KIIT-MiTA

High-Resolution Drone Imagery for Military Object Detection and Recognition

About Dataset

Drone Images for Military Object Detection

This dataset contains 1,700 high-resolution images captured by drones, annotated in the YOLO format for object detection tasks.
The images span 7 distinct classes, including:

Artilary
Missile
Radar
M. Rocket Launcher
Soldier
Tank
Vehicle
🧩 Key Features
Suitable for military and security-focused research.
Includes detailed bounding box annotations for each object class.
Ideal for training YOLO-based models or other object detection architectures.
⚙️ Usage
This dataset is intended for educational and research purposes only.
Commercial use is strictly prohibited.
If used, proper attribution to the creator is required.

📄 Citation
If you use this dataset in your research, please cite the following paper:

S. Chakrabarty, R. Chatterjee, S. Chakraborty, S. Roy Shuvo, and R. Chowdhury,
"Drones in Defense: Real-Time Vision-Based Military Target Surveillance and Tracking,"
2025 3rd International Conference on Intelligent Systems, Advanced Computing and Communication (ISACC),
Silchar, India, 2025, pp. 508–513,
doi: 10.1109/ISACC65211.2025.10969335

🔗 Paper Link
📘 View on IEEE Xplore

⚠️ Note
This dataset was used as part of the above research publication.
Please ensure proper citation if you use it in any academic or research work.