Introducing suspicion levels: why we replaced binary alerts

Most retail security systems give you a binary signal: something happened, or nothing happened. That sounds simple, but in practice it creates two equally bad failure modes — too many alerts that staff start ignoring, or too few that let real incidents slip through.

When we designed Lexerus, we wanted a system that gives staff context, not just noise.

The three levels

NONE is the baseline. No unusual activity detected. The camera is watching but not raising any flags.

ALERT means the system has flagged something suspicious — but it hasn't been verified yet. This could be unusual movement patterns, extended dwell time in a monitored zone, or a detection that needs a second look. You receive a push notification so you're aware, but the system is still verifying.

CONFIRMED is the action level. This means a suspicious behaviour — such as concealment — has been positively verified by our AI verification layer. Only confirmed detections are treated as real incidents. This is the threshold at which your team should investigate.

Why verification matters

A single video frame is unreliable. Lighting changes, camera noise, and normal movements like adjusting clothing can all produce false detections. Our system requires a behaviour to be flagged across multiple frames and then verified by an independent AI review before it escalates to CONFIRMED — not just a single-frame anomaly.

This two-stage approach (detect, then verify) dramatically reduces false positives compared to single-frame detection. Your staff trust that a CONFIRMED alert means something worth investigating.

The result

Three clear levels. No ambiguity. ALERT means "be aware", CONFIRMED means "go check". Staff know exactly what to do at each level, and alert fatigue drops because the noise has been filtered out before it reaches them.