Simuletic/UAV-Aerial-View-Battle-Tank-Detection-Dataset

UAV & Aerial View: Battle Tank Detection Dataset (free sample)

A free sample of 164 synthetic UAV and drone-view images of battle tanks in desert, village and field terrain, with 716 YOLO bounding boxes for battle_tank, for training and testing aerial vehicle detection models. Views range from oblique flyovers to straight top-down. Made by Simuletic, which builds synthetic security and defence datasets for computer vision.

Full dataset and licence: simuletic.com/datasets/uav-aerial-battle-tank-detection-dataset

	Free sample (this page)	Full dataset
Images	164 PNG and JPG, 512 to 1,024 px square	3,000 images
Classes	battle_tank	battle_tank
Labels	YOLO .txt, 716 boxes, 1–10 per image	YOLO
View	UAV and drone cameras, oblique to top-down; desert, villages, fields	balanced mix of terrain, altitudes and camera pitches; haze and dust plumes
Licence	CC BY-NC 4.0	one-time licence: research $149 / commercial $590 per 1,000 images

What's in the images

Main battle tanks and other heavy tracked armoured vehicles seen from UAV and drone cameras, from oblique flyover angles down to straight top-down (nadir) views. Real aerial imagery of armoured vehicles is restricted and hard to collect, which leaves few public training sets for this task.

• Terrain. Dry desert, Eastern European villages, open fields and seasonal vegetation, with atmospheric haze.
• Several targets per image. 1 to 10 tanks per image, 4.4 on average.
• Small objects. 99% of the boxes cover less than 5% of the image area.

Classes and boxes

Class	YOLO id	Boxes in sample
battle_tank	0	716

The included data.yaml and classes.txt name this class military tank; it is the same class (id 0).

Folder layout

Drone & Aerial View: Battle Tank Detection Dataset/
  images/        # 164 images: 74 image<N>.png and 90 image_<N>.jpg
  labels/        # one YOLO .txt per image, one line per tank
  data.yaml      # class map (0: military tank)
  classes.txt

Images are square: 75 at 1,024 px, 15 at 768 px and 74 at 512 px. The folder name contains spaces, & and :, so quote it in paths.

Each YOLO label line is class_id x_center y_center width height, normalised to 0–1. The first two of seven tanks in image_00000.txt:

0 0.519 0.577 0.108 0.164
0 0.704 0.166 0.082 0.051

Quick start (Ultralytics YOLO)

Create a data.yaml:

path: "Drone & Aerial View: Battle Tank Detection Dataset"
train: images
val: images   # small sample: use the same images, or split them yourself
names:
  0: battle_tank

Then train:

from ultralytics import YOLO

model = YOLO("yolo11n.pt")
model.train(data="data.yaml", epochs=50, imgsz=640)

Use cases

• Vehicle detection and tracking on UAV and drone cameras
• Aerial surveillance and situational-awareness research
• Detection of vehicles near buildings, tree lines and in desert terrain
• Sim-to-real (Sim2Real) domain adaptation experiments
• Edge inference tests on drone-mounted hardware (for example Jetson, Hailo or Coral)

How the images were made

The images are fully synthetic (computer-generated) by Simuletic: no real people, places or incidents are depicted, so there are no privacy or consent issues.

Limitations. Synthetic images do not cover every camera, lens, compression artefact or scene you will meet in deployment. Validate on your own footage, and for production combine synthetic and real data.

Full dataset and custom data

The full UAV & Aerial View: Battle Tank Detection Dataset has 3,000 images, YOLO, class battle_tank. Licence: research $149 / commercial $590 per 1,000 images. Larger sets, other scenes, camera angles or classes: see custom data or write to contact@simuletic.com.

This free sample is also on Kaggle.

More synthetic datasets from Simuletic

• Military Drone Swarm & Saturation Attack Dataset
• Airport Drone Threat & Safety Dataset
• Long-Distance Wildfire & Smoke Detection Dataset
• All datasets: simuletic.com/datasets · Hugging Face · Kaggle · GitHub

Licence

This free sample is licensed under CC BY-NC 4.0: you may share and adapt it for non-commercial purposes, as long as you credit Simuletic. The full dataset is sold under a separate one-time research or commercial licence.

Citation

@misc{simuletic_tank_2026,
  author       = {{Simuletic}},
  title        = {UAV & Aerial View: Battle Tank Detection Dataset},
  year         = {2026},
  publisher    = {Simuletic},
  howpublished = {\url{https://huggingface.co/datasets/Simuletic/UAV-Aerial-View-Battle-Tank-Detection-Dataset}},
  note         = {Free sample. Full dataset: \url{https://simuletic.com/datasets/uav-aerial-battle-tank-detection-dataset}}
}

Questions or feedback: contact@simuletic.com