Despite the current vacation season, the last weeks were busy here at Sinergise. We would love to share some of the highlights with you.
There are also plenty of opportunities to meet us in person at various events listed below. During the PhiWeek in Frascati, Italy, we will host a dedicated session on Data Cubes - Getting the Most Out of EO Data With the Help of Data Cube Services and ML Tools on September 12th. For those attending, please reserve your seat by marking this side event in your registration.
The Wonderful World of Custom Scripts
The success of the Sentinel Hub Custom Script Contest organized this Spring motivated us to repeat it in the Autumn. The winning scripts of the Contest are at your disposal on our GitHub repository. Stay tuned and follow us on Twitter to find out more about the contest.
As preparation for the next challenge, we have created a short Custom Script Tutorial which will help you create beautiful and informative satellite images for your own purpose. With a little help of our friendly pdf tutorial you will be able to create masterpieces in no time.
Visit our Flickr account to see the images in full resolution.
The Sentinel-3 SLSTR imager provides global and regional Sea and Land Surface Temperature measurements. It is useful for climate change monitoring, vegetation monitoring, land and sea surface temperature monitoring, and active fire detection.
EO Browser provides data acquired in nadir view in the descending pass. The entire dataset including the oblique view is available through API requests. Measurements represent the top of atmosphere (TOA) reflectance or brightness temperature. Spatial resolution of the available product is generated at two resolutions, 500 m resolution for solar reflectance bands and 1 km resolution for thermal infrared bands. Using two satellites enables you to inspect the data on a daily basis.
July's heatwave as seen on July 23, 2019 over France with the Sentinel-3 SLSTR's thermal infrared band S9, representing the brightness temperature. See it in EO Browser.
Land Cover Monitoring System
Upcoming blog post
If you liked our trilogy about land cover classification with eo-learn (part 1, part 2 and part 3), we have excellent news. Coming up in August is a blog post about the next level in land observation - the Land Cover Monitoring System. The system connects our latest research about land cover classification, crop type classification and change detection with data processing tools from eo-learn. This is joined with the latest Bring Your Own Data functionality of Sentinel Hub services and Geopedia. Finally, there is also a web application for land monitoring. Stay tuned to find out how and where we made it production-ready.
The blog post will cover everything, from the ground truth reference data, data processing pipelines in Python, data storage, to front end application.
Example: Crowdsourcing the water detection dataset for BlueDot application
Calling all in need of a real-life machine learning application for satellite data! Experience the ClassificationApp, a web-application tool to help you annotate satellite images in an easy and intuitive way. Its recent major feature improvement is to create your own labelling campaigns and make them available to others for crowdsourcing. For more information about its advantages and how to use it, follow the link below. It will also take you through the dedicated public campaign Water-Body Segmentation Correction. It was created for the needs of the BlueDot Water Observatory application which already has the coverage of about 10 000 water bodies.
Example of task suggested by the water body segmentation campaign.
The project has started and one of the first steps is already complete. Check out the Sentinel Hub's functionality, Bring Your Own Data, which is useful for satellite imagery, aerial imagery, machine learning results and other raster data. To learn more about it and the project itself, see our blog.
The announcement of the Awarded Projects
After announcing our support of GEO - Amazon Programme back in April, we are looking forward to seeing the results of the awarded projects. The resources will help 21 projects from 17 developing countries to realize the potential of EO for sustainable development. Our interest is particularly focused on those using Sentinel Hub services to empower their EO applications. Continue reading!
The Sentinel Hub API is being used by IBISA to develop a service of crop protection for smallholder farmers in developing countries
To satisfy the challenge of low-cost operations, IBISA combined the use of blockchain and EO data. With a help of the Sentinel Hub service they managed to avoid the need for ground truth data to do field assessments. To learn more about IBISA's need for EO data, risk modelling and their Watchers Platform for "crowd-watching" we recommend reading our guest blog post by Jean-Baptiste Playnet.
A group of plots in true color (left), NDVI (middle) and NDVI anomaly (right).
BlueDot and Land Cover Classification Application at Living Planet Symposium
Our research team presented at #LPS19 in Milan, Italy in May
We had a presentation about our Query Planet project, under the session of Machine Learning with EO Big Data at Scale. In this talk we presented the need for a generic set of tools for collecting and annotating data, as well as handling all this data in an efficient and elegant manner. We showcased some of the tools which can be used to easily and intuitively create beautiful and useful applications, such as the BlueDot Water Observatory or various applications of land cover classification.
Check out the posters at the given links and try out the tools for yourself. We will be happy to hear all about it.
To meet us in person and discuss your needs, send us an email to [email protected] and schedule a meeting with us at the following events:
H2020 Big Data Hackathon, Frascati, Italy
Please do not hesitate to send us any feedback. You can also use our Forum
or meet us in person at the upcoming events. We look forward to hearing from you.
The Sentinel Hub team