Related Experiment Video
Updated: May 6, 2026
13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
9.0K
IoTKITs: A novel dataset for IoT education kit recognition
Thanh-Thien Nguyen1,2, Anh-Tuan Nguyen Do3, Duc-Lung Vu1,2
1University of Information Technology, Ho Chi Minh City, Vietnam.
Data in Brief
|June 11, 2025
Summary
This study introduces IoTKITs, a new dataset for identifying Internet of Things (IoT) education kits (KITs). It provides a valuable resource for advancing KIT classification and related educational technology research.
Area of Science:
- Computer Science
- Educational Technology
- Machine Learning
Background:
- Publicly available datasets for identifying Internet of Things (IoT) education kits (KITs) are scarce.
- Accurate classification of IoT KITs is crucial for educational and embedded systems applications.
Purpose of the Study:
- Introduce IoTKITs, a novel, well-annotated dataset for IoT KIT identification and classification.
- Provide a benchmark for evaluating object detection models on IoT KITs.
Main Methods:
- Collected and annotated over 3,000 high-resolution images of various IoT education kits.
- Included popular KIT designs like Arduino Uno, Arduino Nano, and ESP32.
- Evaluated state-of-the-art object detection models (YOLOv5, YOLOv7, Faster R-CNN, SSD) on the dataset.
Main Results:
- Established baseline performance metrics for object detection models on the IoTKITs dataset.
- Demonstrated the dataset's suitability for training and evaluating KIT classification models.
Conclusions:
- The IoTKITs dataset addresses a critical gap in resources for IoT education research.
- Facilitates advancements in KIT classification, embedded systems, and smart learning environments.
