YO-CNN: A Hybrid Deep Learning Model for Prescription Mapping in Precision Agriculture
Mathematical and Computational Applications · 06.11.2026 · 31 (6) · 250
DOI: 10.3390/mca31060250 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mathematical and Computational Applications, online erschienen am 06.11.2026. Verfasst von Syeda Iqra Hassan und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Modern agriculture is encountering a dilemma about increasing food production with minimum environmental damage. Current approaches to combating pests and monitoring the state of the soil are inadequate in terms of efficiency and require too much effort. Although the use of unmanned aerial vehicles (UAVs) in combination with artificial intelligence (AI) has partially resolved the situation, solutions are still too diverse and not optimally integrated with geographic information systems (GIS). This study proposes a lightweight, domain-adapted detector based on the YOLO framework, called YO-CNN. The algorithm is meant to be used in the field to instantly recognize early signs of crop stress, such as pest infestation or low plant density, by processing images obtained from UAVs. The results indicate that the method is suitable for practical applications such as prescription mapping. The main contributions are domain-specific model adaptation, the compact size of the solution that allows its implementation on hardware platforms in the field, and integrated analysis that allows prescriptions to be made. An experiment was conducted on a rice field of 40 acres in Tando Muhammad Khan, Pakistan, to evaluate the model. A set of 4200 annotated images (3000 pest and 1200 plant-stand images) was used to train and validate the algorithm. On the held-out validation images, the pest detector reached a precision of about 0.90, a recall of about 0.90, an mAP@0.5 of about 0.93 and an mAP@0.5:0.95 of about 0.69. In addition, the detections were used to generate prescription maps, which agronomists inspected qualitatively. Validation on independent farms and seasons, and field trials of the resulting prescriptions, are still required.
Zitieren
APA
Hassan, S. I., Zia, M. Y. I., Khan, M. U., Rashid, M., Alam, M. M., & Su’ud, M. M. (2026). YO-CNN: A Hybrid Deep Learning Model for Prescription Mapping in Precision Agriculture. Mathematical and Computational Applications, 31(6), 250. https://doi.org/10.3390/mca31060250
Vancouver
Hassan SI, Zia MYI, Khan MU, Rashid M, Alam MM, Su’ud MM. YO-CNN: A Hybrid Deep Learning Model for Prescription Mapping in Precision Agriculture. Mathematical and Computational Applications. 2026;31(6):250. doi:10.3390/mca31060250
BibTeX
@article{Hassan2026,
author = {Hassan, Syeda Iqra and Zia, Muhammad Yousuf Irfan and Khan, Muhammad Umar and Rashid, Muhammad and Alam, Muhammad Mansoor and Su’ud, Mazliham Mohd},
title = {YO-CNN: A Hybrid Deep Learning Model for Prescription Mapping in Precision Agriculture},
journal = {Mathematical and Computational Applications},
year = {2026},
volume = {31},
number = {6},
pages = {250},
doi = {10.3390/mca31060250}
}
EUVIMED