Deep Learning-Based Portable Image Analysis System for Real-Time Detection of Vespa velutina

Honeybees pollinate over 75% of the total food resources produced annually, and they produce valuable hive products, such as bee pollen, propolis, royal jelly. However, species Asian hornet (Vespa velutina) feed on more than 85% honeybees, causing a decline in their population considerable damage to beekeepers Korea. To prevent portable real-time monitoring system was developed that detects V. velutina individuals notifies users presence. This designed with focus portability ease installation, can be found various areas apiary sites. detect velutina, an improved convolutional neural network YOLOv5s trained 1960 high-resolution (3840×2160) image data. At confidence threshold ≥0.600 intersection union ≥0.500, performance terms detection accuracy, precision, recall, F1 score, mean average precision high. A distance-based comparison showed able while three beehives. During field test beehives, could 83.3% during hunting activities send alarms registered mobile application users.