Turning movement counting

Do you want to know the traffic flow at intersections and traffic circles? The Swarm Perception Platform provides the solution to capture the number of vehicles traveling from an origin zone (origin) to a destination zone (destination), broken down by vehicle type (class).

What data can be generated?

For this use case, our software provides you with all the necessary data for source-destination traffic analyses - the number of vehicles from one zone to another, including classification according to the standards TLS (technical delivery conditions for route stations) issued by the German Highway Research Institute (BASt).

What needs to be considered for a successful analysis?

Camera setup

In the table below you will find detailed information on the requirements, settings and positioning of the camera.

Setting	Recommended
Pixels per Meter (PPM) Pixels Per Meter is a unit of measurement used to define the amount of potential image detail a camera provides at a given distance.	>30 PPM for object classes car, truck >60 PPM for object classes person, bicycle, motorbike Using the camera parameters defined below ensures that the minimum required PPM value is achieved. Tip: Use the lens calculator from Axis or a generic lens calculator.
Video resolution	1280 × 720 Pixel
Video log/Codec	RTSP/H264
Focal length	2.8 mm - 12 mm
Camera mounting - distance to object center	Object classes car, truck 5-30 m (2.8 mm focal length) 35-100 m (12 mm focal length) Object classes person, bicycle, scooter 3-12 m (2.8 mm focal length) 25-50 m (12 mm focal length)
Camera mounting - mounting height	Up to 10 m Note: Mounting higher can help to minimize obstruction by large objects.
Camera mounting - vertical angle to the object	<50 ° Note: The correct vertical angle to the object should be achieved by setting the correct distance to the vehicle and the optimum mounting height.
Camera mounting - horizontal angle to the object	0 ° - 360 °

Specifications for the environment

Object speed	<50 km/h
Day/Night/Lighting	During the day/well lit/night vision
Indoor/outdoor use	Outdoor
Expected accuracy (from origin to destination when all environmental, hardware and camera requirements are met)	origin-destination counts and classification main classes: >85 % (for vehicles) Classification main classes: >95 % classification sub classes: >85 %

Hardware specifications

Supported products	BMA, mobile BMA, B101, B401
Frames Per Second (fps)	25