European Space Agency enters into new partnership with Dotphoton to to ensure reliability of Earth Observation data
We’re excited to announce Dotphoton’s new partnership with the European Space Agency (ESA) to continue the development of a metrologically correct image compression technology for space applications, in order to improve reliability and scalability of AI-based analysis for Earth Observation (EO) data.

The purpose of the new contract is to further develop and extensively test Dotphoton's metrological compression technology for EO image data. The final product not only aimed at improving the performance and reliability of raw image data, but also reducing CO2 emissions associated with large data and reducing costs in processing and storing raw image data by ESA in satellite-to-ground image transition, one of the vital cost drivers for all modern missions.
“The world is relying more and more on remote sensing satellite data, but those daily petabytes come at an extremely high cost. This is accelerating the paradigm shift from the traditional 'raw data delivery' to the more desirable 'information delivery'. Here is where Dotphoton’s technology combined with the unique knowledge of our payloads can achieve a major accomplishment: preserve the image quality of our data while attaining compression ratios only possible with high information loss in the past"

Roberto Camarero, Onboard Payload Data Processing Engineer at ESA.
Dotphoton's metrologically correct compression combines the highest quality of lossless image formats with the flexibility and speed of lossy formats.

Eugenia Balysheva, CEO Dotphoton
“Dotphoton's metrologically correct compression combines the highest quality of lossless image formats with the flexibility and speed of lossy formats."
Dotphoton continues to work with ESA under GSTP (General Support Technology Programme), the signed contract will bring Dotphoton's technology to the Technology Readiness Level (TRL) 5.
Why this is the big step for the space industry?

ESA is the world leader in Earth Observation (EO). EO data plays a crucial role in studying, analysing, understanding and predicting natural processes and human-related activities.

Recently, the term ‘big Earth data’ emerged to describe massive EO datasets that confronts analysts and their traditional workflows with a range of challenges (Sudmanns et al., 2019). It is believed that in order for the data to be ‘big’, the following 5 ‘V’s’ need to be present: volume of data, velocity (how quickly data is generated, processed and moved around), variety (diversity of forms and types), and veracity (quality and accuracy of data), and value the data provides for further analysis. Just like traditional big data, the main challenge in big Earth Data is an increase in data volume and handling of the data, as well as its veracity.

According to the Copernicus Sentinel Data Access Annual Report released in September 2021, 8.6 PB (7.65 PiB) of user-level data from Sentinel satellites was published in 2020, including 4.7 PB (4.15 PiB) from Sentinel 2. That is 23.6 TB and 12.9 per day respectively. Data volume keeps growing steadily year on year since the start of Operations in 2014, accounting for 2% of all published data volume in 2015, to 20% in 2018, and 31% in 2020.

Dotphoton has achieved up to 7:1 compression for Earth Observation images, and with images being up to 7 times smaller, it brings about a large impact on how data could be stored and handled. Previously, ESA and Dotphoton stated that “The activity (collaboration between Dotphoton and ESA) confirmed that raw images maintained their quality even when compressed down to a ratio of 10:1. Ultimately, this could reduce costs for data transfer and data storage for ESA’s Earth Observation programme”.
EO is one of the first markets where the amount of images produced exceeds the AI processing power. The amount of data is extraordinary. However, as noted by Google researchers in a recent paper, “data is the most under-valued and de-glamorised aspect of AI”. While emphasis has been placed on analysis algorithms and handling of data, the quality of the data hasn’t been as high as it could have been.

Typical compression algorithms such as JPEG2000 are developed for human eye analysis, not machine vision and AI. Such traditional compression methods often slow down or even prevent processing of large data sets, and may also generate artefacts, which in turn could create flaws in the analyses and substantially limit the results.

In contrast, while reducing the file size, Dotphoton compression preserves its original raw information. This allows for a much higher image data quality (even with existing systems) that is essential for a reliable and fast AI-based image analysis.
EO is one of the first markets where the amount of images produced exceeds the AI processing power. The amount of data is extraordinary. However, as noted by Google researchers in a recent paper, “data is the most under-valued and de-glamorised aspect of AI”. While emphasis has been placed on analysis algorithms and handling of data, the quality of the data hasn’t been as high as it could have been.

Typical compression algorithms such as JPEG2000 are developed for human eye analysis, not machine vision and AI. Such traditional compression methods often slow down or even prevent processing of large data sets, and may also generate artefacts, which in turn could create flaws in the analyses and substantially limit the results.

In contrast, while reducing the file size, Dotphoton compression preserves its original raw information. This allows for a much higher image data quality (even with existing systems) that is essential for a reliable and fast AI-based image analysis.
About Dotphoton
Dotphoton is a Swiss technology company that develops innovative data management solutions for large image data to improve its reliability, scalability and sustainability. Company’s deep knowledge and latest insights from quantum physics ensures it stays ahead as the highly reliable partner in the imaging field, trusted by the European Space Agency, Bosch, leading camera manufacturers and the most innovative biotech centres and labs across the world.
Contact
Eugenia Balysheva
Dotphoton AG
eugenia.balysheva@dotphoton.com
+41 0 41 552 50 00