María Fernanda Calderón Vega

Dr. María Fernanda Calderón Vega

orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 June 2025 in Electronics
Verified
Journal Article
Open access
PDF file
Luis Jácome GalarzaLuis Jácome Galarza
Luis Jácome Galarza
MRMiguel Realpe
Miguel Realpe
MVMarlon Santiago Viñán-Ludeña
Marlon Santiago Viñán-Ludeña
MVMaría Fernanda Calderón Vega
María Fernanda Calderón Vega
SLSilvia Alexandra Jaramillo Luzuriaga
Silvia Alexandra Jaramillo Luzuriaga
A more accurate crop yield estimation is essential for optimizing agricultural productivity and resource management. Traditional machine learning models, such as linear regression and convolutional neural networks (CNNs), often struggle to integrate multimodal data sources effectively, limiting their predictive accuracy. In this study, we propose the AgriTransformer model, a transformer-based model that enhances crop yield prediction by leveraging attention mechanisms for multimodal data fusion. The AgriTransformer model incorporates tabular agricultural data and vegetation indices (VI), allowing dynamic feature interaction and improved interpretability. Experimental results have demonstrated that AgriTransformer significantly outperforms conventional approaches, achieving an R2 of 0.919, compared to 0.884 for the best-performing linear regression model. The findings highlight the importance of structured tabular data in yield estimation, while VI serves as a complementary feature that increases the prediction capability and confidence. This study highlights the potential of transformer-based architectures in precision agriculture, offering a scalable and adaptable framework for crop yield forecasting. The AgriTransformer model enhances predictive accuracy and generalization across diverse agricultural conditions by prioritizing relevant features through attention mechanisms.

4 July 2023
Verified
Conference Paper
Closed access
PZPabelco Zambrano
Pabelco Zambrano
MVMaría Fernanda Calderón Vega
María Fernanda Calderón Vega
HVHéctor Villegas
Héctor Villegas
JPJonathan Paillacho
Jonathan Paillacho
DPDoménica Pazmiño
Doménica Pazmiño
MRMiguel Realpe
Miguel Realpe
Crop monitoring and diagnosis are crucial for efficient agricultural production, and unmanned aerial vehicle (UAV) Remote Sensing can assist in achieving this goal. This article offers an automated Systematic Literature Review (SLR) of UAV Remote Sensing for crop monitoring and diagnosis. This review analyzes the primary scientific applications and trends in this area using Deep Learning techniques to automatically select relevant articles and validate them through full reading. The SLR collected over 800 papers, of which 64 met the selection process. The articles selected by Deep Learning classifiers were successfully cataloged with high accuracy in pre-selecting articles for review. F1 scores of 93% were achieved in tests with unpublished examples for the classifier model. The review of the 64 primary studies reported a peak in UAV Remote Sensing applications in 2020, attributed to the increasing diffusion of precision farming practices with technological equipment. The UAV Remote Sensing application objectives included crop monitoring, pest and disease detection, yield prediction, and plant nutrition. Artificial Intelligence, particularly Machine Learning and Deep Learning, are widely used for UAV Remote Sensing analysis. The NDVI is the most applied vegetation index for crop condition assessment and monitoring. The proposed solution for automating the literature selection process for precision agriculture-related scientific articles can be used in other areas that require extensive literature review.

6 June 2022 in La Granja
Verified
Journal Article
Open access
PDF file
UAUlises Gildardo Quiroz Antúnez
Ulises Gildardo Quiroz Antúnez
AMAlejandro Ismael Monterroso-Rivas
Alejandro Ismael Monterroso-Rivas
MVMaría Fernanda Calderón Vega
María Fernanda Calderón Vega
AGAdán Guillermo Ramirez Garcia
Adán Guillermo Ramirez Garcia
Se evaluó las zonas de aptitud actual y futura para los cultivos de cacao (Theobroma cacao L.) y café (Coffea arabica L.), en escenarios de cambio climático en la región del Soconusco, Chiapas, México y se definieron las áreas donde aumentó o disminuyó la aptitud de la tierra. Lo anterior permite tener información para recomendar su manejo y conservación. Se emplearon dos metodologías: ponderación de factores y factores limitantes; ambas con base en los valores óptimos y extremos de tolerancia de cada cultivo para cada requerimiento edafoclimático. Con la primera, se determinaron las áreas potenciales para el establecimiento de los cultivos; con la segunda, se diferenciaron las combinaciones de las variables ambientales que tuvieron lugar dentro del área de estudio. Se evaluaron escenarios de cambio climático y se determinó la distribución y superficies de aptitud. Además, se realizó una comparación entre aptitud actual y futura de ambos cultivos para cuantificar el impacto del cambio climático. Los resultados indican que la aptitud futura de la tierra tendrá una disminución de las áreas potenciales para el establecimiento de ambos cultivos de entre 4,5 y 4,8% para café y de 7,2 a 9,3% para cacao.

5 December 2022 in Económicas CUC
Verified
Journal Article
Open access
PDF file
AGAdán Guillermo Ramirez Garcia
Adán Guillermo Ramirez Garcia
IEIrma Cecilia Castillo Escalante
Irma Cecilia Castillo Escalante
MVMaría Fernanda Calderón Vega
María Fernanda Calderón Vega
DMDayana Duffus Miranda
Dayana Duffus Miranda
AHAna Arelis Pirela Hernández
Ana Arelis Pirela Hernández
En el municipio de Álamos, Sonora México el agua es un recurso escaso por lo que se tiene un interés muy especial en la conservación del mismo. El objetivo del presente trabajo fue analizar las percepciones de los habitantes de siete comunidades ubicadas en el Área de Protección de Flora y Fauna Sierra de Álamos Río Cuchujaqui para estimar el valor económico del recurso agua y determinar su disponibilidad de pago. La investigación tiene un enfoque metodológico cuantitativo, con alcance de carácter exploratorio, descriptivo y con diseño de campo. Se usó el método de valoración contingente para poder asignar un monto al valor económico del recurso agua, así como la investigación acción participativa para analizar, describir y comprender el fenómeno estudiado. El instrumento de utilizado fue la entrevista. De los resultados encontrados destaca que la disponibilidad de pago en general para las comunidades analizadas es de $94.27/persona y la comunidad que tiene la mayor disposición en cuanto a pagar por el recurso es Güirocaba con la cantidad de $1,000.00/persona. Se concluye que, la información que ofrece este estudio puede emplearse para focalizar los esfuerzos de promoción de programas como el de pagos por servicios ambientales hidrológicos y la propuesta de políticas públicas que contribuyan a la ampliación del servicio y el uso sustentable del recurso agua.

1 January 2017
Verified
Conference Paper
Closed access
PDF file
ASAdriana Santos
Adriana Santos
CGCarlos Gonzalez
Carlos Gonzalez
MMMaría Fernanda Miño
María Fernanda Miño
CPCinthia Párraga
Cinthia Párraga
MVMaría Fernanda Calderón Vega
María Fernanda Calderón Vega

1 January 2016
Verified
Conference Paper
Open access
PDF file
PSPriscila Serrano
Priscila Serrano
MVMaría Fernanda Calderón Vega
María Fernanda Calderón Vega
Ecosystems play an important role for the environmental and sustainability, that makes mandatory to comprehend the effects of them due to climate change to assess the conservation status, vulnerability and take mitigations measures. This paper aims to analyse the Holdridge life zone changes in the Guayas province (Ecuador) by using geospatial analysis between 1978 and 2000. The methodology was structured in three components that were: climatic indicator determination of precipitation and temperature, Holdridge life zone classification and Holdridge life zone changes evidence occurred in the studied years. The results showed that in 2000 classification had six life zones which are also included in the 1978's work, unlike the latter, lower montane moist forest and tropical thorn stepped have disappeared, while dry tropical forest increased its surface on a share of the tropical very dry forest. At the same time, the pre montane moist forest and the tropical moist forest recorded a minimal increase in their surface. In addition, the most affected ecosystems were conservative and protective zones. Keywords-Geographic Information System (GIS), Holdridge life zone, climate change.