Diagnosing Hard-to-See 'Honeybee Mites' with AI: Contributing to the Prevention of Mass Honeybee Colony Deaths

A technology that can quickly and accurately diagnose 'honeybee mites', a pest considered a major cause of mass honeybee colony deaths during winter, has been developed domestically for the first time in the world.

The Rural Development Administration announced on August 20 that it has jointly developed an AI-based 'real-time honeybee mite detection device' with Kangwon National University.

A real-time detection device called 'BeeSion' that can automatically determine the presence of honeybee mites within 30 seconds by capturing images of honeycomb using AI technology. (Photo by Rural Development Administration)

Honeybee mites are pests that parasitize honeybees inside the hive, directly damaging their development or transmitting viruses that cause diseases and lead to colony collapse. They are known as the most serious pest affecting honeybees worldwide, and are difficult to control.

An official from the Rural Development Administration explained, "Honeybee mites live inside the hive, making them very difficult to observe with the naked eye. In particular, observation and control become even more challenging in the high temperatures of summer, making it easy to miss the optimal control period. Even experienced beekeepers require more than 30 minutes to carefully inspect a single hive, and for elderly beekeepers, it is especially difficult to detect honeybee mites outdoors in the hot and humid summer."

To address this, the Rural Development Administration developed a real-time detection device called 'BeeSion', which can automatically determine the presence of honeybee mites within 30 seconds by capturing images of the honeycomb using AI technology.

This device can simultaneously analyze 16 types of pest and growth information, including not only honeybee mites but also chalkbrood (a disease where larvae dry out and harden), honeybees infected with diseases, bees with wing deformities, and abnormal larvae. Depending on the level of infection, it provides scientific control standards such as recommendations for control, caution stage, and intensive control.

The accuracy of honeybee mite analysis reaches 97.8%. Its simple design allows even elderly or novice users to make accurate and rapid diagnoses. The Rural Development Administration expects that by using this device, beekeepers will be able to detect honeybee mites and other pests, as well as abnormal signs in honeybees, in advance at the beekeeping site, enabling proactive management and reducing colony losses.

The Rural Development Administration has already filed a patent application for the device. The technology will be transferred to industry this year, and production of the product will begin. After on-site demonstrations, the device is planned to be distributed in earnest to beekeepers nationwide starting in 2028.

Bang Hyesun, Director of the Agricultural Biology Division at the Rural Development Administration, stated, "This achievement is the first case of integrating data and AI technology into beekeeping, which has traditionally relied on experience, and will serve as a turning point for precision management and automation of pest monitoring. We will continue to advance digital-based proactive monitoring systems to protect honeybees and enhance the competitiveness of the beekeeping industry."

This content was produced with the assistance of AI translation services.