The way bees are kept can affect the health of the colony, thereby increasing or decreasing losses. Similarly, methods of diagnosis and prevention of bee diseases can influence the health of the colony. Therefore, it is important to signal in time the course of key life parameters in the hive and to monitor the manifestations of hive activities in the area of the hive entrance. That is why early signaling is important in the course of key life parameters in the hive and monitoring the manifestations of hive activities in the area of the hive entrance. Monitoring the condition of the colony using modern digital technologies and information on the course of key life parameters in the colony, including the detection of the Varroa mite, are useful not only for remote control of hives, but especially for possible early intervention of beekeepers - beginners, experienced hobby beekeepers, professional beekeepers and all those who need to control hives remotely. The solution aims in particular to contribute to improving the health status of bees, especially by early signaling of the course of key life parameters in the hive, monitoring the manifestations of colony activity in the area of the entrance to the hive and refining the detection of the parasite - Varroa destructor mite - on the hive mat. The technical solution is developed in cooperation with the Institute of Applied Informatics at the University of South Bohemia in Č. Budějovice. The new colony monitoring system uses the latest findings from artificial intelligence, neural networks, deep learning and IOT technologies. Thanks to this, it is possible to automatically monitor variables such as the temperature and humidity in the hive, the temperature in a tuft of bees, the temperature, humidity and barometric pressure outside the hive, bee sounds, hive weight, CO2 concentration in the hive, as well as the visual monitoring of the hives by the camera system - the frequency of bees entering and leaving the hives. Beekeepers can have an overview not only of the condition of the colony, the conditions inside the hive and the surrounding area, but they can also monitor their colonies, for example, using GPS in case of hive theft or using a scale to monitor the growth, to detect honey theft, swarming, etc. Using artificial intelligence, neural networks and deep learning, the detection of a queen bee on a honeycomb or the detection of a pathogen on a hive mat can then be refined.
