Estimating Arabica Coffee Productivity Using Planetscope Images and Artificial Neural Networks
DOI:
https://doi.org/10.25186/.v21i.2401Abstract
Research on orbital remote sensing and artificial neural networks (ANN) for estimating coffee productivity has advanced in Brazil, but applications remain scarce in states such as Bahia. This study aimed to develop and validate a model to estimate Arabica coffee (Coffea arabica L.) yield using PlanetScope images and ANN, based on two commercial areas with contrasting production systems (rainfed and irrigated). Georeferenced productivity samples (2 per hectare) were collected during the 2024 harvest, and PlanetScope images were analyzed from August 2023 to May 2024. In R software, vegetation indices related to vigor, nutritional status, and water stress were calculated, and their minimum, mean, and maximum values were extracted. Most variables followed a normal distribution (Shapiro–Wilk test, p < 0.05), enabling Pearson’s correlation analysis. The best correlation was observed in Santa Vera, where the CCCI index of 08/20/2023 showed a moderate positive correlation with yield (r = 0.66; p = 0.001). After ANN training in SPSS software, the
model achieved very high predictive performance in both areas, with R² and adjusted R² above 0.99, low errors (MAE = 0.031–0.072; RMSE = 0.042–0.087; relative RMSE = 0.029–0.043), and almost no bias (BIAS = 0.001; relative BIAS = 0). Among the vegetation indices, CCCI was the most relevant, followed by NDRE, NDVI, NDWI, and EVI. Overall, the findings demonstrate that combining PlanetScope imagery with ANN provides a highly accurate approach for estimating Arabica coffee productivity under different production systems in Bahia.
Key words: Precision Agriculture; Coffee Farming; Artificial Intelligence; Coffea arabica L.
References
Abreu Júnior, C. A. M. et al. (2022). Estimating coffee plant yield based on multispectral images and machine learning models. Agronomy, 12(12):3195.
Agência Sebrae. (2024). Café da Chapada Diamantina (BA) é a mais nova Indicação Geográfica brasileira. Available in: https://agenciasebrae.com.br/cultura-empreendedora/cafe-da-chapada-diamantina-ba-e-a-mais-nova-indicacao-geografica-brasileira/.
Alves, S. A. (2023). A inserção da cafeicultura no planalto da conquista - Bahia: transformações sociais e econômicas da região.
Geofronter, 9(1).
Anderson, T. W. (1984). Estimating linear statistical relationships. Annals of Statistics, 12(1):1-45.
Auravant. (2024). Auravant for Education. Available in: <https://www.auravant.com/auravant-for-education/.>
Barbosa, B. D. S. et al. (2021). UAV-based coffee yield prediction utilizing feature selection and deep learning. Smart Agricultural Technology, 1:100010.
Bolaños, J., Corrales, J. C., & Campo, L.V. (2023). Feasibility of early yield prediction per coffee tree based on multispectral aerial imagery: Case of arabica coffee crops in Cauca-Colombia. Remote Sensing, 15(1):282.
BoiarskiI, B., & Hasegawa, H. (2019). Comparison of NDVI and NDRE indices to detect differences in vegetation and chlorophyll content. Journal of Mechanics of Continua and Mathematical Sciences, 14:20-29.
Chemura, A. et al. (2018). Mapping spatial variability of foliar nitrogen in coffee (Coffea arabica L.) plantations with multispectral Sentinel-2 MSI data. ISPRS Journal of Photogrammetry and Remote Sensing, 138:1-11.
Chlingaryan, A., Sukkarieh, S., & Whelan, B. (2018). Machine learning approaches for crop yield prediction and nitrogen status estimation in precision agriculture: A review. Computers and Electronics in Agriculture, 151:61-69.
Companhia Nacional de Abastecimento - CONAB. (2025). Acompanhamento da safra brasileira de café, v. 11 – Safra 2024, n.4- Quarto levantamento. Brasília, p. 1-53. Available in: https://www.gov.br/conab/pt-br/atuacao/informacoes-agropecuarias/safras/safra-de-cafe/4o-levantamento-de-cafe-safra-2024/boletim-cafe-janeiro-2025.
Dias, T. L. et al. (2019). Evapotranspiração e coeficiente de cultura do cafeeiro irrigado a partir de imagens de sensores orbitais. IRRIGA, 1(1):31–37.
Dias, C. G. et al. (2024). Climate risks and vulnerabilities of the Arabica coffee in Brazil under current and future climates considering new CMIP6 models. Science of The Total Environment, 907:167753.
Dalchiavon, F. C., & Carvalho, M. de P. (2012). Correlação linear e espacial dos componentes de produção e produtividade da soja. Semina: Ciências Agrárias, 33(2):541-552.
Dhika, H. R. et al. (2024). Predicting coffee productivity using artificial neural networks. BIS Information Technology and Computer Science, 1:V124007.
Fagundes, R. B., & Bolfe, Édson L. (2022). Diagnosis about the perspectives of precision applications of coffee growing technologies in municipalities of Bahia, Brazil. Coffee Science, 17:e172034.
Fabri, T. M. P. et al. (2023). Comportamento de índices de vegetação em cafeeiros utilizando sensor multi-espectral em aeronave remotamente pilotada. Revista Ifes Ciência, 9(1):1-15.
Formaggio, A. R., & Sanches, I. D. A. (2017). Sensoriamento remoto em agricultura. São Paulo: Oficina de Textos, 285p.
Gao, Bo-Cai. (1996). NDWI-A normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sensing of Environment, 58(3):257-266.
Gitelson, A. A. et al. (1996). Detection of red-edge position and chlorophyll content by reflectance measurements near 700 nm. Journal of Plant Physiology, 148:501-508.
Gitelson, A. A. et al. (2005), Remote estimation of canopy chlorophyll content in crops. Geophysical Research Letters, 32(8):L08403.
Ramirez, G. M., & Zullo Júnior, J. (2010). Topografia, fotogrametria e sensoriamento remoto. Engenharia Agrícola, 30(3):468-479.
Huete, A. R. (1988). A soil-adjusted vegetation index (SAVI). Remote Sensing of Environment, 25(3):295-309.
Huete, A. et al. (2002). Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sensing of Environment, 83(1-2):195-213.
Ioannidis, T. D. et al. (2017). The power of bias in economics research. The Economic Journal, 127(605):F236-F265.
Fitzgerald, G. F. et al. (2006). Spectral and thermal sensing for nitrogen and water status in rainfed and irrigated wheat environments. Precision Agriculture, 7(4):233-248.
Luna, D. R. et al. (2020). Análisis de los índices de vegetación NDVI, GNDVI y NDRE para la caracterización del cultivo de café (Coffea arábica). Ingeniaría y desarrollo, 38(2):2145-9371.
Martins, R. N. et al. (2021). A novel vegetation index for coffee ripeness monitoring using aerial imagery. Remote Sensing, 13(2):263.
Martello, M. et al. (2022). Coffee-yield estimation using high-resolution time-series satellite images and machine learning. AgriEngineering, 4(4):888-902.
Mosomtai, G. et al. (2021). Functional land cover scale for three insect pests with contrasting dispersal strategies in a fragmented coffee-based landscape in Central Kenya. 2020. Agriculture, Ecosystems & Environment, 319:107558.
Koaudio, L. et al. (2021). Performance of a process-based model for predicting robusta coffee yield at the regional scale in Vietnam. Ecological Modelling, 443:0304-3800.
Kittichotsatsawat, Y. et al. (2023). Forecasting arabica coffee yields by auto-regressive integrated moving average and machine learning approaches. AIMS Agriculture and Food, 8(4):1052-1070.
Oliveira, D. T. et al. (2020). Convolutional neural networks in predicting cotton yield from images of commercial fields. Computers and Electronics in Agriculture, 171:105307.
Pereira, A. (2019). Análise temporal do índice NDVI utilizando o Google Earth Engine: Estudo de caso na cafeicultura. In C. C. Santos. Book: O Semiárido Brasileiro e suas especificidades. Ponta Grossa, PR: Atena Editora, [recurso eletrônico].
Planet Labs. (2025). PlanetScope product specifications. Available in: <https://assets.planet.com/docs/Planet_PSScene_Imagery_Product_Spec_letter_screen.pdf.>.
Putra, B. T. W., & Soni, P. (2018). Enhanced broadband greenness in assessing Chlorophyll a and b, Carotenoid, and Nitrogen in Robusta coffee plantations using a digital camera. Precision Agriculture, 19:238-256.
Qi, J. et al. (1994). A modified soil adjusted vegetation index. Remote Sensing of Environment, 48(2):119-126.
Rodrigues, S. A. et al. (2021). Efeito da chuva de granizo em variedades do café arábica por meio de índices de vegetação. Agrarian, 14(54):433-441.
Kogan, F. N. (2012). Historical Perspectives on AVHRR NDVI and Vegetation Drought Monitoring. In B. D. Wardlow., M. C. Anderson., & J. P. Verdin, (Ed.). Remote sensing of drought: innovative monitoring approaches. Boca Raton: CRC Press/Taylor & Francis.
Ruiz, I. et al. (2022). On the combined use of phenological metrics derived from different PlanetScope vegetation indices for classifying savannas in Brazil. Remote Sensing Applications: Society and Environment, 26:100764.
Santana, L. S. et al. (2022). Precision coffee growing: a review. Coffee Science, 17:e172007.
Santos, S. A. d. et al. (2023). Evaluation of the water conditions in coffee plantations using RPA. AgriEngineering, 5(1):65-84.
Santos, L. M. d. et al. (2022). RPA images in the mapping of the chlorophyll index of coffee plants. Sustainability, 14:13118.
Santinato, F. et al. (2016). Análise espacial da produtividade do cafeeiro arábica utilizando diferentes densidades amostrais. Revista Brasileira de Engenharia Agrícola e Ambiental, 20(12):1075-1080.
Superintendência de Estudos Econômicos e Sociais da Bahia- SEI. (s.d). Perfil dos Territórios de Identidade da Bahia. Available in: <https://sei.ba.gov.br/index.php?option=com_content&view=category&id=1518&Itemid=284&lang=pt>.
Silva J. B. C. da., & Pinto, P. A. L. de A. (2024). Impactos socioeconômicos das mudanças climáticas na produção do café: uma revisão sistemática da literatura. 2024. Revista Brasileira de Climatologia, 35(20):155-178.
Silva, F. M., & Alves, M. C. Cafeicultura de Precisão. Lavras: Editora UFLA, 2013.
Silvestre, M. R., & Bezerra, M. I. S. (2015). Modelos de decomposição temporal e de regressão harmônica: uma comparação para a série mensal da temperatura mínima e máxima média de Presidente Prudente (SP). Revista Brasileira de Meteorologia, 30(4):457-466.
Sousa, A. (2019). Coeficiente de correlação de Pearson e coeficiente de correlação de Spearman. O que medem e em que situações devem ser utilizados?. «Correio dos Açores: Matemática», 19p.
Souza, J. B. C. et al. (2022). Integrating satellite and UAV data to predict peanut maturity upon artificial neural networks. Agronomy, 12(7):1512.
Tedesco, D. et al. (2021). Use of remote sensing to characterize the phenological development and to predict sweet potato yield in two growing seasons. European Journal of Agronomy, 129:126337.
Thao, N. T. T. T. et al. (2022). Early prediction of coffee yield in the central highlands of vietnam using a statistical approach and satellite remote sensing vegetation biophysical variables. Remote Sensing, 14(13):2975.
Tuia, D. et al. (2022). Perspectives in machine learning for wildlife conservation. Nature Communications, 13:792.
Viet, L. V., &Thuy, T. T. T. (2023). Improving the quality of coffee yield forecasting in Dak Lak Province, Vietnam, through the utilization of remote sensing data. Environmental Research Communications, 5:095011.
Willmott, C. J., & Matsuura, K. (2005). Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Research, 30(1):79-82.
Zanella, M. A. et al. (2024). Coffee yield prediction using high-resolution satellite imagery and crop nutritional status in Southeast Brazil. Remote Sensing Applications: Society and Environment, 33:101092.
Zanzarini, F. V. et al. (2013). Correlação espacial do índice de vegetação (NDVI) de imagem Landsat/ETM+ com atributos do solo. Revista Brasileira de Engenharia Agrícola e Ambiental, 17(6):608-614.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Coffee Science - ISSN 1984-3909The copyright of the articles published in this journal belongs to the authors, with the journal holding the first publication rights. As the articles are published in this journal under open access, they may be freely used, with proper attribution, for educational and non-commercial purposes
