BeeHealthAI: Deep Learning-Based Detection of Varroa Mites and Wax Moth Damage from Images
Supervisor Name
Ahmad Amarneh
Supervisor Email
a.amarneh@ptuk.edu.ps
University
Palestine Technical University Kadoorie
Research field
Artificial Intelligence
Bio
Dr. Ahmad Alashqar Amarneh is an agricultural scientist, researcher, and university academic with over 20 years of experience in agricultural research, plant health, integrated pest management, and agricultural extension. He holds a PhD from the University of Córdoba, Spain, specializing in crop diversification and plant disease management. His expertise includes sustainable agriculture, plant pathology, pest and disease control, and the development of innovative, climate-resilient agricultural systems. He is committed to advancing scientific research, promoting sustainable farming practices, and strengthening agricultural innovation and food security.
Description
This project aims to develop an artificial intelligence (AI)-based mobile application for the early detection of visible signs of honeybee pests and hive damage, focusing on *Varroa destructor* mites and wax moth infestations. These threats significantly affect honeybee health, colony productivity, and beekeeping sustainability. The proposed system integrates deep learning, computer vision, and mobile technologies to analyze images of bees and honeycombs and identify potential pest-related indicators. Image preprocessing techniques will be applied to improve data quality, while Convolutional Neural Networks (CNNs) and transfer learning models, such as MobileNetV2 and EfficientNet, will be trained and evaluated for image classification. A user-friendly mobile application, developed using Flutter, will allow beekeepers to capture or upload images, receive AI-generated predictions with confidence scores, and access pest-specific inspection recommendations. A FastAPI-based backend and database will manage image processing, screening results, and hive inspection records. The system is intended as a preliminary screening tool rather than a definitive diagnostic solution. Ultimately, the project seeks to support beekeepers in identifying potential threats, improving hive monitoring, and making timely, informed decisions to promote sustainable beekeeping practices.