CC BY 4.0Estira, Marydel C.Bernal, Yhoebe Rae C.Acocos, Kyla Q.Garcia, Jasmin M.Kimayong, Princess Joy D.2026-04-222026-04-222025-11Bernal, Y. R. C., Acocos, K. Q., Garcia, J. M., & Kimayong, P. J. D. (2025). UNAScan: Applying deep learning for early detection of sugarcane leaves diseases for basi production in Naguilian, La Union.[Unpublished Undergraduate Thesis]. Don Mariano Marcos Memorial State University - Mid La Union Campus, City of San Fernando, La Union. Lakasa ti Sirib, DMMMSU Institutional Repository.https://lakasa.dmmmsu.edu.ph/handle/123456789/1362Full textThis study was conducted to support sugarcane farmers in detecting and managing leaf diseases that affect crop quality and yield, especially for Basi production. The capstone project aimed to build a dataset of sugarcane leaf images and train an optimized deep learning model using the MobileNetV3 architecture. The modeling and development process followed the CRISP-DM framework, ensuring structured stages from data understanding to model evaluation. The trained model was then integrated into both web and mobile platforms through the Agile Methodology. A descriptive and developmental research design was applied with twenty (20) respondents: five (5) agricultural staff, five (5) Basi producers, and ten (10) sugarcane farmersfrom Naguilian, La Union. The system achieved a 98% accuracy rate and a System Usability Scale (SUS)score of 85.6, indicating excellent usability. Respondents described the system as efficient, user-friendly, and highly valuable for early detection and effective management of common sugarcane leaf diseases.xii, 82 p.: ill. (col.).EnglishSugarcane productsSugarcane products--AnalysisDeep learning (Machine learning)Sugarcane leaf scaldThesis