|
|
Registro completo
|
Biblioteca (s) : |
INIA La Estanzuela. |
Fecha : |
29/10/2020 |
Actualizado : |
21/03/2023 |
Tipo de producción científica : |
Artículos en Revistas Indexadas Internacionales |
Autor : |
AHMAD, R.; YANG, B.; ETTLIN, G.; BERGER, A.; RODRÍGUEZ-BOCCA, P. |
Afiliación : |
REHAAN AHMAD, Cupertino High School, 10100 Finch Avenue, Cupertino, CA 95014, USA.; Cupertino High School, 10100 Finch Avenue, Cupertino, CA 95014, USA.; GUILLERMO ETTLIN, Facultad de Ingeniería, Instituto de Computación, Universidad de la República, Julio Herrera y Reissig 565, Montevideo 11300, Uruguay.; ANDRES GUSTAVO BERGER RICCA, INIA (Instituto Nacional de Investigación Agropecuaria), Uruguay; PABLO RODRÍGUEZ BOCCA, Facultad de Ingeniería, Instituto de Computación, Universidad de la República, Julio Herrera y Reissig 565, Montevideo 11300, Uruguay. |
Título : |
A machine-learning based ConvLSTM architecture for NDVI forecasting. |
Fecha de publicación : |
2023 |
Fuente / Imprenta : |
International Transactions in Operational Research, 2023, Volume 30, Issue 4, Pages 2025 - 2048. doi: https://doi.org/10.1111/itor.12887 |
ISSN : |
0969-6016 (print); 1475-3995 (electronic) |
DOI : |
10.1111/itor.12887 |
Idioma : |
Inglés |
Notas : |
Article history: Received 24 September 2019; Received in revised form 7 August 2020; Accepted 5 October 2020: First published 22 October 2020. -- Corresponding author: Rodríguez-Bocca, P.; Facultad de Ingeniería, Instituto de Computación, Universidad de la República, Julio Herrera y Reissig 565, Montevideo, Uruguay; email:prbocca@fing.edu.uy -- FUNDING: This research was partially supported by the "Comisión Sectorial de Investigación Científica (CSIC), UDELAR" and the "Programa de Desarrollo de las Ciencias Básicas (PEDECIBA)" of Uruguay. Some of the calculations reported in this paper were performed in ClusterUY, a newly installed platform for high-performance scientific computing at the National Supercomputing Center, Uruguay. -- Special Issue: OR and Big Data in Agriculture. |
Contenido : |
Abstract:Normalized difference vegetation index (NDVI) is an essential remote measurement for agricultural studies because of its strong correlation with crop growth and yield. Accurate and comprehensive NDVI forecasts thus provide effective future projections of crop yield for precise agricultural planning and budgeting. Previous recurrent neural network (RNN) based forecasting methodologies have only performed single-pixel or large-area-average NDVI predictions. We present an alternative RNN-based deep-learning architecture, the convolutional long short-term memory (ConvLSTM), to supply much more comprehensive and detailed NDVI forecasts. In this paper, a single ConvLSTM is capable of 10,000-pixel field-level NDVI predictions, providing a more practical methodology for agricultural producers than single-pixel studies. We compare our model to the parametric crop growth model (PCGM), another multipixel field-level NDVI forecasting technique. We test each model over the same set of soybean crop field pixels with the root mean square error (RMSE) metric. The training configuration of each model is defined by the number of seasons of historical data used for weight optimization. When the best training configuration of the model found is used, the ConvLSTM obtains an RMSE of 0.0782, outperforming the PCGM?s RMSE of 0.0989 (an improvement of 0.0207 in precision represents a large gain in the accuracy of production volume prediction when projected into large production areas). Finally, by comparing the ConvLSTM predictions with the ground truth data over the entire target region rather than just the soybean crop pixels, we discover that the ConvLSTM can also predict NDVI values over the nonsoybean crop as effectively. © 2020 The Authors. MenosAbstract:Normalized difference vegetation index (NDVI) is an essential remote measurement for agricultural studies because of its strong correlation with crop growth and yield. Accurate and comprehensive NDVI forecasts thus provide effective future projections of crop yield for precise agricultural planning and budgeting. Previous recurrent neural network (RNN) based forecasting methodologies have only performed single-pixel or large-area-average NDVI predictions. We present an alternative RNN-based deep-learning architecture, the convolutional long short-term memory (ConvLSTM), to supply much more comprehensive and detailed NDVI forecasts. In this paper, a single ConvLSTM is capable of 10,000-pixel field-level NDVI predictions, providing a more practical methodology for agricultural producers than single-pixel studies. We compare our model to the parametric crop growth model (PCGM), another multipixel field-level NDVI forecasting technique. We test each model over the same set of soybean crop field pixels with the root mean square error (RMSE) metric. The training configuration of each model is defined by the number of seasons of historical data used for weight optimization. When the best training configuration of the model found is used, the ConvLSTM obtains an RMSE of 0.0782, outperforming the PCGM?s RMSE of 0.0989 (an improvement of 0.0207 in precision represents a large gain in the accuracy of production volume prediction when projected into large production areas). Fin... Presentar Todo |
Palabras claves : |
CONVLSTM NEURAL NETWORKS; DEEP LEARNING; NORMALIZED DIFFERENCE VEGETATION INDEX; OPTIMIZATION; PREDICTIVE ANALYSIS. |
Asunto categoría : |
F01 Cultivo |
URL : |
https://onlinelibrary.wiley.com/doi/epdf/10.1111/itor.12887
|
Marc : |
LEADER 03430naa a2200265 a 4500 001 1061440 005 2023-03-21 008 2023 bl uuuu u00u1 u #d 022 $a0969-6016 (print); 1475-3995 (electronic) 024 7 $a10.1111/itor.12887$2DOI 100 1 $aAHMAD, R. 245 $aA machine-learning based ConvLSTM architecture for NDVI forecasting.$h[electronic resource] 260 $c2023 500 $aArticle history: Received 24 September 2019; Received in revised form 7 August 2020; Accepted 5 October 2020: First published 22 October 2020. -- Corresponding author: Rodríguez-Bocca, P.; Facultad de Ingeniería, Instituto de Computación, Universidad de la República, Julio Herrera y Reissig 565, Montevideo, Uruguay; email:prbocca@fing.edu.uy -- FUNDING: This research was partially supported by the "Comisión Sectorial de Investigación Científica (CSIC), UDELAR" and the "Programa de Desarrollo de las Ciencias Básicas (PEDECIBA)" of Uruguay. Some of the calculations reported in this paper were performed in ClusterUY, a newly installed platform for high-performance scientific computing at the National Supercomputing Center, Uruguay. -- Special Issue: OR and Big Data in Agriculture. 520 $aAbstract:Normalized difference vegetation index (NDVI) is an essential remote measurement for agricultural studies because of its strong correlation with crop growth and yield. Accurate and comprehensive NDVI forecasts thus provide effective future projections of crop yield for precise agricultural planning and budgeting. Previous recurrent neural network (RNN) based forecasting methodologies have only performed single-pixel or large-area-average NDVI predictions. We present an alternative RNN-based deep-learning architecture, the convolutional long short-term memory (ConvLSTM), to supply much more comprehensive and detailed NDVI forecasts. In this paper, a single ConvLSTM is capable of 10,000-pixel field-level NDVI predictions, providing a more practical methodology for agricultural producers than single-pixel studies. We compare our model to the parametric crop growth model (PCGM), another multipixel field-level NDVI forecasting technique. We test each model over the same set of soybean crop field pixels with the root mean square error (RMSE) metric. The training configuration of each model is defined by the number of seasons of historical data used for weight optimization. When the best training configuration of the model found is used, the ConvLSTM obtains an RMSE of 0.0782, outperforming the PCGM?s RMSE of 0.0989 (an improvement of 0.0207 in precision represents a large gain in the accuracy of production volume prediction when projected into large production areas). Finally, by comparing the ConvLSTM predictions with the ground truth data over the entire target region rather than just the soybean crop pixels, we discover that the ConvLSTM can also predict NDVI values over the nonsoybean crop as effectively. © 2020 The Authors. 653 $aCONVLSTM NEURAL NETWORKS 653 $aDEEP LEARNING 653 $aNORMALIZED DIFFERENCE VEGETATION INDEX 653 $aOPTIMIZATION 653 $aPREDICTIVE ANALYSIS 700 1 $aYANG, B. 700 1 $aETTLIN, G. 700 1 $aBERGER, A. 700 1 $aRODRÍGUEZ-BOCCA, P. 773 $tInternational Transactions in Operational Research, 2023, Volume 30, Issue 4, Pages 2025 - 2048. doi: https://doi.org/10.1111/itor.12887
Descargar
Esconder MarcPresentar Marc Completo |
Registro original : |
INIA La Estanzuela (LE) |
|
Biblioteca
|
Identificación
|
Origen
|
Tipo / Formato
|
Clasificación
|
Cutter
|
Registro
|
Volumen
|
Estado
|
Volver
|
|
Registro completo
|
Biblioteca (s) : |
INIA La Estanzuela. |
Fecha actual : |
04/03/2020 |
Actualizado : |
05/09/2022 |
Tipo de producción científica : |
Artículos en Revistas Indexadas Internacionales |
Circulación / Nivel : |
Internacional - -- |
Autor : |
MENA, E.; STEWART, S.; MONTESANO, M.; PONCE DE LEÓN, I. |
Afiliación : |
EILYN MENA, Departamento de Biología Molecular, Instituto de Investigaciones Biológicas Clemente Estable, Montevideo, Uruguay.; SILVINA MARIA STEWART SONEIRA, INIA (Instituto Nacional de Investigación Agropecuaria), Uruguay; MARCOS MONTESANO, Departamento de Biología Molecular, Instituto de Investigaciones Biológicas Clemente Estable, Montevideo, Uruguay.; INÉS PONCE DE LEÓN, Departamento de Biología Molecular, Instituto de Investigaciones Biológicas Clemente Estable, Montevideo, Uruguay. |
Título : |
Soybean stem canker caused by diaporthe caulivora; pathogen diversity, colonization process, and plant defense activation. |
Fecha de publicación : |
2020 |
Fuente / Imprenta : |
Frontiers in Plant Science, 23 January 2020, Volume 10, Article number 1733. OPEN ACCESS. Doi: https://doi.org/10.3389/fpls.2019.01733 |
DOI : |
10.3389/fpls.2019.01733 |
Idioma : |
Inglés |
Notas : |
Article history:Received: 20 May 2019.//Accepted: 09 December 2019.// Published: 23 January 2020.
The authors thank Ricardo Larraya for technical assistance and Andrés Di Paolo for assistance in confocal microscopy analysis.This work was supported by Agencia Nacional de Investigación e Innovación (ANII) (grant RTS-1-2014, and graduate fellowships), and Programa de Desarrollo de las Ciencias Básicas (PEDECIBA) Uruguay.The datasets generated for this study can be found in the GeneBank database (MK483139-MK483213, MK507892, and MN584748-MN584826). |
Contenido : |
Abstract:Soybean is an important crop in South America, and its production is limited by fungal diseases caused by species from the genus Diaporthe, including seed decay, pod and stem blight, and soybean stem canker (SSC). In this study, we focused on Diaporthe species isolated from soybean plants with SSC lesions in different parts of Uruguay. Diaporthe diversity was determined by sequencing the internal transcribed spacer (ITS) regions of ribosomal RNA and a partial region of the translation elongation factor 1-alpha gene (TEF1?). Phylogenetic analysis showed that the isolates belong to five defined groups of Diaporthe species, Diaporthe caulivora and Diaporthe longicolla being the most predominant species present in stem canker lesions. Due to the importance of D. caulivora as the causal agent of SSC in the region and other parts of the world, we further characterized the interaction of this pathogen with soybean. Based on genetic diversity of D. caulivora isolates evaluated with inter-sequence single repetition (ISSR), three different isolates were selected for pathogenicity assays. Differences in virulence were observed among the selected D. caulivora isolates on susceptible soybean plants. Further inspection of the infection and colonization process showed that D. caulivora hyphae are associated with trichomes in petioles, leaves, and stems, acting probably as physical adhesion sites of the hyphae. D. caulivora colonized the stem rapidly reaching the phloem and the xylem at 72 h post-inoculation (hpi), and after 96 hpi, the stem was heavily colonized. Infected soybean plants induce reinforcement of the cell walls, evidenced by incorporation of phenolic compounds. In addition, several defense genes were induced in D. caulivora?inoculated stems, including those encoding a pathogenesis-related protein-1 (PR-1), a PR-10, a ?-1,3-glucanase, two chitinases, two lipoxygenases, a basic peroxidase, a defensin, a phenylalanine-ammonia lyase, and a chalcone synthase. This study provides new insights into the interaction of soybean with D. caulivora, an important pathogen causing SSC, and provides information on the activation of plant defense responses. MenosAbstract:Soybean is an important crop in South America, and its production is limited by fungal diseases caused by species from the genus Diaporthe, including seed decay, pod and stem blight, and soybean stem canker (SSC). In this study, we focused on Diaporthe species isolated from soybean plants with SSC lesions in different parts of Uruguay. Diaporthe diversity was determined by sequencing the internal transcribed spacer (ITS) regions of ribosomal RNA and a partial region of the translation elongation factor 1-alpha gene (TEF1?). Phylogenetic analysis showed that the isolates belong to five defined groups of Diaporthe species, Diaporthe caulivora and Diaporthe longicolla being the most predominant species present in stem canker lesions. Due to the importance of D. caulivora as the causal agent of SSC in the region and other parts of the world, we further characterized the interaction of this pathogen with soybean. Based on genetic diversity of D. caulivora isolates evaluated with inter-sequence single repetition (ISSR), three different isolates were selected for pathogenicity assays. Differences in virulence were observed among the selected D. caulivora isolates on susceptible soybean plants. Further inspection of the infection and colonization process showed that D. caulivora hyphae are associated with trichomes in petioles, leaves, and stems, acting probably as physical adhesion sites of the hyphae. D. caulivora colonized the stem rapidly reaching the phloem and the xyl... Presentar Todo |
Palabras claves : |
CELL WALL; DEFENSE GENES; DIAPORTHE CAULIVORA; DISEASE SYMPTOMS; INTERNAL TRANSCRIBED SPACER (ITS) RIBOSOMAL RNA (RDNA); PATHOGEN COLONIZATION; SOYBEAN STEM CANKER; TRANSLATION ELONGATION FACTOR 1-ALPHA GENE (TEF1a). |
Thesagro : |
ENFERMEDADES DE LAS PLANTAS; PATÓGENOS; SOJA. |
Asunto categoría : |
H20 Enfermedades de las plantas |
URL : |
http://www.ainfo.inia.uy/digital/bitstream/item/16699/1/fpls-10-01733.pdf
https://www.frontiersin.org/articles/10.3389/fpls.2019.01733/pdf
|
Marc : |
LEADER 03822naa a2200313 a 4500 001 1060886 005 2022-09-05 008 2020 bl uuuu u00u1 u #d 024 7 $a10.3389/fpls.2019.01733$2DOI 100 1 $aMENA, E. 245 $aSoybean stem canker caused by diaporthe caulivora; pathogen diversity, colonization process, and plant defense activation.$h[electronic resource] 260 $c2020 500 $aArticle history:Received: 20 May 2019.//Accepted: 09 December 2019.// Published: 23 January 2020. The authors thank Ricardo Larraya for technical assistance and Andrés Di Paolo for assistance in confocal microscopy analysis.This work was supported by Agencia Nacional de Investigación e Innovación (ANII) (grant RTS-1-2014, and graduate fellowships), and Programa de Desarrollo de las Ciencias Básicas (PEDECIBA) Uruguay.The datasets generated for this study can be found in the GeneBank database (MK483139-MK483213, MK507892, and MN584748-MN584826). 520 $aAbstract:Soybean is an important crop in South America, and its production is limited by fungal diseases caused by species from the genus Diaporthe, including seed decay, pod and stem blight, and soybean stem canker (SSC). In this study, we focused on Diaporthe species isolated from soybean plants with SSC lesions in different parts of Uruguay. Diaporthe diversity was determined by sequencing the internal transcribed spacer (ITS) regions of ribosomal RNA and a partial region of the translation elongation factor 1-alpha gene (TEF1?). Phylogenetic analysis showed that the isolates belong to five defined groups of Diaporthe species, Diaporthe caulivora and Diaporthe longicolla being the most predominant species present in stem canker lesions. Due to the importance of D. caulivora as the causal agent of SSC in the region and other parts of the world, we further characterized the interaction of this pathogen with soybean. Based on genetic diversity of D. caulivora isolates evaluated with inter-sequence single repetition (ISSR), three different isolates were selected for pathogenicity assays. Differences in virulence were observed among the selected D. caulivora isolates on susceptible soybean plants. Further inspection of the infection and colonization process showed that D. caulivora hyphae are associated with trichomes in petioles, leaves, and stems, acting probably as physical adhesion sites of the hyphae. D. caulivora colonized the stem rapidly reaching the phloem and the xylem at 72 h post-inoculation (hpi), and after 96 hpi, the stem was heavily colonized. Infected soybean plants induce reinforcement of the cell walls, evidenced by incorporation of phenolic compounds. In addition, several defense genes were induced in D. caulivora?inoculated stems, including those encoding a pathogenesis-related protein-1 (PR-1), a PR-10, a ?-1,3-glucanase, two chitinases, two lipoxygenases, a basic peroxidase, a defensin, a phenylalanine-ammonia lyase, and a chalcone synthase. This study provides new insights into the interaction of soybean with D. caulivora, an important pathogen causing SSC, and provides information on the activation of plant defense responses. 650 $aENFERMEDADES DE LAS PLANTAS 650 $aPATÓGENOS 650 $aSOJA 653 $aCELL WALL 653 $aDEFENSE GENES 653 $aDIAPORTHE CAULIVORA 653 $aDISEASE SYMPTOMS 653 $aINTERNAL TRANSCRIBED SPACER (ITS) RIBOSOMAL RNA (RDNA) 653 $aPATHOGEN COLONIZATION 653 $aSOYBEAN STEM CANKER 653 $aTRANSLATION ELONGATION FACTOR 1-ALPHA GENE (TEF1a) 700 1 $aSTEWART, S. 700 1 $aMONTESANO, M. 700 1 $aPONCE DE LEÓN, I. 773 $tFrontiers in Plant Science, 23 January 2020, Volume 10, Article number 1733. OPEN ACCESS. Doi: https://doi.org/10.3389/fpls.2019.01733
Descargar
Esconder MarcPresentar Marc Completo |
Registro original : |
INIA La Estanzuela (LE) |
|
Biblioteca
|
Identificación
|
Origen
|
Tipo / Formato
|
Clasificación
|
Cutter
|
Registro
|
Volumen
|
Estado
|
Volver
|
Expresión de búsqueda válido. Check! |
|
|