Limitations of transfer learning for chilean cherry tree health monitoring: when lab results do not translate to the orchard

Vista sencilla de ítem
dc.contributor.authorHidalgo, Mauricio
dc.contributor.authorYanine, Fernando
dc.contributor.authorGalleguillos Silva, Renato Bruno
dc.contributor.authorLagos, Miguel
dc.contributor.authorKumar Sahoo, Sarat
dc.contributor.authorParedes, Rodrigo
dc.date.accessioned2025-08-25T16:56:44Z
dc.date.available2025-08-25T16:56:44Z
dc.date.issued2025-08-13
dc.description.abstractChile, which accounts for 27% of global cherry exports (USD 2.26 billion annually), faces a critical industry challenge in crop health monitoring. While automated sensors monitor environmental variables, phytosanitary diagnosis still relies on manual visual inspection, leading to detection errors and delays. Given this reality and the growing use of AI models in agriculture, our study quantifies the theory–practice gap through comparative evaluation of three transfer learning architectures (namely, VGG16, ResNet50, and EfficientNetB0) for automated disease identification in cherry leaves under both controlled and real-world orchard conditions. Our analysis reveals that excellent laboratory performance does not guarantee operational effectiveness: while two of the three models exceeded 97% controlled validation accuracy, their field performance degraded significantly, reaching only 52% in the best-case scenario (ResNet50). These findings identify a major risk in agricultural transfer learning applications: strong laboratory performance does not ensure real-world effectiveness, creating unwarranted confidence in model performance under real conditions that may compromise crop health management.
dc.identifier.citationProcesses, Vol 13, N° 8, 2559 (2025) p. 1-16.
dc.identifier.doihttps://doi.org/10.3390/pr13082559
dc.identifier.issne2227-9717
dc.identifier.orcidhttps://orcid.org/0000-0003-3191-3673
dc.identifier.orcidhttps://orcid.org/0000-0003-1086-0840
dc.identifier.orcidhttps://orcid.org/0000-0001-6065-5218
dc.identifier.urihttps://hdl.handle.net/20.500.12254/4277
dc.language.isoen
dc.publisherMDPI
dc.rightsAtribución-NoComercial-CompartirIgual 3.0 Chile (CC BY-NC-SA 3.0 CL)
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/3.0/cl/
dc.subjectTransfer learning
dc.subjectImage classification
dc.subjectField performance
dc.subjectCherry tree health
dc.subjectQuality assurance
dc.titleLimitations of transfer learning for chilean cherry tree health monitoring: when lab results do not translate to the orchard
dc.typeArticle
Archivos
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
processes-13-02559.pdf
Tamaño:
1.65 MB
Formato:
Adobe Portable Document Format
Descripción:
Texto completo
Descargar
Bloque de licencias
Mostrando 1 - 1 de 1
No hay miniatura disponible
Nombre:
license.txt
Tamaño:
347 B
Formato:
Item-specific license agreed upon to submission
Descripción:
Descargar
Colecciones
Artículos de Revistas