Abstract
This paper introduces IoTKITs, a novel and well-annotated dataset specifically designed for the identification and classification of IoT education kits (KITs), addressing the scarcity of publicly available datasets in this domain. The dataset comprises over 3,000 high-resolution images of various KITs, including popular designs such as Arduino Uno, Arduino Nano, ESP32, and others, with detailed annotations for object detection tasks. To establish baselines, we evaluated state-of-the-art object detection models, including YOLOv5, YOLOv7, Faster R-CNN, and SSD, on the dataset. IoTKITs is designed to advance KIT classification research and foster applications in education, embedded systems, and smart learning environments.