No Cover Image

Journal article 281 views 29 downloads

Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning

Bo Zhang Orcid Logo, Li Zhang Orcid Logo, Yong Pang Orcid Logo, Peter North Orcid Logo, Min Yan Orcid Logo, Hongge Ren, Linlin Ruan Orcid Logo, Zhenyu Yang, Bowei Chen Orcid Logo

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Volume: 17, Pages: 1 - 13

Swansea University Author: Peter North Orcid Logo

  • 64010_VOR.pdf

    PDF | Version of Record

    2023 The Authors. This work is licensed under a Creative Commons Attribution NonCommercial-NoDerivatives 4.0 License

    Download (4.14MB)

Check full text

DOI (Published version): 10.1109/jstars.2023.3290680

Abstract

NASA's (National Aeronautics and Space Administration) ICESat-2 with a Photon Counting LiDAR (Light Detection And Ranging) Sensor was successfully launched in September 2018. The sensor uses an advanced detection system called the Advanced Topographic Laser Altimeter System (ATLAS). The ATLAS s...

Full description

Published in: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
ISSN: 1939-1404 2151-1535
Published: Institute of Electrical and Electronics Engineers (IEEE) 2024
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa64010
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2023-08-01T10:47:48Z
last_indexed 2023-08-01T10:47:48Z
id cronfa64010
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>64010</id><entry>2023-08-01</entry><title>Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning</title><swanseaauthors><author><sid>fc45a0cb36c24d6cf35313a8c808652f</sid><ORCID>0000-0001-9933-6935</ORCID><firstname>Peter</firstname><surname>North</surname><name>Peter North</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2023-08-01</date><deptcode>SGE</deptcode><abstract>NASA's (National Aeronautics and Space Administration) ICESat-2 with a Photon Counting LiDAR (Light Detection And Ranging) Sensor was successfully launched in September 2018. The sensor uses an advanced detection system called the Advanced Topographic Laser Altimeter System (ATLAS). The ATLAS sensor detects signal photons at high speed and is highly sensitive. However, the sensor also extracts a large amount of background photon noise coming from the atmosphere, ground, sun, or other radiation. This condition is particularly evident in forest areas. Therefore, the method of filtering noise is of great significance for any use of the data. Without human intervention, automatic machine learning can form a set of processes needed for classification, namely feature selection, model selection, and model evaluation. This method offers convenient calculation, transferability, applicability, and interpretability. This study proposes an automatic machine learning approach to utilize data for forestry applications to improve data availability compared to NASA's official product. We used only 10% of the sample points for training on five datasets in the forest region and compared the performance of the classifiers. First, we conclude that the integrated learning performance generally outperforms single models, and the mean F1 score of all tests is approximately 0.9. The mean F1 score of the Stacked Ensembles model is 0.957 ahead of the other models. The top three variables used in training models are kNNdist5, kNNdist10, and h. These three variables could explain 51.6% of the components of the models. Over the regions tested, the proposed method could improve the proportion of signals correctly identified by 6.4%, 12.2%, 2.7%, 9.3%, and 1.4% in five datasets. The model performs better in low signal-to-noise (SNR) datasets less than 7.5. Then, compared to distinguishing noise photons, the optimal classifiers did better classifying signal photons from noise. The classifiers could correct misclassified labels in ATL08 products and show good stability in different conditions. A new method for the separation of forest signal from noise has been demonstrated, which uses only a very limited number of sample points for training, ensuring operational efficiency and training accuracy. The method would be largely unaffected by differences in topography, noise distribution, and SNR. Moreover, the classifiers demonstrated the ability to correctly identify signals considered noise photons in the ICESat-2 Level-3A ATL08 product. Overall, our method gave improved performance to the official ATL08 product over the regions tested, and can further improve data availability.</abstract><type>Journal Article</type><journal>IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing</journal><volume>17</volume><journalNumber/><paginationStart>1</paginationStart><paginationEnd>13</paginationEnd><publisher>Institute of Electrical and Electronics Engineers (IEEE)</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1939-1404</issnPrint><issnElectronic>2151-1535</issnElectronic><keywords>ICESat-2/ATLAS, Space-borne LiDAR, Automated Machine Learning, Photon Point Cloud Filtering</keywords><publishedDay>1</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-01-01</publishedDate><doi>10.1109/jstars.2023.3290680</doi><url/><notes/><college>COLLEGE NANME</college><department>Geography</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SGE</DepartmentCode><institution>Swansea University</institution><apcterm/><funders>National Key Research and Development Program of China (Grant Number: 2020YFE0200800). National Natural Science Foundation of China (Grant Number: 42001361).</funders><projectreference/><lastEdited>2024-04-08T20:45:02.5515736</lastEdited><Created>2023-08-01T11:35:16.3676580</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Biosciences, Geography and Physics - Geography</level></path><authors><author><firstname>Bo</firstname><surname>Zhang</surname><orcid>0000-0002-7226-1088</orcid><order>1</order></author><author><firstname>Li</firstname><surname>Zhang</surname><orcid>0000-0002-5880-7507</orcid><order>2</order></author><author><firstname>Yong</firstname><surname>Pang</surname><orcid>0000-0002-9760-6580</orcid><order>3</order></author><author><firstname>Peter</firstname><surname>North</surname><orcid>0000-0001-9933-6935</orcid><order>4</order></author><author><firstname>Min</firstname><surname>Yan</surname><orcid>0000-0001-7234-1590</orcid><order>5</order></author><author><firstname>Hongge</firstname><surname>Ren</surname><order>6</order></author><author><firstname>Linlin</firstname><surname>Ruan</surname><orcid>0009-0003-4602-5326</orcid><order>7</order></author><author><firstname>Zhenyu</firstname><surname>Yang</surname><order>8</order></author><author><firstname>Bowei</firstname><surname>Chen</surname><orcid>0000-0002-6377-1094</orcid><order>9</order></author></authors><documents><document><filename>64010__29244__9bb2b60d7e014a8cb8aa7057f5bdc2b5.pdf</filename><originalFilename>64010_VOR.pdf</originalFilename><uploaded>2023-12-12T09:02:43.9754462</uploaded><type>Output</type><contentLength>4342247</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>2023 The Authors. This work is licensed under a Creative Commons Attribution NonCommercial-NoDerivatives 4.0 License</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling v2 64010 2023-08-01 Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning fc45a0cb36c24d6cf35313a8c808652f 0000-0001-9933-6935 Peter North Peter North true false 2023-08-01 SGE NASA's (National Aeronautics and Space Administration) ICESat-2 with a Photon Counting LiDAR (Light Detection And Ranging) Sensor was successfully launched in September 2018. The sensor uses an advanced detection system called the Advanced Topographic Laser Altimeter System (ATLAS). The ATLAS sensor detects signal photons at high speed and is highly sensitive. However, the sensor also extracts a large amount of background photon noise coming from the atmosphere, ground, sun, or other radiation. This condition is particularly evident in forest areas. Therefore, the method of filtering noise is of great significance for any use of the data. Without human intervention, automatic machine learning can form a set of processes needed for classification, namely feature selection, model selection, and model evaluation. This method offers convenient calculation, transferability, applicability, and interpretability. This study proposes an automatic machine learning approach to utilize data for forestry applications to improve data availability compared to NASA's official product. We used only 10% of the sample points for training on five datasets in the forest region and compared the performance of the classifiers. First, we conclude that the integrated learning performance generally outperforms single models, and the mean F1 score of all tests is approximately 0.9. The mean F1 score of the Stacked Ensembles model is 0.957 ahead of the other models. The top three variables used in training models are kNNdist5, kNNdist10, and h. These three variables could explain 51.6% of the components of the models. Over the regions tested, the proposed method could improve the proportion of signals correctly identified by 6.4%, 12.2%, 2.7%, 9.3%, and 1.4% in five datasets. The model performs better in low signal-to-noise (SNR) datasets less than 7.5. Then, compared to distinguishing noise photons, the optimal classifiers did better classifying signal photons from noise. The classifiers could correct misclassified labels in ATL08 products and show good stability in different conditions. A new method for the separation of forest signal from noise has been demonstrated, which uses only a very limited number of sample points for training, ensuring operational efficiency and training accuracy. The method would be largely unaffected by differences in topography, noise distribution, and SNR. Moreover, the classifiers demonstrated the ability to correctly identify signals considered noise photons in the ICESat-2 Level-3A ATL08 product. Overall, our method gave improved performance to the official ATL08 product over the regions tested, and can further improve data availability. Journal Article IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 17 1 13 Institute of Electrical and Electronics Engineers (IEEE) 1939-1404 2151-1535 ICESat-2/ATLAS, Space-borne LiDAR, Automated Machine Learning, Photon Point Cloud Filtering 1 1 2024 2024-01-01 10.1109/jstars.2023.3290680 COLLEGE NANME Geography COLLEGE CODE SGE Swansea University National Key Research and Development Program of China (Grant Number: 2020YFE0200800). National Natural Science Foundation of China (Grant Number: 42001361). 2024-04-08T20:45:02.5515736 2023-08-01T11:35:16.3676580 Faculty of Science and Engineering School of Biosciences, Geography and Physics - Geography Bo Zhang 0000-0002-7226-1088 1 Li Zhang 0000-0002-5880-7507 2 Yong Pang 0000-0002-9760-6580 3 Peter North 0000-0001-9933-6935 4 Min Yan 0000-0001-7234-1590 5 Hongge Ren 6 Linlin Ruan 0009-0003-4602-5326 7 Zhenyu Yang 8 Bowei Chen 0000-0002-6377-1094 9 64010__29244__9bb2b60d7e014a8cb8aa7057f5bdc2b5.pdf 64010_VOR.pdf 2023-12-12T09:02:43.9754462 Output 4342247 application/pdf Version of Record true 2023 The Authors. This work is licensed under a Creative Commons Attribution NonCommercial-NoDerivatives 4.0 License true eng http://creativecommons.org/licenses/by/4.0/
title Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning
spellingShingle Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning
Peter North
title_short Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning
title_full Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning
title_fullStr Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning
title_full_unstemmed Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning
title_sort Improved Forest Signal Detection for Space-Borne Photon-Counting LiDAR Using Automatic Machine Learning
author_id_str_mv fc45a0cb36c24d6cf35313a8c808652f
author_id_fullname_str_mv fc45a0cb36c24d6cf35313a8c808652f_***_Peter North
author Peter North
author2 Bo Zhang
Li Zhang
Yong Pang
Peter North
Min Yan
Hongge Ren
Linlin Ruan
Zhenyu Yang
Bowei Chen
format Journal article
container_title IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
container_volume 17
container_start_page 1
publishDate 2024
institution Swansea University
issn 1939-1404
2151-1535
doi_str_mv 10.1109/jstars.2023.3290680
publisher Institute of Electrical and Electronics Engineers (IEEE)
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Biosciences, Geography and Physics - Geography{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Biosciences, Geography and Physics - Geography
document_store_str 1
active_str 0
description NASA's (National Aeronautics and Space Administration) ICESat-2 with a Photon Counting LiDAR (Light Detection And Ranging) Sensor was successfully launched in September 2018. The sensor uses an advanced detection system called the Advanced Topographic Laser Altimeter System (ATLAS). The ATLAS sensor detects signal photons at high speed and is highly sensitive. However, the sensor also extracts a large amount of background photon noise coming from the atmosphere, ground, sun, or other radiation. This condition is particularly evident in forest areas. Therefore, the method of filtering noise is of great significance for any use of the data. Without human intervention, automatic machine learning can form a set of processes needed for classification, namely feature selection, model selection, and model evaluation. This method offers convenient calculation, transferability, applicability, and interpretability. This study proposes an automatic machine learning approach to utilize data for forestry applications to improve data availability compared to NASA's official product. We used only 10% of the sample points for training on five datasets in the forest region and compared the performance of the classifiers. First, we conclude that the integrated learning performance generally outperforms single models, and the mean F1 score of all tests is approximately 0.9. The mean F1 score of the Stacked Ensembles model is 0.957 ahead of the other models. The top three variables used in training models are kNNdist5, kNNdist10, and h. These three variables could explain 51.6% of the components of the models. Over the regions tested, the proposed method could improve the proportion of signals correctly identified by 6.4%, 12.2%, 2.7%, 9.3%, and 1.4% in five datasets. The model performs better in low signal-to-noise (SNR) datasets less than 7.5. Then, compared to distinguishing noise photons, the optimal classifiers did better classifying signal photons from noise. The classifiers could correct misclassified labels in ATL08 products and show good stability in different conditions. A new method for the separation of forest signal from noise has been demonstrated, which uses only a very limited number of sample points for training, ensuring operational efficiency and training accuracy. The method would be largely unaffected by differences in topography, noise distribution, and SNR. Moreover, the classifiers demonstrated the ability to correctly identify signals considered noise photons in the ICESat-2 Level-3A ATL08 product. Overall, our method gave improved performance to the official ATL08 product over the regions tested, and can further improve data availability.
published_date 2024-01-01T20:44:57Z
_version_ 1795797021868687360
score 11.012678

Similar Items