| |
|
|
| Registros recuperados : 1 | |
| 1. |  | SALAZAR, D. U.; DEMATTÊ, J. A. M.; VICENTE, L. E.; GUIMARÃES, C. C.; SOUZA, A. B. de; INFORSATO, L.; CARVALHO, H. W. L. de. Spectral range (FRX-VIS-NIR-SWIR-MIR) interaction on the organic matter prediction. In: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science: beyond food and fuel: abstracts. Viçosa, MG: SBCS, 2018. Trabalho 392.| Biblioteca(s): Embrapa Meio Ambiente. |
|    |
| Registros recuperados : 1 | |
|
|
|
Registro Completo
|
Biblioteca(s): |
Embrapa Meio Ambiente. |
|
Data corrente: |
10/01/2019 |
|
Data da última atualização: |
30/01/2019 |
|
Tipo da produção científica: |
Resumo em Anais de Congresso |
|
Autoria: |
SALAZAR, D. U.; DEMATTÊ, J. A. M.; VICENTE, L. E.; GUIMARÃES, C. C.; SOUZA, A. B. de; INFORSATO, L.; CARVALHO, H. W. L. de. |
|
Afiliação: |
Diego Fernando Urbina Salazar, Department of Soil Science/ESALQ/USP; José A. M. Demattê, Department of Soil Science/ESALQ/USP; LUIZ EDUARDO VICENTE, CNPMA; Clecia C. Guimarães, Department of Soil Science/ESALQ/USP; Arnaldo Barros de Souza, Department of Soil Science/ESALQ/USP; Leonardo Inforsato, CENA/USP; Hudson Wallace Pereira de Carvalho, CENA/USP. |
|
Título: |
Spectral range (FRX-VIS-NIR-SWIR-MIR) interaction on the organic matter prediction. |
|
Ano de publicação: |
2018 |
|
Fonte/Imprenta: |
In: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science: beyond food and fuel: abstracts. Viçosa, MG: SBCS, 2018. Trabalho 392. |
|
Idioma: |
Português |
|
Conteúdo: |
Organic matter (OM) is an important indicator of soil quality and, therefore, must be quantified quickly and efficiently. Besides traditional methods, and countless studies with sensors have been advanced, but few integratively. This work aimed to develop models for OM content prediction using the FRX-VIS-NIR-SWIR-MIR regions separately and together. A total of 22 soil samples were collected in the state of São Paulo, Brazil (80-100 cm). Each sample was oven dried at 45 ° C for 48 hours, milled and sieved to a diameter of 0.150 mm and had a series of 8 treatments, considering the addition of humified organic material (MOH) in different amounts (0, 5, 10, 15, 20, 30, 40, 50 g). Afterwards, the treated samples were read in the sensors FIELDSPEC-PRO (VISNIRSWIR), ALPHA FTIR (MIR), and portable FRX for the extraction of the spectral responses. For the creation of the best model by Partial least squares regression (PLSR), 70% of the samples were used for calibration and 30% for validation and different preprocessing techniques were used: transformation of the reflectance data for absorbance, smoothing, first derivative Savitzky-Golay (SGD), standard normal variation (SNV) and multiplicative signal correction (MSC). The spectral data were tested separately and together using the absorbance transformation as a fixed preprocessing to which other was added. The calibration results for all models presented R² greater than 0.8 for the best preprocessing, which varies according to the spectral region analyzed. The best validation occurred for the preprocessed model with only absorbance and smoothing using the VIS-NIR-SIWIR-MIR spectral regions and presented values of R², RMSE and RPIQ equal to 0.8, 6.8 and 4.55, respectively. When using all the regions (RX-VIS-NIR-SWIR-MIR) simultaneously, validation values similar to those cited above (R² = 0.8, RMSE = 6.96 and RPIQ = 4.45)were obtained, indicating that the X-ray region does not influence the improvement of the prediction of OM. Models created only with a the spectral x-ray region presented the worst validation values. OM considerably influences on the spectral response from visible to medium infrared regions and, in this manner, they can be used for more accurate prediction of its values. MenosOrganic matter (OM) is an important indicator of soil quality and, therefore, must be quantified quickly and efficiently. Besides traditional methods, and countless studies with sensors have been advanced, but few integratively. This work aimed to develop models for OM content prediction using the FRX-VIS-NIR-SWIR-MIR regions separately and together. A total of 22 soil samples were collected in the state of São Paulo, Brazil (80-100 cm). Each sample was oven dried at 45 ° C for 48 hours, milled and sieved to a diameter of 0.150 mm and had a series of 8 treatments, considering the addition of humified organic material (MOH) in different amounts (0, 5, 10, 15, 20, 30, 40, 50 g). Afterwards, the treated samples were read in the sensors FIELDSPEC-PRO (VISNIRSWIR), ALPHA FTIR (MIR), and portable FRX for the extraction of the spectral responses. For the creation of the best model by Partial least squares regression (PLSR), 70% of the samples were used for calibration and 30% for validation and different preprocessing techniques were used: transformation of the reflectance data for absorbance, smoothing, first derivative Savitzky-Golay (SGD), standard normal variation (SNV) and multiplicative signal correction (MSC). The spectral data were tested separately and together using the absorbance transformation as a fixed preprocessing to which other was added. The calibration results for all models presented R² greater than 0.8 for the best preprocessing, which varies according to the s... Mostrar Tudo |
|
Palavras-Chave: |
Soil sensing; Spectral library. |
|
Thesaurus NAL: |
Environment; Organic matter; Spectroscopy. |
|
Categoria do assunto: |
X Pesquisa, Tecnologia e Engenharia |
|
URL: |
https://ainfo.cnptia.embrapa.br/digital/bitstream/item/190317/1/RA-VicenteLE-21WCSS-2018-Trabalho-392.pdf
|
|
Marc: |
LEADER 03087nam a2200241 a 4500
001 2103530
005 2019-01-30
008 2018 bl uuuu u00u1 u #d
100 1 $aSALAZAR, D. U.
245 $aSpectral range (FRX-VIS-NIR-SWIR-MIR) interaction on the organic matter prediction.$h[electronic resource]
260 $aIn: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science: beyond food and fuel: abstracts. Viçosa, MG: SBCS, 2018. Trabalho 392.$c2018
520 $aOrganic matter (OM) is an important indicator of soil quality and, therefore, must be quantified quickly and efficiently. Besides traditional methods, and countless studies with sensors have been advanced, but few integratively. This work aimed to develop models for OM content prediction using the FRX-VIS-NIR-SWIR-MIR regions separately and together. A total of 22 soil samples were collected in the state of São Paulo, Brazil (80-100 cm). Each sample was oven dried at 45 ° C for 48 hours, milled and sieved to a diameter of 0.150 mm and had a series of 8 treatments, considering the addition of humified organic material (MOH) in different amounts (0, 5, 10, 15, 20, 30, 40, 50 g). Afterwards, the treated samples were read in the sensors FIELDSPEC-PRO (VISNIRSWIR), ALPHA FTIR (MIR), and portable FRX for the extraction of the spectral responses. For the creation of the best model by Partial least squares regression (PLSR), 70% of the samples were used for calibration and 30% for validation and different preprocessing techniques were used: transformation of the reflectance data for absorbance, smoothing, first derivative Savitzky-Golay (SGD), standard normal variation (SNV) and multiplicative signal correction (MSC). The spectral data were tested separately and together using the absorbance transformation as a fixed preprocessing to which other was added. The calibration results for all models presented R² greater than 0.8 for the best preprocessing, which varies according to the spectral region analyzed. The best validation occurred for the preprocessed model with only absorbance and smoothing using the VIS-NIR-SIWIR-MIR spectral regions and presented values of R², RMSE and RPIQ equal to 0.8, 6.8 and 4.55, respectively. When using all the regions (RX-VIS-NIR-SWIR-MIR) simultaneously, validation values similar to those cited above (R² = 0.8, RMSE = 6.96 and RPIQ = 4.45)were obtained, indicating that the X-ray region does not influence the improvement of the prediction of OM. Models created only with a the spectral x-ray region presented the worst validation values. OM considerably influences on the spectral response from visible to medium infrared regions and, in this manner, they can be used for more accurate prediction of its values.
650 $aEnvironment
650 $aOrganic matter
650 $aSpectroscopy
653 $aSoil sensing
653 $aSpectral library
700 1 $aDEMATTÊ, J. A. M.
700 1 $aVICENTE, L. E.
700 1 $aGUIMARÃES, C. C.
700 1 $aSOUZA, A. B. de
700 1 $aINFORSATO, L.
700 1 $aCARVALHO, H. W. L. de
Download
Esconder MarcMostrar Marc Completo |
|
Registro original: |
Embrapa Meio Ambiente (CNPMA) |
|
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
Fechar
|
| Nenhum registro encontrado para a expressão de busca informada. |
|
|
