Comparative study of Convolutional Neural Networks performance and efficiency with YOLOv8 models applied for pest detection purposes in bean plants.

Authors

  • Julio Cesar De Dios Garcia Universidad Politecnica de Pachuca
  • Nantai Nava Nolazco Universidad Politécnica de Pachuca, Centro de Bachillerato Tecnológico industrial y de servicios No. 222
  • Ernesto Monroy Cruz Tecnológico Nacional de México Campus Atitalaquia, Centro de Estudios Tecnológicos industrial y de servicios No. 26
  • Zamora Galvan Marlon Universidad Politécnica de Pachuca, Centro de Bachillerato Tecnológico industrial y de servicios No. 222
  • Luis Rodolfo Garcia Carrillo Klipsch School of Electrical and Computer Engineering at New Mexico State University, USA
  • Victor Adrian Macias Martinez Centro de Bachillerato Tecnológico industrial y de servicios No. 222
  • Noe Ordaz Rodriguez Centro de Bachillerato Tecnológico industrial y de servicios No. 222
  • Hugo Enrique Orozco Garcia Centro de Bachillerato Tecnológico industrial y de servicios No. 222

DOI:

https://doi.org/10.61467/2007.1558.2025.v16i2.603

Keywords:

Neural network, Convolution Neural Network (CNN), YoloV8, Pest Detection

Abstract

Neural Networks have significantly evolved, particularly in their application to computer vision. This paper presents a comprehensive comparison of different versions of YOLOv8, such as YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x for the detection of pests in bean plants, leveraging the capabilities of Convolutional Neural Networks.

To train the neural network using different versions of YOLOv8, identical conditions were applied, such as the amount of environment light, the number of labeled images, epochs, and batch size.

The results indicate that, as the complexity of the YOLO model increases, the training time escalates significantly. This increase corresponds to a more detailed data processing approach in advanced models. The results also provide insight on which model emerges as the most balanced option, offering the highest precision without compromising too much on speed. 

One of the models achieves the highest precision,  making it reliable for accurate object detection but the speed is slow compared with other models. Otherwise exceptional precision makes it ideal for tasks where accurate identification is critical. The slight reduction in speed does not significantly hinder its overall performance in contexts where precision and detection distance are prioritized.

Author Biographies

Julio Cesar De Dios Garcia, Universidad Politecnica de Pachuca

Julio Cesar DE DIOS GARCIA was born in Apan, Hidalgo, Mexico, in 1993. He received the B.S. degree in Mechatronics Engineering in 2015 and the M.S. degree in Mechatronics in 2019, both from the Polytechnic University of Pachuca, Hidalgo, Mexico. He conducted his master's research under the supervision of Ph.D Filiberto Muñoz Palacios at the Center for Research and Advanced Studies (CINVESTAV) in the area of unmanned aerial vehicles.

He carried out a research stay at Texas A&M University-Corpus Christi, under the supervision of Ph.D Luis Rodolfo Garcia Carrillo. Professionally, he worked in the maintenance area at EGO Electronics in 2015 and as an engineering developer at Mecatroniks in 2016. Since 2019, he has served as an associate professor of high school education at the Centro de Bachillerato Tecnológico Industrial y de Servicios No 222. Concurrently, he has been a part-time professor at the Polytechnic University of Pachuca.

His research interests include computer vision systems, neural networks, unmanned aerial vehicles, obstacle avoidance algorithms, and embedded systems.

Ernesto Monroy Cruz, Tecnológico Nacional de México Campus Atitalaquia, Centro de Estudios Tecnológicos industrial y de servicios No. 26

Ernesto Monroy Cruz received the Master degree in Mechatronics from Universidad Politécnica de Pachuca in 2015. From January to April 2014, he was a visiting scholar at the Unmanned Systems Laboratory from the Texas A&M University - Corpus Christi, USA. He received the PhD. degree in Advanced Manufacturing from the Centro de Tecnología Avanzada (CIATEQ), Mexico, in 2023, where he was advised by Professor Luis Rodolfo García Carrillo from the New Mexico State University. At present he serves as an Assistant Professor with Tecnologico Nacional de Mexico Campus Atitalaquia in Hidalgo, Mexico.

Luis Rodolfo Garcia Carrillo , Klipsch School of Electrical and Computer Engineering at New Mexico State University, USA

Luis Rodolfo GARCIA CARRILLO was born in Gomez Palacio, Durango, Mexico in 1980. He received the B.S. degree in Electronic Engineering in 2003, and the M.S. degree in Electrical Engineering in 2007, both from the Institute of Technology of La Laguna, Coahuila, Mexico. He received the Ph.D. degree in Control Systems from the University of Technology of Compiegne, France, in 2011, where he was advised by Professor Rogelio Lozano. From 2012 to 2013, he was a Postdoctoral Researcher at the Center for Control, Dynamical systems and Computation (CCDC) at the University of California, Santa Barbara, where he was working with Professor Joao Hespanha. During this time he was also a Researcher in the Institute for Collaborative Biotechnologies (ICB). From 2013 to 2020 he served as an Assistant Professor of Engineering at Texas A&M University - Corpus Christi (TAMUCC), At present he serves as an Assistant Professor with the Klipsch School of Electrical and Computer Engineering at New Mexico State University, in Las Cruces, NM, USA.

His research interests include control systems, multi-agent systems, game theory, and the use of computer vision in feedback control.

Noe Ordaz Rodriguez , Centro de Bachillerato Tecnológico industrial y de servicios No. 222

Noé Ordaz Rodríguez was born in Mexico City, Mexico in 1972. He received the B.S. of Electronic Systems Engineer with Honors in 1997 from the Monterrey Institute of Technology, Mexico City campus. He specialized in Mechatronics at the CNAD. From 1997 to 2000 he worked at Infosel and TV AZTECA in the systems area, occupying positions ranging from technical support to LAN network administrator. From 2001 to 2003, he was systems and operations manager at SISMART. In 2003 he joined the DGETI as a teacher at CBTIS 32 and later joined CBTIS 222 in Pachuca Hidalgo, Mexico, at the Mechanic Academy. During this time he has obtained the status of research professor in the years 2018, 2019, 2020, 2021, 2022, 2023 and 2024.

His research interests include control systems, AI systems and integration of mechatronic projects in community problems.

Downloads

Published

2025-03-25

How to Cite

De Dios Garcia, J. C., Nava Nolazco, N., Monroy Cruz, E., Galvan Marlon, Z., Garcia Carrillo , L. R., Macias Martinez, V. A., Ordaz Rodriguez , N., & Orozco Garcia, H. E. (2025). Comparative study of Convolutional Neural Networks performance and efficiency with YOLOv8 models applied for pest detection purposes in bean plants. International Journal of Combinatorial Optimization Problems and Informatics, 16(2), 112–122. https://doi.org/10.61467/2007.1558.2025.v16i2.603

Issue

Vol. 16 No. 2 (2025)

Section

Articles

License

Copyright (c) 2025 International Journal of Combinatorial Optimization Problems and Informatics

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.