Solutions

Forklift–pedestrian detection from existing cameras

Make forklift and pedestrian interactions visible: real-time warnings when people and vehicles share space, and data on where near-misses happen.

Forklift–pedestrian detection uses computer vision to detect forklifts and people in the same camera view, estimate their proximity or zone overlap, and trigger an alert or record a near-miss when a defined rule is broken.

The problem

Many warehouse incidents happen where pedestrian walkways cross vehicle routes: aisle ends, dock doors, blind corners. Near-misses in these places are rarely reported, so the risk stays invisible until an accident occurs.

Painted lines and mirrors help, but they do not tell safety managers how often the rules are broken or which crossing needs attention first.

How it works

  1. 1Existing cameras covering aisles, crossings and docks are connected over RTSP/ONVIF.
  2. 2The model detects forklifts (and other configured vehicles) and people in each frame and tracks them over time.
  3. 3Rules define pedestrian-only zones, vehicle-only zones and proximity conditions per camera.
  4. 4When a rule is broken, an alert is raised and the event is logged with a clip for near-miss analysis.

Existing IP cameras / NVR

RTSP / ONVIF streams

Edge device on-site

AI models + rules; video stays local

Events

Time, camera, zone, evidence frame

Dashboard, alerts, API

Notifications, relays, REST/webhook

Typical on-premise deployment. Cloud and hybrid options are also available.

What can be detected

  • Pedestrian and forklift inside the same zone at the same time
  • Pedestrian entering a vehicle-only zone
  • Forklift entering a pedestrian-only walkway
  • Near-miss hotspots by location and time of day

Typical scenarios

  • Aisle ends and blind intersections in warehouses
  • Dock doors and loading ramps
  • Production halls where internal logistics share space with operators
  • Local warnings via beacon or siren at a specific crossing

Deployment options

On-premise / edge

Analysis runs on hardware inside your network. Video does not leave the site; only events and metrics are stored or shared.

Cloud

Suitable for distributed, low-camera-count sites where central management matters more than keeping video on-site.

Hybrid

Video is processed on-site, while events, dashboards and multi-site reporting are managed centrally.

Working with existing cameras

  • Wide-angle views over crossings work well; the whole interaction area should be in frame.
  • Higher frame rates help with fast-moving vehicles.
  • Proximity is estimated from the 2D image and zone calibration; it is not a certified distance measurement.
  • Most IP cameras and NVRs that provide RTSP or ONVIF streams can be connected, regardless of brand.
  • Before a pilot, sample footage from each candidate camera is reviewed for resolution, angle, distance and lighting.
  • Where a view is not suitable, the recommendation is usually to reposition the camera or add one — not to replace the whole system.

Integration

  • Dashboard, mobile and e-mail alerts
  • Relay outputs for local beacons or sirens
  • REST API / webhook for WMS or EHS reporting

Privacy and security

  • No face recognition is used in safety and operations analytics; events describe what happened, not who.
  • On-premise processing keeps video inside your network.
  • Evidence clips, retention periods and user access are configurable and should follow your GDPR/KVKK policies.
  • Workplace monitoring should be introduced with clear employee information and, where applicable, consultation with employee representatives.

Limitations

  • This is a monitoring and awareness layer, not a replacement for vehicle-mounted safety systems or a safety-rated interlock.
  • Heavy occlusion by racks or loads can reduce detection quality; camera placement is reviewed before the pilot.

How a pilot works

  1. 01

    Discovery

    We agree on the problem, the owner, the sites and what a successful outcome would look like.

  2. 02

    Camera assessment

    Sample footage from candidate cameras is reviewed for angle, resolution, distance and lighting.

  3. 03

    Use-case definition

    Events, zones, rules, alert recipients and success metrics are written down before deployment.

  4. 04

    Pilot deployment

    Edge hardware or cloud processing is set up and connected to the selected cameras.

  5. 05

    Validation

    Detections are checked against reviewed footage; true detections, false alarms and misses are counted per camera.

  6. 06

    Report

    Results, limitations and recommended changes are documented and shared with all stakeholders.

  7. 07

    Rollout decision

    Based on measured results, you decide whether and how to expand to more cameras, sites or use cases.

Frequently asked questions

Do forklifts need tags or sensors?

No. Detection is camera-based, so vehicles and people do not need wearables or tags.

Can an alert trigger a local warning light?

Yes, events can drive relay outputs for beacons or sirens near the crossing. The exact wiring is defined with your maintenance team.

Will it slow down operations with false alarms?

Zone and duration rules are tuned during the pilot and false alarms are measured per camera. Rules are only rolled out once the alarm rate is acceptable to the operations team.

Start with the cameras you already have

Send us a few frames from your cameras and the problem you want to solve. We will tell you honestly what is feasible.