Aerial Cows: 1,723 Images for Livestock Detection

1,723 aerial images annotated for cow detection form this dataset, part of the Roboflow 100 benchmark. It was created by LibreYOLO and last updated on January 20, 2026. The dataset contains a single object class and is split into 1,084 training, 340 validation, and 299 test images.

Use Cases

• Training object detection models based on aerial imagery.
• Benchmarking model performance on livestock detection tasks.
• Developing agricultural monitoring tools based on the single 'cow' class.
• Evaluating model generalization across the provided train/validation/test splits.

Strengths

• 1,723 total annotated images provide a substantial base for model training.
• Clear data splits of 1,084 training, 340 validation, and 299 test images facilitate standard machine learning workflows.
• Part of the multi-domain Roboflow 100 benchmark, suggesting a standardized format.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset focuses on a single class ('cow'), which may limit its utility for multi-class detection tasks.

Provenance

Source

LibreYOLO

Collection Method

Part of the Roboflow 100 benchmark collection; specific gathering method is unknown.

Freshness

Last updated 2026-01-20 23:21:25; freshness should be verified.