The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data
ExtremeEarth is a European H2020 project; it aims at developing analytics techniques and technologies that combine Copernicus satellite data with information and knowledge extraction, and exploiting them on ESA’s Food Security and Polar Thematic Exploitation Platforms. The current publication focuse...
Main Authors: | , , , |
---|---|
Format: | Conference Object |
Language: | unknown |
Published: |
2022
|
Subjects: | |
Online Access: | https://elib.dlr.de/186539/ https://lps22.esa.int/frontend/index.php |
id |
ftdlr:oai:elib.dlr.de:186539 |
---|---|
record_format |
openpolar |
spelling |
ftdlr:oai:elib.dlr.de:186539 2024-05-19T07:48:24+00:00 The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data Dumitru, Corneliu Octavian Schwarz, Gottfried Yao, Wei Datcu, Mihai 2022 https://elib.dlr.de/186539/ https://lps22.esa.int/frontend/index.php unknown Dumitru, Corneliu Octavian und Schwarz, Gottfried und Yao, Wei und Datcu, Mihai (2022) The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data. Living Planet Symposium 2022, 2022-05-23 - 2022-05-27, Bonn, Germany. EO Data Science Konferenzbeitrag NonPeerReviewed 2022 ftdlr 2024-04-25T00:59:55Z ExtremeEarth is a European H2020 project; it aims at developing analytics techniques and technologies that combine Copernicus satellite data with information and knowledge extraction, and exploiting them on ESA’s Food Security and Polar Thematic Exploitation Platforms. The current publication focuses on the Polar case for which a large training dataset has been generated and demonstrates the use of different machine learning/deep learning techniques (e.g., Compression-based pattern recognition, Cascaded learning for semantic labelling, Explainable AI for SAR sea-ice content discovery, Physics-aware deep hybrid architecture). The solution proposed in the project is an active learning approach that represents a simple way to generate semantically annotated datasets from given Sentinel-1/Sentinel-2 images. Active Learning is a form of supervised machine learning in which the learning algorithm is able to interactively query some (human) information source to obtain the desired image classification outputs at new data points. The key idea behind active learning is that a machine learning algorithm (e.g., a Support Vector Machine) can achieve greater accuracy with fewer training labels if it’s allowed to choose itself among the data from which it learns. In this case, relevance feedback is included; this supports users to search for additional images of interest in a large repository. Further, any new image content classes that do not exist yet can be defined by expert users based on their specific knowledge; however, different users can give different meaning to the new classes. In this case, the number of semantic classes that can be extracted with the proposed active learning method is not fixed (as for many current state-of-the-art classification methods), but the classes are defined interactively by the users. In our case, we apply our active learning scheme to sequences of pre-selected satellite images. The idea behind this active learning approach is to obtain high classification accuracies (between 85% and ... Conference Object Sea ice German Aerospace Center: elib - DLR electronic library |
institution |
Open Polar |
collection |
German Aerospace Center: elib - DLR electronic library |
op_collection_id |
ftdlr |
language |
unknown |
topic |
EO Data Science |
spellingShingle |
EO Data Science Dumitru, Corneliu Octavian Schwarz, Gottfried Yao, Wei Datcu, Mihai The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data |
topic_facet |
EO Data Science |
description |
ExtremeEarth is a European H2020 project; it aims at developing analytics techniques and technologies that combine Copernicus satellite data with information and knowledge extraction, and exploiting them on ESA’s Food Security and Polar Thematic Exploitation Platforms. The current publication focuses on the Polar case for which a large training dataset has been generated and demonstrates the use of different machine learning/deep learning techniques (e.g., Compression-based pattern recognition, Cascaded learning for semantic labelling, Explainable AI for SAR sea-ice content discovery, Physics-aware deep hybrid architecture). The solution proposed in the project is an active learning approach that represents a simple way to generate semantically annotated datasets from given Sentinel-1/Sentinel-2 images. Active Learning is a form of supervised machine learning in which the learning algorithm is able to interactively query some (human) information source to obtain the desired image classification outputs at new data points. The key idea behind active learning is that a machine learning algorithm (e.g., a Support Vector Machine) can achieve greater accuracy with fewer training labels if it’s allowed to choose itself among the data from which it learns. In this case, relevance feedback is included; this supports users to search for additional images of interest in a large repository. Further, any new image content classes that do not exist yet can be defined by expert users based on their specific knowledge; however, different users can give different meaning to the new classes. In this case, the number of semantic classes that can be extracted with the proposed active learning method is not fixed (as for many current state-of-the-art classification methods), but the classes are defined interactively by the users. In our case, we apply our active learning scheme to sequences of pre-selected satellite images. The idea behind this active learning approach is to obtain high classification accuracies (between 85% and ... |
format |
Conference Object |
author |
Dumitru, Corneliu Octavian Schwarz, Gottfried Yao, Wei Datcu, Mihai |
author_facet |
Dumitru, Corneliu Octavian Schwarz, Gottfried Yao, Wei Datcu, Mihai |
author_sort |
Dumitru, Corneliu Octavian |
title |
The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data |
title_short |
The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data |
title_full |
The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data |
title_fullStr |
The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data |
title_full_unstemmed |
The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data |
title_sort |
extremeearth project: a solution for generating reliable semantically annotated data |
publishDate |
2022 |
url |
https://elib.dlr.de/186539/ https://lps22.esa.int/frontend/index.php |
genre |
Sea ice |
genre_facet |
Sea ice |
op_relation |
Dumitru, Corneliu Octavian und Schwarz, Gottfried und Yao, Wei und Datcu, Mihai (2022) The ExtremeEarth project: A Solution for Generating Reliable Semantically Annotated Data. Living Planet Symposium 2022, 2022-05-23 - 2022-05-27, Bonn, Germany. |
_version_ |
1799466647978770432 |