Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on Landcover
The purpose of this study was to construct artificial intelligence (AI) training datasets based on multi-resolution remote sensing and analyze the results through learning algorithms in an attempt to apply machine learning efficiently to (quasi) real-time changing landcover data. Multi-resolution da...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-05-01
|
| Series: | Frontiers in Remote Sensing |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/frsen.2022.832753/full |
| _version_ | 1797963771777384448 |
|---|---|
| author | Seong-Hyeok Lee Moung-Jin Lee |
| author_facet | Seong-Hyeok Lee Moung-Jin Lee |
| author_sort | Seong-Hyeok Lee |
| collection | DOAJ |
| description | The purpose of this study was to construct artificial intelligence (AI) training datasets based on multi-resolution remote sensing and analyze the results through learning algorithms in an attempt to apply machine learning efficiently to (quasi) real-time changing landcover data. Multi-resolution datasets of landcover at 0.51- and 10-m resolution were constructed from aerial and satellite images obtained from the Sentinel-2 mission. Aerial image data (a total of 49,700 data sets) and satellite image data (300 data sets) were constructed to achieve 50,000 multi-resolution datasets. In addition, raw data were compiled as metadata in JavaScript Objection Notation format for use as reference material. To minimize data errors, a two-step verification process was performed consisting of data refinement and data annotation to improve the quality of the machine learning datasets. SegNet, U-Net, and DeeplabV3+ algorithms were applied to the datasets; the results showed accuracy levels of 71.5%, 77.8%, and 76.3% for aerial image datasets and 88.4%, 91.4%, and 85.8% for satellite image datasets, respectively. Of the landcover categories, the forest category had the highest accuracy. The landcover datasets for AI training constructed in this study provide a helpful reference in the field of landcover classification and change detection using AI. Specifically, the datasets for AI training are applicable to large-scale landcover studies, including those targeting the entirety of Korea. |
| first_indexed | 2024-04-11T01:33:28Z |
| format | Article |
| id | doaj.art-436b1a93d99e42658666ffb21e115bd0 |
| institution | Directory Open Access Journal |
| issn | 2673-6187 |
| language | English |
| last_indexed | 2024-04-11T01:33:28Z |
| publishDate | 2022-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Remote Sensing |
| spelling | doaj.art-436b1a93d99e42658666ffb21e115bd02023-01-03T09:17:07ZengFrontiers Media S.A.Frontiers in Remote Sensing2673-61872022-05-01310.3389/frsen.2022.832753832753Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on LandcoverSeong-Hyeok Lee0Moung-Jin Lee1Division for Environmental Planning, Korea Environment Institute, Sejong, South KoreaCenter for Environmental Data Strategy, Korea Environment Institute, Sejong, South KoreaThe purpose of this study was to construct artificial intelligence (AI) training datasets based on multi-resolution remote sensing and analyze the results through learning algorithms in an attempt to apply machine learning efficiently to (quasi) real-time changing landcover data. Multi-resolution datasets of landcover at 0.51- and 10-m resolution were constructed from aerial and satellite images obtained from the Sentinel-2 mission. Aerial image data (a total of 49,700 data sets) and satellite image data (300 data sets) were constructed to achieve 50,000 multi-resolution datasets. In addition, raw data were compiled as metadata in JavaScript Objection Notation format for use as reference material. To minimize data errors, a two-step verification process was performed consisting of data refinement and data annotation to improve the quality of the machine learning datasets. SegNet, U-Net, and DeeplabV3+ algorithms were applied to the datasets; the results showed accuracy levels of 71.5%, 77.8%, and 76.3% for aerial image datasets and 88.4%, 91.4%, and 85.8% for satellite image datasets, respectively. Of the landcover categories, the forest category had the highest accuracy. The landcover datasets for AI training constructed in this study provide a helpful reference in the field of landcover classification and change detection using AI. Specifically, the datasets for AI training are applicable to large-scale landcover studies, including those targeting the entirety of Korea.https://www.frontiersin.org/articles/10.3389/frsen.2022.832753/fulllandcoverAItraining datasetsannotationmachine learning |
| spellingShingle | Seong-Hyeok Lee Moung-Jin Lee Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on Landcover Frontiers in Remote Sensing landcover AI training datasets annotation machine learning |
| title | Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on Landcover |
| title_full | Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on Landcover |
| title_fullStr | Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on Landcover |
| title_full_unstemmed | Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on Landcover |
| title_short | Comparisons of Multi Resolution Based AI Training Data and Algorithms Using Remote Sensing Focus on Landcover |
| title_sort | comparisons of multi resolution based ai training data and algorithms using remote sensing focus on landcover |
| topic | landcover AI training datasets annotation machine learning |
| url | https://www.frontiersin.org/articles/10.3389/frsen.2022.832753/full |
| work_keys_str_mv | AT seonghyeoklee comparisonsofmultiresolutionbasedaitrainingdataandalgorithmsusingremotesensingfocusonlandcover AT moungjinlee comparisonsofmultiresolutionbasedaitrainingdataandalgorithmsusingremotesensingfocusonlandcover |