This study presents the development and evaluation of a mobile-based diagnostic tool for identifying coffee pests and diseases through image processing, targeting coffee farmers in Kalinga Province, Philippines. The system integrates traditional visual inspection with advanced image recognition algorithms, including K-means clustering, support vector machine and inceptionV3, to enable early and accurate detection of major pests such as the coffee berry borer (Hypothenemus hampei (Ferrari)) and coffee stem borer (Xylotrechus quadripes Chevrolat), as well as common diseases including coffee leaf rust (Hemileia vastatrix Berk. & Broome) and coffee berry disease (Colletotrichum kahawae J.M. Waller & Bridge). Pilot testing was conducted among farmers, academic researchers and IT experts. System quality was evaluated using the ISO/IEC 25010 software quality model, covering functionality, reliability, usability, efficiency, maintainability and portability. Findings revealed high compliance across all quality characteristics (overall mean = 4.52), indicating strong system performance. Farmers reported improved satisfaction in pest and disease management and improved access to timely treatment recommendations. The study recommends expanding the image dataset to improve detection accuracy and developing an iOS-compatible version to broaden accessibility. This project demonstrates the potential of combining image processing and mobile technology to empower coffee farmers, support sustainable pest and disease management and enhance crop productivity in resource-limited agricultural contexts.
C A Giarhard (Tue,) studied this question.