Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra
Soil spectroscopy has experienced a tremendous increase in soil property characterisation, and can be used not only in the laboratory but also from the space (imaging spectroscopy). Partial least squares (PLS) regression is one of the most common approaches for the calibration of soil properties usi...
Main Authors: | , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden
2018
|
Subjects: | |
Online Access: | http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-232271 http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-232271 http://www.qucosa.de/fileadmin/data/qucosa/documents/23227/remotesensing-09-01299.pdf |
id |
ndltd-DRESDEN-oai-qucosa.de-bsz-14-qucosa-232271 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-DRESDEN-oai-qucosa.de-bsz-14-qucosa-2322712018-06-07T03:26:53Z Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra Liu, Lanfa Ji, Min Buchroithner, Manfred F. PLS Gradienten-verstärkte Entscheidungsbäume Bodenspektroskopie LUCAS TU Dresden Publikationsfond PLS gradient-boosted decision trees soil spectroscopy LUCAS TU Dresden Publishing Fund ddc:620 rvk:ZI 0001 Soil spectroscopy has experienced a tremendous increase in soil property characterisation, and can be used not only in the laboratory but also from the space (imaging spectroscopy). Partial least squares (PLS) regression is one of the most common approaches for the calibration of soil properties using soil spectra. Besides functioning as a calibration method, PLS can also be used as a dimension reduction tool, which has scarcely been studied in soil spectroscopy. PLS components retained from high-dimensional spectral data can further be explored with the gradient-boosted decision tree (GBDT) method. Three soil sample categories were extracted from the Land Use/Land Cover Area Frame Survey (LUCAS) soil library according to the type of land cover (woodland, grassland, and cropland). First, PLS regression and GBDT were separately applied to build the spectroscopic models for soil organic carbon (OC), total nitrogen content (N), and clay for each soil category. Then, PLS-derived components were used as input variables for the GBDT model. The results demonstrate that the combined PLS-GBDT approach has better performance than PLS or GBDT alone. The relative important variables for soil property estimation revealed by the proposed method demonstrated that the PLS method is a useful dimension reduction tool for soil spectra to retain target-related information. Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden Molecular Diversity Preservation International (MDPI), 2018-06-06 doc-type:article application/pdf http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-232271 urn:nbn:de:bsz:14-qucosa-232271 http://www.qucosa.de/fileadmin/data/qucosa/documents/23227/remotesensing-09-01299.pdf Remote sensing 2017, 9(12), ISSN: 2072-4292. DOI: 10.3390/rs9121299 eng |
collection |
NDLTD |
language |
English |
format |
Article |
sources |
NDLTD |
topic |
PLS Gradienten-verstärkte Entscheidungsbäume Bodenspektroskopie LUCAS TU Dresden Publikationsfond PLS gradient-boosted decision trees soil spectroscopy LUCAS TU Dresden Publishing Fund ddc:620 rvk:ZI 0001 |
spellingShingle |
PLS Gradienten-verstärkte Entscheidungsbäume Bodenspektroskopie LUCAS TU Dresden Publikationsfond PLS gradient-boosted decision trees soil spectroscopy LUCAS TU Dresden Publishing Fund ddc:620 rvk:ZI 0001 Liu, Lanfa Ji, Min Buchroithner, Manfred F. Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra |
description |
Soil spectroscopy has experienced a tremendous increase in soil property characterisation, and can be used not only in the laboratory but also from the space (imaging spectroscopy). Partial least squares (PLS) regression is one of the most common approaches for the calibration of soil properties using soil spectra. Besides functioning as a calibration method, PLS can also be used as a dimension reduction tool, which has scarcely been studied in soil spectroscopy. PLS components retained from high-dimensional spectral data can further be explored with the gradient-boosted decision tree (GBDT) method. Three soil sample categories were extracted from the Land Use/Land Cover Area Frame Survey (LUCAS) soil library according to the type of land cover (woodland, grassland, and cropland). First, PLS regression and GBDT were separately applied to build the spectroscopic models for soil organic carbon (OC), total nitrogen content (N), and clay for each soil category. Then, PLS-derived components were used as input variables for the GBDT model. The results demonstrate that the combined PLS-GBDT approach has better performance than PLS or GBDT alone. The relative important variables for soil property estimation revealed by the proposed method demonstrated that the PLS method is a useful dimension reduction tool for soil spectra to retain target-related information. |
author2 |
Molecular Diversity Preservation International (MDPI), |
author_facet |
Molecular Diversity Preservation International (MDPI), Liu, Lanfa Ji, Min Buchroithner, Manfred F. |
author |
Liu, Lanfa Ji, Min Buchroithner, Manfred F. |
author_sort |
Liu, Lanfa |
title |
Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra |
title_short |
Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra |
title_full |
Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra |
title_fullStr |
Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra |
title_full_unstemmed |
Combining Partial Least Squares and the Gradient-Boosting Method for Soil Property Retrieval Using Visible Near-Infrared Shortwave Infrared Spectra |
title_sort |
combining partial least squares and the gradient-boosting method for soil property retrieval using visible near-infrared shortwave infrared spectra |
publisher |
Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden |
publishDate |
2018 |
url |
http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-232271 http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-232271 http://www.qucosa.de/fileadmin/data/qucosa/documents/23227/remotesensing-09-01299.pdf |
work_keys_str_mv |
AT liulanfa combiningpartialleastsquaresandthegradientboostingmethodforsoilpropertyretrievalusingvisiblenearinfraredshortwaveinfraredspectra AT jimin combiningpartialleastsquaresandthegradientboostingmethodforsoilpropertyretrievalusingvisiblenearinfraredshortwaveinfraredspectra AT buchroithnermanfredf combiningpartialleastsquaresandthegradientboostingmethodforsoilpropertyretrievalusingvisiblenearinfraredshortwaveinfraredspectra |
_version_ |
1718692373464088576 |