Weapon and Threat Detection Dataset for Object Detection Models

A cleaned and restructured object detection dataset for identifying weapons, aggressors, and security-related items. The dataset was created by Subh775 and was last updated on October 2, 2025. It is formatted for use with the Hugging Face datasets library and training frameworks like Ultralytics for YOLO models.

Use Cases

• Train object detection models to identify weapons based on the described security-related items.
• Fine-tune YOLO models for threat detection using the pre-formatted dataset.
• Benchmark performance of security-focused computer vision models on a dedicated dataset.

Strengths

• The dataset has been cleaned and restructured for immediate use, as stated in the description.
• It is specifically formatted for compatibility with the Hugging Face datasets library and Ultralytics YOLO frameworks.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Derivative work based on the weapon-detection Object Detection Model from Hugging Face.

Freshness

Last updated 2025-10-02 12:13:18; freshness should be verified.