibilities to use remotely sensed time-series data to detect crop damage on agricultural parcels.

The initial stage of the research was carried out with the cooperation of local farmers, who provided information about the general condition of agricultural parcels and granted access to the croplands for field surveying. During the in-person visits that took place early in July 2021, a visual assessment of various winter and spring vegetation types was performed and georeferenced images from across the fields were taken. Extra attention was paid to the difference between crop condition within damaged and healthy areas.


Figure 1. Damaged and healthy areas of winter wheat.

The results of the preliminary analysis on a test subset of agricultural parcels vary. The number of parcels in the current stage of analysis was quite limited and it caused several issues. Nevertheless, using simple unsupervised classification techniques it is possible to obtain a spatial representation of the damaged areas through the analysis of specific features of winter rapeseed at particular stages of its development.

At the next stage of the analysis, the work will be continued in the scope of supervised classification techniques combined with individual approaches to various agricultural crop types.

Anton Kostiukhin, software developer

Newsletter, April 2022

1. Advanced speckle filtering
Our raster processing got a major upgrade that significantly improves the quality of our backscatter and coherence rasters. We analyzed, modified and combined multiple published methods when designing a new filter for KappaOne. Although the filter adds significant load to the computations, the output raster will become much sharper with it as the edge structures are preserved much better.

To demonstrate the impact of a new filter to the backscatter images, we chose an area that has comb-shaped parcel on it (see the bottom right corner of figure 1). The narrow “teeth” of the comb are about 20 meters in width. Although “teeth” were also visible previously, their sharpness on KappaOne output is drastically higher.


Figure 1: Backscattering in VV (red) and VH (green) polarization previously (left) and with KappaOne (right) speckle filter.

The newly created KappaOne filter allows improving the quality of coherence images as well. On figure 2 the same area is used for demonstration as previously. Now the comb-like structure is gone, as for coherence more averaging is needed to produce meaningful output. There are large high coherence areas on the previous coherence image, which are created by objects that create extremely high coherence. One of those areas is marked by an ellipse on figure 2. Our novel filter helps to significantly shrink the area where those coherence values dominate. As a result, we can produce an image with much finer details on it.


Figure 2: Previously computed coherence image in VV polarization (left) and the one with KappaOne speckle filter (right).
 
Mihkel Veske, software developer
 
2. Towards production: Synthetic NDVI (SNDVI) rasters
The ultimate test of any machine learning model waits in production. Hence in the last month, we have focused on operationalizing the NDVI model as a layer of a new Analysis-Ready Data (ARD) product – KappaOne. In the ARD Germany demo alone, we prepared 35 rasters (each spanning 2500 sqkm) with a newly implemented semi-automated pipeline, that takes collocated Sentinel-1 and -2 products through preprocessing, inference, and post-processing tasks, to create NDVI or SNDVI rasters. 31 of these rasters in Northern Germany were entirely or partially created by our generative model due to cloudy Sentinel-2 images. For larger demo areas such as these, the production objective is to have a new NDVI or SNDVI image within 24 hours of newly acquired data from Sentinel-2.


Figure 3: Rasters created for Germany ARD demo (NDVI – green circles, SNDVI – white circles)

However, we are tackling new challenges at this stage, foremost of which is validating the SNDVI rasters. Validation involves evaluating the quality of images generated by our model, and the accuracy of parcel-level SNDVI compared with available NDVI. We have made significant progress in the former as reported in previous issues, however by the next issue on SNDVI we plan to have a comparable SNDVI timeseries (in addition to partial NDVI timeseries) in the demo map to validate newly created rasters. Most importantly, we want to learn more about the model as we scale and validate on new data and unseen areas, to improve the accuracy and reliability of predictions.
Hudson Taylor Lekunze, data analyst


Newsletter, March 2022

1. KappaMask for Sentinel-2
Phase II of the KappaMask development has been going in full power for a few months already! We started introducing the model new areas such as deserts in Africa and Australia, tropics in South America and even Arctic regions. Previously KappaMask was trained only for Northern terrestrial summer conditions, but now we are also preparing ground truth data for all the seasons. And thanks to the hard and precise work of our labellers, the first results of the new KappaMask model on the previously unseen areas are out and you can examine them on the figures below.


Figure 1. KappaMask predictions in water conditions. Yellow – clouds, blue – semi-transparent clouds.
 
Previously, KappaMask misclassified water as cloud shadows or semi-transparent clouds, or even as invalid pixels. Now, the model predictions look reasonable, however errors still occur.

On Figure 2, KappaMask predictions are presented in desert areas. You can see that most of the clouds and shadows are predicted correctly. Although, some of the terrain conditions are misclassified as cloud shadows and missed semi-transparent clouds are presented. Therefore, there is a lot of work still to be done!

Figure 2. KappaMask predictions in desert areas. Yellow – clouds, green – cloud shadows, blue – semi-transparent clouds.

 
Olga Wold, geospatial data quality specialist
 
2. Success at Kemira and Valtra Hackathons
KappaZeta participated in Kemira and Valtra hackathons on 9-10 and 22-23 of March in Finland. We teamed up with Lauri Karp and the partnership resulted in brilliant results. Our joint-team won the Kemira Hackathon with a concept of monitoring the end-to-end supply chains of next generation bio-based renewable materials with radar satellite data.

Few weeks later the same team got an honorable mention in Valtra Hackathon with a concept of automatic log generation of farming events (ploughing, sowing, mowing, harvesting etc.) for non-smart tractors.

 
More info at:
https://www.jamk.fi/fi/uutiset/2022/kemirahackathonin-voittajaksi-virolainen-biocarbon-satelliittidatan-avulla
and
https://www.jamk.fi/fi/uutiset/2022/traktorinkuljettajista-kimiraikkosia-personal-trainer-sovelluksen-avulla

Kaupo Voormansik, CEO, SAR expert


Newsletter, February 2022

1. Grazing detection based on Copernicus data
Grazing detection from Copernicus data for agricultural subsidy checks is one of our recently finished projects, carried out in collaboration with the European Space Agency (ESA) and with our partner, Czech EO company Gisat s.r.o. We developed a grazing detection methodology based on Sentinel-1 and Sentinel-2 imagery time series. This work is important for the reform of the Common Agricultural Policy, which will replace on-the-spot grazing checks with satellite monitoring. As about 20% of grasslands in Europe are being grazed, our detection methodology will help to complete the grassland maintenance checks alongside the mowing detection.
 
Fieldwork was a significant part of ground reference data collection required for the development of grazing detection methodology. We have collected the data about grazing and mowing events and grass height weekly during summer 2021 from several grasslands in central Estonia. Summary of the data together with S2 NDVI, NDWI, and PSRI indices and calculated grazing intensity for two grasslands is visualized in the figure below.


Figure 1. Time series of two grasslands in central Estonia for the May - September period.

On these grasslands, cows moved from one field to another throughout the summer, staying on each grassland for up to 5 weeks. In Figure 2 it is visible that grassland V2 was separated by the farmer into two parts. The larger part was mowed on the 30th of June for silage, which is visible from the NDVI, NDWI, and grass height drop on the time series. Animals were located on the smaller part of the V2 before and after the mowing event. Due to the mowing event in between, it is quite hard to distinguish changes caused by grazing activity, even though the grazing intensity is high (4.2 LU/ha).


Figure 2. Sentinel-2 NDVI raster time-series. Base map: Estonian Land Board

Changes in biomass on the grasslands are mostly visible on Sentinel-2 NDVI rasters, as the increase or decrease of NDVI values are more distinguishable by the human eye (Figure 2). In comparison, the visual analysis of Sentinel-1 raster parameters is more complex. The changes in VV or VH coherence values are difficult to detect if the exact location of animals is missing (Figure 3). However, the machine learning approach can be implemented in future works. As grazing is not distinguishable for the human eye on Sentinel-1 images, some of the signatures can be detected by the AI model.

Figure 3. Sentinel-1 coherence VV raster time-series. Base map: Estonian Land Board

 
Jelizaveta Vabištševitš, EO analyst, project manager
 
2. KappaZeta is going to Kemira and Valtra hackathons in Finland
KappaZeta and CarbonEye Global are to deliver new breed innovative solutions that fuse Carbon, Space and Agricultural markets to Kemira and Valtra hackathons in Finland in March 2022. Hackathons are organized by Jyväskylä University (JAMK) in cooperation with global segment leader Kemira Oyj (Chemicals company) and Valtra Inc. (Smart Tractors company). More here. For further enquiries please contact jurgen.lina@kappazeta.ee and lauri.karp@carboneye.ai.


Figure 4. Biomass View in KappaZeta ARD Demo.

Read more about the hackathons:
https://www.jamk.fi/en/project/biopaavo/biopaavo-hackathon/kemirahackathon
https://www.jamk.fi/en/project/biopaavo/biopaavo-hackathon/valtrahackathon

Kaupo Voormansik, CEO, SAR expert


Newsletter, January 2022

1. KappaOne
We are happy to announce KappaOne, a WMS service for selected Sentinel 1 based raster images and an API for parcel statistics. The WMS service gives an opportunity to either look at your particular area of interest in a web browser, like many mapping solutions currently work, or use the WMS directly in analytics program like QGIS.


Figure 1. The logo of KappaOne.

Currently, one can see backscattering in VH, a coherence image of VV in 6/12 interval, and RGB composite of previous two plus backscatter in VV. In addition, we provide a custom product – synthetic NDVI, a product where an AI algorithm derives current NDVI from Sentinel 1 image, using history of Sentinel 1 and Sentinel 2 data. This way, an NDVI can be given for dates when clouds do not permit optical satellites to get useful data.


Figure 2. KappaOne ARD demo.

For parcels, by which we mean a continuous agricultural area with size larger than 0.5 hectares, we provide statistical information, which can be viewed either in the web map or accessed directly via an API. The statistics uses the same information that is available in raster but aggregated over the parcel area. This data, represented as time series, allows detection of biomass changes and farming events like haying, harvesting and ploughing, but also natural events of significant influence, like flooding or hail damage.


Figure 3. KappaOne ARD demo - parcel view.

We provide two examples for the service in our initial demo, one for Germany https://demodev2.kappazeta.ee/ard_demo/ and one for Estonia https://demodev2.kappazeta.ee/ard_ee/. We encourage you to test the services and let us know what you think.
The KappaOne service development is funded by the InCubed program of ESA ɸ-lab.
 
Andres Luhamaa, KappaOne product owner
 
2. sNDVI rasters now available on our Demo Map
Since the last issue, we have prepared NDVI images for a 50×50km demo area in Northern Germany spanning 5 months (May to September). We replace cloudy image tiles with inferred synthetic NDVI (sNDVI) images, using historical NDVI and Synthetic Aperture Radar (SAR) data from the same area as inputs for the model. A sub-section of this demo area is pictured below (Figure 4), with the color scheme (ranging from brown through yellow to green) showing increasing areas of healthy vegetation, estimated by NDVI.
Figure 4. Synthetic NDVI Map of subarea in Germany Demo. Brown through yellow to Green, shows increasing healthy vegetation. Image source: KappaZeta Germany ARD Layers (September 29, 2021)


This demo is available in KappaZeta web map environment and includes other SAR data - backscatter and coherence - over the same area.

Interaction with red-outlined parcels, reveals rich temporal information of selected fields, as averaged parcel statistics over the entire demo period (April – December 2021) and can be used to compare accuracy of synthetic NDVI images. For now, not all areas have NDVI images for the entire period since historical inputs are required for image synthesis and some areas did not have sufficient cloud-free historical images for prediction. Months with dense NDVI coverage are May, June, and September.

An evaluation of model accuracy over some test images (212 images sized at 512px) from this period, shows synthetic images have a mean structural similarity (or reconstruction accuracy) of 82% and absolute pixel error of 6.86%. The next challenge is making all synthetic images available, even for areas with insufficient historical data.
 
Hudson Taylor Lekunze, Data analyst