AI video analytics

Technology

AI video analytics

Definition

AI video analytics, also called video content analysis (VCA), uses computer vision models to automatically detect, classify and track things in camera footage, such as people, vehicles, PPE and behaviours, and turn them into alerts and data.

Older video analytics relied on pixel changes and simple rules, which made them prone to false alarms from shadows, rain and moving trees. Modern deep-learning models recognise objects and actions, so they can tell a person from a fox or a hard hat from a bare head.

Analytics can run on the camera, on a local box, or in the cloud. On remote sites, running detection on the device means it keeps working when the connection drops.

Typical uses include intrusion detection, PPE and safety monitoring, vehicle counting and ANPR, footfall and occupancy, and searching recorded footage by event instead of by hour.

How Scopis AI helps

Scopis AI is a hardware-agnostic video analytics platform: AI agents for safety, security and operations on existing ONVIF cameras and solar 4G units.

Go deeper: Adding AI to your existing CCTV: what you actually need

Next step

See Scopis on your own sites.

A 30 minute walkthrough of the platform on live deployments: detection, deterrence, zone control and the alert history behind them. No obligation, no deck.