| |
|
|
| Registros recuperados : 3 | |
| 1. |  | BÉGUÉ, A.; ARVOR, D.; BELLON, B.; BETBEDER, J.; ABELLEYRA, D. de; FERRAZ, R. P. D.; LEBOURGEOIS, V.; LELONG, C.; SIMÕES, M.; VERÓN, S. R. Remote sensing and cropping practices: a review. Remote Sensing, v. 10, n. 1, Jan. 2018.| Biblioteca(s): Embrapa Solos. |
|    |
| 2. | | WALDNER, F.; BELLEMANS, N.; HOCHMAN, Z.; NEWBY, T.; ABELLEYRA, D. de; VERÓN, S. R.; BARTALEV, S.; LAVRENIUK, M.; KUSSUL, N.; LE MAIRE, G.; SIMÕES, M.; SKAKUN, S.; DEFOURNY, P. Roadside collection of training data for cropland mapping is viable when environmental and management gradients are surveyed. International Journal of Applied Earth Observation and Geoinformation, v. 80, p. 82-93, Aug. 2019.| Biblioteca(s): Embrapa Solos. |
|  |
| 3. | | WALDNER, F.; SCHUCKNECHT, A.; LESIV, M.; GALLEGO, J.; SEE, L.; PÉREZ-HOYOS, A.; D'ANDRIMONT, R.; DE MAET, T.; LASO BAYAS, J. C.; FRITZ, S.; LEO, O.; KERDILES, H.; DÍEZ, M.; VAN TRICHT, K.; GILLIAMS, S.; SHELESTOV, A.; LAVRENIUK, M.; SIMÕES, M.; FERRAZ, R. P. D.; BELLÓN, B.; BÉGUÉ, A.; HAZEU, G.; STONACEK, V.; KOLOMAZNIK, J.; MISUREC, J.; VERÓN, S. R.; ABELLEYRA, D. de; PLOTNIKOV, D.; MINGYONG, L.; SINGHA, M.; PATIL, P.; ZHANG, M.; DEFOURNY, P. Conflation of expert and crowd reference data to validate global binary thematic maps. Remote Sensing of Environment, v. 221, p. 235-246, Feb. 2019.| Biblioteca(s): Embrapa Solos. |
| |
| Registros recuperados : 3 | |
|
|
|
 | Acesso ao texto completo restrito à biblioteca da Embrapa Solos. Para informações adicionais entre em contato com cnps.biblioteca@embrapa.br. |
Registro Completo
|
Biblioteca(s): |
Embrapa Solos. |
|
Data corrente: |
18/04/2019 |
|
Data da última atualização: |
08/11/2019 |
|
Tipo da produção científica: |
Artigo em Periódico Indexado |
|
Circulação/Nível: |
A - 1 |
|
Autoria: |
WALDNER, F.; BELLEMANS, N.; HOCHMAN, Z.; NEWBY, T.; ABELLEYRA, D. de; VERÓN, S. R.; BARTALEV, S.; LAVRENIUK, M.; KUSSUL, N.; LE MAIRE, G.; SIMÕES, M.; SKAKUN, S.; DEFOURNY, P. |
|
Afiliação: |
FRANÇOIS WALDNER, CSIRO AGRICULTURE & FOOD, AUSTRALIA; NICOLAS BELLEMANS, UNIVERSITE CATHOLIQUE DE LOUVAIN, BELGIUM; ZVI HOCHMAN, CSIRO AGRICULTURE & FOOD, AUSTRALIA; TERENCE NEWBY, AGRICULTURAL RESEARCH COUNCIL, SOUTH AFRICA; DIEGO DE ABELLEYRA, INSTITUTO DE CLIMA Y AGUA, INSTITUTO NACIONAL DE TECNOLOGIA AGROPECUARIA, ARGENTINA; SANTIAGO R. VERÓN, INSTITUTO DE CLIMA Y AGUA, INSTITUTO NACIONAL DE TECNOLOGÍA AGROPECUARIA, ARGENTINA; SERGEY BARTALEV, SPACE RESEARCH INSTITUTE OF RUSSIAN ACADEMY OF SCIENCES, RUSSIA; MYKOLA LAVRENIUK, SPACE RESEARCH INSTITUTE NAS AND SSA, UKRAINE; NATALIIA KUSSUL, SPACE RESEARCH INSTITUTE NAS AND SSA, UKRAINE; GUERRIC LE MAIRE, CIRAD / ECO &SOLS / UNIV MONTPELLIER, FRANCE; MARGARETH GONCALVES SIMOES, CNPS; SERGII SKAKUN, DEPARTMENT OF GEOGRAPHICAL SCIENCES, UNIVERSITY OF MARYLAND, USA; PIERRE DEFOURNY, UNIVERSITÉ CATHOLIQUE DE LOUVAIN, BELGIUM. |
|
Título: |
Roadside collection of training data for cropland mapping is viable when environmental and management gradients are surveyed. |
|
Ano de publicação: |
2019 |
|
Fonte/Imprenta: |
International Journal of Applied Earth Observation and Geoinformation, v. 80, p. 82-93, Aug. 2019. |
|
DOI: |
10.1016/j.jag.2019.01.002 |
|
Idioma: |
Inglês |
|
Conteúdo: |
Cropland maps derived from satellite imagery have become a common source of information to estimate food production, support land use policies, and measure the environmental impacts of agriculture. Cropland classification models are typically calibrated with data collected from roadside surveys which enable the sampling of large areas at a relatively low cost. However, there is a risk of providing biased data as environmental and management gradients may not be fully captured from road networks, thereby violating the assumption of representativeness of calibration data. Despite being widely adopted, the potential biases of roadside sampling have so far not been thoroughly addressed. In this study, we looked for evidence of these biases by comparing three sampling strategies: Random sampling, Roadside sampling, and Transect sampling - a spatially constrained variant of Roadside sampling. In these three strategies, non-cropland data are randomly distributed as they can be photo-interpreted. Based on reference maps at 30 m in four study sites, we followed a Monte Carlo approach to generate multiple realizations of each sampling strategy for ten sample sizes. The effect of the sampling strategy was then assessed in terms of representativeness of the data set collected and accuracy of the resulting maps. Results showed that data sets obtained from Roadside sampling were significantly less representative than those obtained from Random sampling but the resulting maps were only marginally less accurate (2% difference). Transect sampling captured systematically less variability than Random or Roadside sampling which led to differences in accuracy as large as 15%. The effect of sample size on accuracy varied across sites but generally leveled off after reaching 3000 pixels. Augmenting the size of Transect samples improved the classification accuracy but not sufficiently to match the performance of the other sampling strategies. Finally, we found that Random and Roadside training sets with similar representativeness yield comparable accuracy. Therefore, we conclude that roadside sampling can be a viable source of training data for cropland mapping if the range of environmental and management gradients is surveyed. This underlines the importance of survey planning to identify those routes that capture most variability. MenosCropland maps derived from satellite imagery have become a common source of information to estimate food production, support land use policies, and measure the environmental impacts of agriculture. Cropland classification models are typically calibrated with data collected from roadside surveys which enable the sampling of large areas at a relatively low cost. However, there is a risk of providing biased data as environmental and management gradients may not be fully captured from road networks, thereby violating the assumption of representativeness of calibration data. Despite being widely adopted, the potential biases of roadside sampling have so far not been thoroughly addressed. In this study, we looked for evidence of these biases by comparing three sampling strategies: Random sampling, Roadside sampling, and Transect sampling - a spatially constrained variant of Roadside sampling. In these three strategies, non-cropland data are randomly distributed as they can be photo-interpreted. Based on reference maps at 30 m in four study sites, we followed a Monte Carlo approach to generate multiple realizations of each sampling strategy for ten sample sizes. The effect of the sampling strategy was then assessed in terms of representativeness of the data set collected and accuracy of the resulting maps. Results showed that data sets obtained from Roadside sampling were significantly less representative than those obtained from Random sampling but the resulting maps were only mar... Mostrar Tudo |
|
Thesagro: |
Agricultura; Agricultura de Precisão; Amostragem; Tamanho. |
|
Thesaurus NAL: |
Accuracy; Agriculture; Classification; Samplers; Sampling. |
|
Categoria do assunto: |
P Recursos Naturais, Ciências Ambientais e da Terra |
|
Marc: |
LEADER 03464naa a2200385 a 4500
001 2108335
005 2019-11-08
008 2019 bl uuuu u00u1 u #d
024 7 $a10.1016/j.jag.2019.01.002$2DOI
100 1 $aWALDNER, F.
245 $aRoadside collection of training data for cropland mapping is viable when environmental and management gradients are surveyed.$h[electronic resource]
260 $c2019
520 $aCropland maps derived from satellite imagery have become a common source of information to estimate food production, support land use policies, and measure the environmental impacts of agriculture. Cropland classification models are typically calibrated with data collected from roadside surveys which enable the sampling of large areas at a relatively low cost. However, there is a risk of providing biased data as environmental and management gradients may not be fully captured from road networks, thereby violating the assumption of representativeness of calibration data. Despite being widely adopted, the potential biases of roadside sampling have so far not been thoroughly addressed. In this study, we looked for evidence of these biases by comparing three sampling strategies: Random sampling, Roadside sampling, and Transect sampling - a spatially constrained variant of Roadside sampling. In these three strategies, non-cropland data are randomly distributed as they can be photo-interpreted. Based on reference maps at 30 m in four study sites, we followed a Monte Carlo approach to generate multiple realizations of each sampling strategy for ten sample sizes. The effect of the sampling strategy was then assessed in terms of representativeness of the data set collected and accuracy of the resulting maps. Results showed that data sets obtained from Roadside sampling were significantly less representative than those obtained from Random sampling but the resulting maps were only marginally less accurate (2% difference). Transect sampling captured systematically less variability than Random or Roadside sampling which led to differences in accuracy as large as 15%. The effect of sample size on accuracy varied across sites but generally leveled off after reaching 3000 pixels. Augmenting the size of Transect samples improved the classification accuracy but not sufficiently to match the performance of the other sampling strategies. Finally, we found that Random and Roadside training sets with similar representativeness yield comparable accuracy. Therefore, we conclude that roadside sampling can be a viable source of training data for cropland mapping if the range of environmental and management gradients is surveyed. This underlines the importance of survey planning to identify those routes that capture most variability.
650 $aAccuracy
650 $aAgriculture
650 $aClassification
650 $aSamplers
650 $aSampling
650 $aAgricultura
650 $aAgricultura de Precisão
650 $aAmostragem
650 $aTamanho
700 1 $aBELLEMANS, N.
700 1 $aHOCHMAN, Z.
700 1 $aNEWBY, T.
700 1 $aABELLEYRA, D. de
700 1 $aVERÓN, S. R.
700 1 $aBARTALEV, S.
700 1 $aLAVRENIUK, M.
700 1 $aKUSSUL, N.
700 1 $aLE MAIRE, G.
700 1 $aSIMÕES, M.
700 1 $aSKAKUN, S.
700 1 $aDEFOURNY, P.
773 $tInternational Journal of Applied Earth Observation and Geoinformation$gv. 80, p. 82-93, Aug. 2019.
Download
Esconder MarcMostrar Marc Completo |
|
Registro original: |
Embrapa Solos (CNPS) |
|
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
Fechar
|
| Nenhum registro encontrado para a expressão de busca informada. |
|
|
