Industrial safety · SquintPRO
Forklift–pedestrian detection on the CCTV you already have
A forklift–pedestrian detection system flags moments when people and forklifts share space they should not. Wearable tags and on-truck sensors work from each person or vehicle. Camera analytics work from fixed CCTV and watch the whole aisle, dock or yard in view. SquintPRO, the video analytics platform Interkey builds, detects people and vehicles on your existing cameras. It alerts when a person comes too close to a moving forklift, when a person enters a zone you mark as forklift-only, or when a vehicle enters a pedestrian-only zone, and sends video evidence to the incident workflow your site already uses.
Published 6 min read
In short
- What it watches: zones you draw on each camera view: forklift aisles, walkways, dock lanes and crossings.
- What triggers an alert: a person too close to a moving forklift, a person in a forklift-only zone, or a vehicle in a pedestrian-only zone, during the hours you set.
- What the alert carries: video evidence, with timestamp, camera, zone and event type.
- What it does not do: stop a truck. It is an alerting and evidence tool, not a collision-avoidance system.
- Where it runs: on-premise, so footage stays on site, or in the cloud with 3, 6 or 12 months' retention.
- How it starts: a camera survey, then a pilot on one site with written success criteria.
Three ways sites detect forklift–pedestrian conflicts
| Approach | What it sees | What it needs | Main limit |
|---|---|---|---|
| Wearable tags and vehicle units | Tagged people near tagged vehicles | A device on every person and every truck, kept charged and worn | Visitors, contractors and anyone without a tag are missed |
| Sensors or cameras on the truck | The area around that one truck | Equipment fitted to each vehicle | Covers each truck's surroundings, not the whole floor |
| Analytics on fixed CCTV | Every person and vehicle in the camera's view, tagged or not, including people who come close to a moving vehicle | Cameras that cover the areas that matter | Only where cameras see. Racking and poor light reduce what they see |
These approaches are not rivals. Many sites use fixed-CCTV analytics to watch shared areas and to produce evidence, and keep on-vehicle systems for the trucks themselves. The right mix depends on your layout and your incident history.
What SquintPRO does on forklift routes
- Detects people and vehicles in each camera view, on the cameras the survey qualifies.
- Applies zone rules in both directions. A person in a forklift-only aisle raises an alert, and so does a vehicle in a pedestrian walkway.
- Alerts on unsafe proximity. SquintPRO also alerts when a person comes too close to a moving forklift, on the views the survey qualifies. Test this alert on your own footage during the pilot.
- Follows your shift pattern. Each zone can be active all the time or only during set shifts. A dock lane, for example, can be off-limits only while loading is under way.
- Lets you define the vehicle. Where reach trucks, tuggers or other site vehicles need their own definition, describe them in text or show a reference image from your own cameras. No retraining is needed. Test each new definition on your footage before you rely on its alerts.
- Sends alerts into your existing workflow, in the video management system, control room or incident system the site already uses, with video evidence and a review step.
- Logs every event with timestamp, camera, zone and event type, and turns the events into reports and KPIs.
- Follows a person or vehicle across cameras on a map of the site, so an investigator can see the route before and after an event.
- Searches recorded footage by description or by a reference image, for example "a yellow forklift in the last three hours".
Zone rules that match how your site runs
| Zone | Rule | When active |
|---|---|---|
| Forklift aisle between racking | Alert if a person enters | During operating shifts |
| Marked pedestrian walkway | Alert if a vehicle enters | Always |
| Loading dock lane | Alert if a person enters | While loading is scheduled |
| Crossing point at a blind corner | Alert if a person is in the crossing | During peak shift changes |
| Charging or parking area for trucks | Alert if a person enters outside maintenance hours | Outside maintenance windows |
Your safety team sets the rules, the zones and who receives each alert. Interkey configures them and tunes them during the pilot.
What camera detection cannot do
- It does not prevent collisions. An alert arrives after someone enters a zone or comes too close to a truck. Physical barriers, marked routes, speed limits and training remain the first line of defence.
- It sees only what cameras see. Racking, stacked goods and blind corners create gaps, and the survey shows where they are.
- Light and distance matter. Dim areas, glare from dock doors and very distant views reduce reliability.
- Someone has to act on the alert. Agree who receives each type of alert, and what they do, before go-live.
- It complements tags and on-truck systems. Camera analytics do not replace either of them.
What the camera survey checks
Interkey surveys your existing cameras first. It sorts views into three groups: usable as they are, usable after adjustment, and needing new hardware. Replacement is normally limited to the views that genuinely need it. On forklift routes the survey looks at:
- Whether each aisle, crossing and dock lane is in a view at all.
- Whether people and trucks are large enough in the frame, and not hidden behind racking.
- Lighting at night and at dock doors.
- Overlap at intersections, so a person can be followed from one view to the next.
From alerts to fewer near misses
Each event is logged, so the data shows where and when zone breaches happen: which crossing, which shift, which hour. Safety teams use it to move a walkway, add a barrier, change a loading schedule or plan a toolbox talk on a specific location. Video evidence makes those conversations concrete. See our near-miss definition for how near misses are usually recorded.
Introduce the system as a tool for spotting hazards, not for watching people. Agree early how footage is handled. Workers who see monitoring as surveillance tend to work around it.
Relatednear-miss definition
Deployment, privacy and licensing
- On-premise or cloud. On-premise processing keeps footage on your site. Cloud deployment keeps recordings for 3, 6 or 12 months.
- No face recognition. SquintPRO detects people and vehicles. It does not identify who a person is.
- Pricing. SquintPRO is priced per use case. Each quote is built from the site, the number of cameras and the use cases in scope.
- Delivery. Interkey carries out the survey, the integration with your video and incident systems, deployment, tuning and operation, from Riyadh.
How to start: one site, written success criteria
The first step is a camera-suitability review of one site. If the review is positive, the next step is a pilot. It covers one site, a defined set of cameras, one or two detection cases, and success criteria written before the start, including an acceptable false-alert rate per shift. Ask for the pilot to be priced separately from the rollout, so the decision to continue rests on results. Read the pilot scoping guide.
RelatedRead the pilot scoping guide
Questions safety teams ask
Do we need new cameras?
Usually not for every view. The survey shows which existing cameras are usable as they are, which need adjustment, and which views need new hardware.
Does it stop the forklift?
No. SquintPRO raises alerts and records evidence. It does not control vehicles. Keep barriers, marked routes and any on-vehicle systems in place.
Does SquintPRO alert when a person gets too close to a forklift?
Yes. Besides zone rules, SquintPRO alerts when a person comes too close to a moving forklift, on the views the survey qualifies. Measure missed detections and false alerts on your own footage during the pilot.
Can a zone be active only during certain shifts?
Yes. Each zone can be active all the time or only during the shifts you set.
Does it identify the person?
No. SquintPRO does not use face recognition. It detects that a person or vehicle is in a zone, not who it is.
How accurate is it?
That depends on your cameras, lighting and layout. Measure missed detections and false alerts on your own footage during the pilot, against criteria you set in writing.
Next
Related pages
Next step
Arrange a camera-suitability review
Tell us about the site, roughly how many cameras cover forklift areas, and where people and trucks share space. We will tell you which views can support zone alerts, where the gaps are, and what a pilot could measure.
Or directly
+966-11-2180999 [email protected]Tawuniya Towers, North Tower, 7th Floor, King Fahad Highway, Olaya, P.O. Box 56835, Riyadh 11564, Saudi Arabia