We hope you are staying safe and healthy even as the current global situation disrupts our everyday life. During the lockdown we are keeping the spirits up and looking for new ways to bring satellite data closer to you.
In this edition we introduce the availability of commercial datasets and a global archive of Sentinel-2 L2A data, a new open source library, new EO Browser features, many useful custom scripts, system updates and more. Last but certainly not least, don't miss a chance to participate in a special edition of Custom Script Contest aiming to help monitor and mitigate the coronavirus COVID-19 situation.
Use the simplicity of the Sentinel Hub to easily access commercial data of higher resolution at an affordable price
We are delighted to launch a new service providing easy access to very high-resolution data, starting with Planet and Airbus and expanding to others soon. The staged roll-out started one week ago and things are running smoothly - we have ingested more than 20 000 square km already. If you are interested to try it out, check the documentation and then contact us.
Read more about it in our blog post.
Calling for remote sensing experts, machine learning scientists and the interested public to contribute ideas
In cooperation with the European Space Agency and the European Commission, we are organizing a special edition of Custom Script Contest, focused on the effects of the COVID-19 crisis. We are in the quest for ideas on how satellite data could help monitor and mitigate the situation for the upcoming months. Use a specially prepared environment where you will find High Resolution Pleiades and PlanetScope data in addition to our standard EO missions’ data offer. Explore a plethora of tools ready to be used for this purpose – starting from EO Browser, Sentinel Hub and xcube services, eo-learn, Jupyter Notebooks and more; something for any skill level and for every analysis.
The opportunity is too important to ignore and just about anyone can help - each and every idea matters. Visit the official Contest web page and read the latest blog post for more!
The processing of global archive of L2A is reaching the end. Once it will complete, all the scenes worldwide since 2017 with less than 90% cloud coverage will be available as atmospherically corrected (L2A, Bottom of the atmosphere) data. At the moment, data processing goes back to July 2017. Stay updated by following our Forum!
An example of non-atmospherically corrected data (L1C) on the left and atmospherically corrected data (L2A) on the right.
We have added a new service, managed cloud-based PostGIS database, called EDC GeoDB, perfect for storage of vector data. You can access it seamlessly using a Python client. The expanded collection of the available Jupyter Notebooks will help you use various APIs with examples on LPIS crop classification, xcube data analysis and much more.
Visit EDC Marketplace to explore or create your own Jupyter Notebooks, data products and applications for yourself and others to use!
Open-source library for satellite imagery powered web applications
Experimenting with MixMatch and Earth observation data
While machine learning is a field that is already appreciably useful in Earth observation, its limits are being constantly expanded. In the past few months, we were looking at how we could use semi-supervised learning - a form of learning, where we use labelled as well as unlabelled data - to improve the performance of deep neural networks in classification of satellite imagery. Besides enabling us to achieve high performance with less labelled data, our results lead to an optimistic implication: that we can use labelled data, sourced only from a small, bounded, geographic region, and still perform well in other (unseen) regions by adding unlabelled samples from those areas.
Diagram of the label guessing process used in MixMatch.
EO Browser updates
With each update, EO Browser becomes a more powerful tool for exploring our Planet. We can easily observe our impact, hopefully giving us a chance to do something about it and to contribute to a brighter future of our environment. Therefore, we are proudly presenting the new features, such as the Education mode, which offers 12 devoted themes, multi-temporal scripting, better control of time intervals, improved time-lapse filters, and five other cool adjustments. Dive in!
Getting started guide
A quick introduction to remote sensing for teachers who would like to use satellite imagery in class is brought to you by Pierre Markuse. Pierre is a freelance remote sensing expert with a special interest in optical remote sensing, and his images are being used by numerous media outlets. To learn more about the usage of satellite data, how to process and visualize it, and about different resources on how to get started, read his guest blog post.
Learn more about the new scripts from the second round of the Sentinel Hub Custom Script Contest
Get inspired by the scripts submitted in the second round of the Contest, presented in our blog post, and submit your own script in the next round. We have received 23 beautiful and useful remote sensing scripts that you can explore globally in EO Browser. The results are published on the official Contest website and the entries are available in our Custom Scripts repository.
Transition to Evalscript Version 3
To simplify the usage of different versions of custom scripts in the processing API and our web applications, EO Browser and Sentinel Playground, we have decided to deprecate versions 1 and 2, and keep only the basic version and version 3. All the scripts in the Custom Script repository using deprecated versions were already converted to V3. For now, you can still use V1 and V2 as before. In the future, as we complete the deprecation phase, your custom scripts will be transformed to V3 before saving your newly created layers in the Configurator. To transform deprecated versions, you can also use our handy converter. Learn more about V3 in our Evalscript documentation.
- Aggregated statistic insight on the historic usage - We have started aggregating the usage statistics of requests and processing units from the Elasticsearch to a persistent PostgreSQL database. Before the aggregated statistics were stored in the database, the statistics were calculated on the fly each time you have asked for it. Because of tens of millions of daily logs, the new approach is faster, less prone to failures, and enables us to display historic statistics further back in time.
You can see this aggregated history of your statistics for executed API and OGC requests under the Account Settings. The data is available since October 2019 and can be exported as Excel and CSV format.
- Copy your configuration to another user - You can now copy your configurations in the Configuration Utility to another Sentinel Hub user simply by following the instructions here.
In this unusual situation caused by the COVID-19 we have taken the necessary measures to maintain the safety of our team while continuing to maintain all operations. Since March 13th we are working from home - it's not much further to the clouds, where we operate our systems, just a bit less social.
We can therefore not invite you to any event in the near future. So please, do send us your feedback via Forum
or social media. We look forward to hearing from you. Stay safe!
The Sentinel Hub team
You received this email because you subscribed to our list. You can unsubscribe at any time. To see previous newsletters, click here.