Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures

Deep learning methods are often used for image classification or local object segmentation. The corresponding test and validation data sets are an integral part of the learning process and also of the algorithm performance evaluation. High and particularly very high-resolution Earth observation (EO)...

Ful tanımlama

Kaydedildi:
Detaylı Bibliyografya
Asıl Yazarlar: Schwarz, Gottfried, Octavian Dumitru, Corneliu, Datcu, Mihai
Materyal Türü: Online
Dil:İngilizce
Baskı/Yayın Bilgisi: InTechOpen 2021
Konular:
Online Erişim:ONIX_20210602_10.5772/intechopen.90910_465
Etiketler: Etiketle
Etiket eklenmemiş, İlk siz ekleyin!
_version_ 1863732195919659008
author Schwarz, Gottfried
Octavian Dumitru, Corneliu
Datcu, Mihai
author_browse Datcu, Mihai
Octavian Dumitru, Corneliu
Schwarz, Gottfried
author_facet Schwarz, Gottfried
Octavian Dumitru, Corneliu
Datcu, Mihai
author_sort Schwarz, Gottfried
collection Directory of Open Access Books
description Deep learning methods are often used for image classification or local object segmentation. The corresponding test and validation data sets are an integral part of the learning process and also of the algorithm performance evaluation. High and particularly very high-resolution Earth observation (EO) applications based on satellite images primarily aim at the semantic labeling of land cover structures or objects as well as of temporal evolution classes. However, one of the main EO objectives is physical parameter retrievals such as temperatures, precipitation, and crop yield predictions. Therefore, we need reliably labeled data sets and tools to train the developed algorithms and to assess the performance of our deep learning paradigms. Generally, imaging sensors generate a visually understandable representation of the observed scene. However, this does not hold for many EO images, where the recorded images only depict a spectral subset of the scattered light field, thus generating an indirect signature of the imaged object. This spots the load of EO image understanding, as a new and particular challenge of Machine Learning (ML) and Artificial Intelligence (AI). This chapter reviews and analyses the new approaches of EO imaging leveraging the recent advances in physical process-based ML and AI methods and signal processing.
format Online
id doab-20.500.12854ir-70471
institution Directory of Open Access Books
language eng
publishDate 2021
publishDateRange 2021
publishDateSort 2021
publisher InTechOpen
publisherStr InTechOpen
record_format ojs
spelling doab-20.500.12854ir-704712025-08-13T14:11:57Z Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures Schwarz, Gottfried Octavian Dumitru, Corneliu Datcu, Mihai Earth observation, synthetic aperture radar, multispectral, machine learning, deep learning thema EDItEUR::U Computing and Information Technology thema EDItEUR::U Computing and Information Technology Deep learning methods are often used for image classification or local object segmentation. The corresponding test and validation data sets are an integral part of the learning process and also of the algorithm performance evaluation. High and particularly very high-resolution Earth observation (EO) applications based on satellite images primarily aim at the semantic labeling of land cover structures or objects as well as of temporal evolution classes. However, one of the main EO objectives is physical parameter retrievals such as temperatures, precipitation, and crop yield predictions. Therefore, we need reliably labeled data sets and tools to train the developed algorithms and to assess the performance of our deep learning paradigms. Generally, imaging sensors generate a visually understandable representation of the observed scene. However, this does not hold for many EO images, where the recorded images only depict a spectral subset of the scattered light field, thus generating an indirect signature of the imaged object. This spots the load of EO image understanding, as a new and particular challenge of Machine Learning (ML) and Artificial Intelligence (AI). This chapter reviews and analyses the new approaches of EO imaging leveraging the recent advances in physical process-based ML and AI methods and signal processing. 2021-02-10T12:58:18Z 2021-06-02T10:13:02Z 2020 chapter ONIX_20210602_10.5772/intechopen.90910_465 https://library.oapen.org/handle/20.500.12657/49351 https://directory.doabooks.org/handle/20.500.12854/70471 eng open access image/jpeg image/jpeg image/jpeg n/a n/a n/a https://library.oapen.org/bitstream/20.500.12657/49351/1/70789.pdf https://library.oapen.org/bitstream/20.500.12657/49351/1/70789.pdf https://library.oapen.org/bitstream/20.500.12657/49351/1/70789.pdf InTechOpen 10.5772/intechopen.90910 10.5772/intechopen.90910 035ecc65-6737-43cf-a13a-6bdf67ce01f4 open access
spellingShingle Earth observation, synthetic aperture radar, multispectral, machine learning, deep learning
thema EDItEUR::U Computing and Information Technology
thema EDItEUR::U Computing and Information Technology
Schwarz, Gottfried
Octavian Dumitru, Corneliu
Datcu, Mihai
Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures
title Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures
title_full Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures
title_fullStr Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures
title_full_unstemmed Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures
title_short Chapter Deep Learning Training and Benchmarks for Earth Observation Images: Data Sets, Features, and Procedures
title_sort chapter deep learning training and benchmarks for earth observation images data sets features and procedures
topic Earth observation, synthetic aperture radar, multispectral, machine learning, deep learning
thema EDItEUR::U Computing and Information Technology
thema EDItEUR::U Computing and Information Technology
topic_facet Earth observation, synthetic aperture radar, multispectral, machine learning, deep learning
thema EDItEUR::U Computing and Information Technology
thema EDItEUR::U Computing and Information Technology
url ONIX_20210602_10.5772/intechopen.90910_465
work_keys_str_mv AT schwarzgottfried chapterdeeplearningtrainingandbenchmarksforearthobservationimagesdatasetsfeaturesandprocedures
AT octaviandumitrucorneliu chapterdeeplearningtrainingandbenchmarksforearthobservationimagesdatasetsfeaturesandprocedures
AT datcumihai chapterdeeplearningtrainingandbenchmarksforearthobservationimagesdatasetsfeaturesandprocedures