Synthetic CCTV Knife Detection Image Data | Knives & Bladed Weapons | YOLO + COCO Annotations

Knives are small, thin and often half-hidden in a hand, which makes them one of the hardest objects for security cameras to catch. Real footage of knife incidents is rare and sensitive. This dataset gives you the volume and variety you need. What's inside - Synthetic CCTV images of people carrying, holding or brandishing knives in shops, stations, car parks, school corridors, nightlife streets and building entrances. - Ceiling, wall and pole-mounted camera views, by day, at night and in monochrome IR mode. - Knives appear small and distant as well as close, held low, raised or partly hidden, the way they look on real security cameras. - Hard negatives (phones, wallets, bottles, umbrellas, shopping bags) so the model learns what not to flag. - Classes: person, knife. - Every person and knife boxed, in YOLO and COCO format, with a train/val split, data.yaml and a dataset card. Volumes - 1,000-image pack available off the shelf (zip download). - Up to 20,000 images available on request, plus 100+ matching CCTV video clips (720p-1080p, 20 FPS, MP4). - Custom packs for your camera angles, sites and knife types. Why synthetic - No real people, faces or locations, so no GDPR or consent issues. - Rare and dangerous scenarios at volume, without staging them. - Every image reviewed for realism and label accuracy before release. Use cases: knife and edged-weapon detection for VMS and video analytics, school, retail and transport security, model benchmarking and false-positive reduction.