Recording isn't protecting
In most facilities, the camera system is a recording system. You go back and look after something has happened. Watching the footage after a theft doesn't bring the goods back. What you actually need is to know while it's happening.
The classic answer is motion sensors and motion-triggered camera alarms. In practice, those get switched off within weeks:
- Branches in the wind, rain, fog, and moving shadows set them off.
- Cats, dogs, birds, and insects keep them firing all night.
- Headlights, passing clouds, and automatic lighting trigger them.
The result is a security team that gets 40 notifications overnight and mutes the system the next morning. A muted alarm is the same as no alarm.
The difference: person detection, not motion
CamGuard AI looks at what is in the image. The model decides whether the object in frame is a person. That one difference cuts false alarms dramatically. A swaying branch isn't a person. A cat isn't a person. A shadow isn't a person.
Vehicle detection works the same way, as a separate rule. A vehicle entering after hours can raise an alarm, while parked vehicles are classified as stationary and don't alarm over and over.
Zone and time rules
The second pillar is how you define rules. In the panel, for each camera you:
- Draw the monitored zone on the image. Watch just the fence line, the doorway, or the storage yard, and leave the neighboring road out of it.
- Set time ranges. A person during shift hours is normal and won't alarm. The same image after hours will.
- Assign different rules per zone. A loading dock can be open 8am to 6pm on weekdays and fully closed on Saturdays.
This lets the system follow the real working rhythm of your facility, so normal staff activity never triggers an alarm.
A real example: A warehouse monitors its main entrance, loading dock, and rear fence line. On weekdays from 7am to 7pm, only the fence line is active, since people on the dock are expected. After 7pm, all three zones go live. At 2:14am a person is detected on the fence line. The on-site device sounds its siren, the security supervisor gets an automated call, and an email with a snapshot arrives.
Extra layers that cut false alarms
Static object separation
Person-shaped objects that stand still, like a mannequin, a poster, a hi-vis jacket on a hook, or a fixed shadow, won't alarm repeatedly. The system separates out detections that never move.
Headlight glare and vehicle overlap
Headlight glare at night and detections that overlap a vehicle are evaluated separately, so vehicle movement doesn't read as a person entering.
Second-stage verification
For critical events, the detection passes a second visual check before it becomes an alarm, so ambiguous frames don't generate needless alerts.
What happens when an alarm fires
- On-site alarm device sounds a siren, flashes the intruder icon red, and shows the location.
- Automated phone call to your contacts.
- SMS and email, with a snapshot of the moment in the email.
- Mobile push to panel users.
- The event video and evidence frame are archived.
Your officer on duty doesn't just hear the alarm. They see it, and can decide within seconds whether a response is actually needed.
Where it helps most
- Warehouses and distribution centers, large areas left empty at night
- Construction sites, where material, cable, and fuel theft is common
- Industrial plants, targeted for scrap, metal, and equipment
- Parking lots and garages, for vehicle and in-vehicle theft
- Farms and agricultural businesses, remote sites with long perimeters
- Property management, for perimeter security and access control
See the use cases page for the full list.
Common questions
How is it different from a motion sensor?
A motion sensor treats every movement as an alarm, including branches, cats, rain, shadows, and insects. CamGuard AI checks whether there's actually a person in the image, which eliminates the empty alarms that ring all night.
Will staff working during the day trigger alarms?
No. You define monitoring hours per camera and per zone. Person detection during shift hours won't alarm. Alarms happen only outside those hours or inside restricted zones.
Can it detect people in the dark?
Yes. Your camera's infrared night vision is enough. The system runs person detection on black-and-white night footage using night-specific filters.
Can it detect vehicles too?
Yes. Vehicle detection works as a separate rule alongside person detection. A vehicle entering after hours can raise an alarm, while parked vehicles are separated out so they don't alarm repeatedly.