Detection rule

Line crossing detection for perimeters and internal zones

Line crossing detection is the most widely used video analytics rule in commercial intruder detection. It defines a virtual tripwire in the camera scene and generates an event when a classified object crosses it in a specified direction.

This guide covers how the rule actually works, how to calibrate it for reliable perimeter and internal use, and the typical false-alarm sources that separate a stable deployment from a noisy one.

Written by Intruder Detect Editorial Team · Reviewed by a commercial security specialist
Definition

What line crossing detection is

Line crossing is a spatial rule applied on top of object classification: a virtual line is drawn in the camera view and a detection event is raised when a classified object crosses it. Modern platforms allow direction constraints (into, out of, both), object type filters (person, vehicle) and time-of-day scheduling.

The rule is defined in image space but back-solved to real-world geometry during calibration, so detection is consistent regardless of where in the frame the crossing occurs.

Applications

Where line crossing is used

Line crossing is the default perimeter analytics rule — a tripwire set on or just inside the fence line detects boundary intrusion before the intruder is on-site. Internally it protects restricted zones, loading bays, plant rooms and stock cages, and it is widely used on unmanned reception areas out-of-hours.

  • Perimeter fence lines — early boundary breach detection
  • Internal restricted zones — server rooms, plant, stock cages
  • Loading bays and yards — out-of-hours vehicle and person detection
  • Unmanned entrances — after-hours reception and lobby coverage
Setup

Calibration and line placement

Line placement matters more than any other single setting. A line placed exactly on a fence produces false alarms from fence movement in wind; a line placed a metre inside the boundary produces reliable detection with substantially lower noise. On internal zones the line should sit inside the zone rather than at the doorway to avoid triggering on approach.

Calibration also sets pixels-per-metre on target at the line, which the classifier uses to reject undersized detections such as small animals and windblown debris.

Rule tuning

Direction and classification filters

Direction filtering is the single most effective false-alarm reduction on line crossing. On a perimeter, only the inward-crossing direction usually matters; outward crossings during the day are legitimate and shouldn't create alarms.

Classification filters restrict detection to relevant object types. On perimeter tripwires, filtering to person and large vehicle typically eliminates the wildlife category of false alarms without missing meaningful intrusion.

Performance

Typical false-alarm sources

The recurrent false-alarm sources on line crossing detection are shadows moving across the line at dawn and dusk, headlights from vehicles on adjacent roads, waterfowl and deer on rural perimeters, wind-moved vegetation crossing the line volume, and reflections from wet surfaces. Each has an established mitigation — schedule masking, direction rules, class filters, line repositioning or supplemental lighting.

Response

Verification and monitored response

A line crossing event should trigger automatic display of the crossing clip at the alarm receiving centre, ideally with the pre-alarm buffer, so an operator can verify the object and direction before escalation. This is what converts detection into monitored response and is the design that qualifies for police response under UK URN policy and equivalent US arrangements.

FAQs

Frequently asked questions

Where should the tripwire line be placed?

Line placement matters more than any other single setting. On perimeters, place the line one to two metres inside the fence to avoid fence-movement triggers. On internal zones, place the line inside the protected area rather than at the doorway so the rule fires on entry rather than approach. Placement should be walk-tested and recorded in the commissioning documentation.

How is line crossing different from zone intrusion?

Line crossing detects the moment of transition across a defined boundary; zone intrusion detects continued presence within a defined area. On perimeters they are often used together — line crossing for the boundary event, zone intrusion for follow-on presence detection. Each rule has its own tuning parameters and neither replaces the other operationally on a well-designed site.

Does line crossing work at night?

Yes, provided the camera has adequate low-light or infrared capability, or is paired with active IR illumination. Classifier accuracy in true darkness on a visible-light camera without IR support falls off sharply. On rural perimeters, dual-spectrum or thermal cameras materially improve overnight reliability compared with visible-light plus IR floodlight designs.

What causes most line crossing false alarms?

The recurring sources are shadows moving across the line at dawn and dusk, headlights from adjacent roads, wildlife on rural perimeters, wind-moved vegetation and reflections from wet surfaces. Each has a defined mitigation — schedule masking, direction filters, class filters, line repositioning or supplemental lighting — and a professionally tuned system addresses these during commissioning rather than in-service.

Can line crossing detect vehicles as well as people?

Yes. Classification filters allow the rule to fire on person, vehicle, or both, and to apply different direction and schedule constraints per class. On mixed sites this is essential — a perimeter tripwire that ignores staff vehicles during the day but detects any pedestrian, and reverses that logic overnight, is a routine and well-supported configuration on modern analytics platforms.

Is line crossing enough on its own for a perimeter?

For most commercial perimeters, line crossing on properly specified cameras with monitored response is the correct primary detection layer. On high-consequence sites — utilities, data centres, defence, critical national infrastructure — a second independent detection technology such as fence-mounted PIDS or radar is normally added to give defence-in-depth against single-point defeat.

How is line crossing verified for police response?

Line crossing events are verified when the camera clip — including the pre-alarm buffer — auto-presents to an operator at an alarm receiving centre for human assessment before escalation. Verified events sustain URN-based police response eligibility in the UK and comparable arrangements in the US. Unverified line crossing alarms alone do not qualify for police response under current policy.

Related

Continue in the intruder detection hub

The intruder detection hub sets out how this technology fits alongside the other layers of a complete commercial design.

Line crossing detection guidance

Speak to a specialist about line crossing detection

Tell us about your site and we'll connect you with a commercial security specialist who understands your detection, monitoring and response requirements.