1 / 11

Change Detection in France Using Sentinel-2 Satellite Images

This project aims to develop a platform with high-performance computing features and cloud technology to process large-scale diverse data. The use case focuses on detecting generic changes in satellite images over the entire France during a one-year period, with a revisit period of 10 days. The project addresses the issue of storage and access, as each Sentinel-2 image has a data volume of about 700 MB and covers a granule of 10,000 x 10,000 pixels.

pnewsome
Download Presentation

Change Detection in France Using Sentinel-2 Satellite Images

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Generic Change Detection all over France Michelle AUBRUN michelle.aubrun@thalesaleniaspace.com

  2. Summary Project presentation Use case presentation Issue encountered

  3. 1 Project presentation

  4. Project : Evolve H2020 : HPC and Cloud enhanced Test-beds for Extracting Value from Large-scale Diverse Data • 19 partners (11 European countries) Objectif : Ability to process unprecedented amounts of data and handle demanding computations via : • Platform with HPC features (HPC world) • Versatile database processing stack for end-to-end workflows (Big Data world) • Easy deployment, access, use and share (Cloud world)

  5. 2 Use case presentation

  6. Use case : Change Detection Satellite image : Sentinel-2 Objectif : Detect generic changes over the entire France during 1 year with a revisit period of 10 days. Lot of applications in many diverse sectors : • Land surveillance (deforestation, urbanization) • Disaster (forest fire, frozen crop) • Climatic change

  7. Use case : Change Detection Encode Sentinel-2 images into a relevant feature spacewith an adequat loss function : • Insensitive to atmospheric conditions Compute a distance metric in this space in order to generate change detection map

  8. 3 Issue encountered

  9. Main issue : Storage and Access One Sentinel-2 image : • Data volume : about 700 Mo • Granule : 10 000 x 10 000 pixels (100km x 100km) Use case objectif : • About 100 granules to cover France territory • 35 revisit for a period of 1 year Storage = About 2,5 To of input data + storage of results (e.g. change detection map)

More Related