AI algorithm

E-Bike and Elevator Entry Detection

Local AI video detection for e-bikes and other non-motor vehicles entering elevator lobbies, elevator cars, and controlled building access areas.

Request a Demo View Products

E-bike entry detection is a video rule for a defined building location. It can be used at elevator lobbies, elevator-car entrances, underground access points, and corridors where non-motor vehicles are restricted. The useful outcome is a time-stamped, reviewable event that allows staff to respond under the property’s own operating rules.

What the camera should see

The strongest view usually captures the vehicle before the elevator door or boundary line. A camera that only sees handlebars, a wheel, or a person pushing the vehicle gives the model less evidence. In tight elevator spaces, a project should test normal passenger traffic, walking bicycles, delivery carts, wheelchairs, and partial occlusion before setting the alarm threshold.

Event handling

Projects can use local audio reminders, snapshots for a property platform, or a review queue for staff. Whether the result triggers an access-control action must be decided with the building and elevator system owners. The AI camera or edge device provides the video event. It does not replace elevator safety systems or fire-protection equipment.

Acceptance points

Acceptance should include the actual camera angles, daytime and nighttime samples, people walking beside a vehicle, different vehicle shapes, and short pass-through cases. The team should agree what counts as an event, how repeated alerts are suppressed, and who reviews exceptions.

Algorithm deployment notes

E-bike and elevator entry: camera position and event definition

An e-bike and elevator entry event is a movement through a defined doorway or protected elevator area. The quality of the result depends on the camera showing the approach, threshold, object, and direction clearly enough to distinguish entry from nearby corridor activity.

Frame the door threshold and approach path

A useful camera view includes the elevator entrance or controlled doorway, part of the approach route, and the floor boundary used by the rule. The object should remain visible for enough frames as it approaches and crosses the line. Door movement and crowding should be included in the test.

Decide which objects are in scope

The building team should define whether the rule applies to e-bikes, scooters, batteries, related carrying behaviour, or a narrower object class. This supports a clearer review process when wheelchairs, prams, luggage trolleys, folded items, or partially covered objects appear in the same view.

Keep evidence suitable for building staff

Event records should identify the building, camera, time, direction, defined zone, image, and related video where available. Access should be restricted to authorised staff and retention should follow the building policy and local privacy requirements.

Test ordinary building traffic

Validate the rule at peak lobby times, with different lighting, door states, object sizes, and occlusions. The aim is to document what the camera can confirm and ensure uncertain events can be reviewed instead of being treated as conclusive automatically.

Related planning: Community and elevator deployment overview. Camera analytics should be used within the building’s agreed access and safety process.

Project inquiry

Tell us your camera environment and required AI analytics.

For the first version this site uses a lightweight email-based inquiry flow. It avoids a database and keeps the English site simple until inquiry volume justifies a dedicated CRM integration.