Telraam's system receives a lot of data from your camera. These data need to be interpreted in order to know in which category (pedestrian, cyclist, car, large vehicle) a given vehicle should be placed. This is not always an easy task. This article will explain how the system works and what the classification of vehicles consists.
First, Telraam distinguishes between pedestrians, two-wheelers and vehicles (= cars + large vehicles) based on the fullness value and axis ratio. Once this distinction is made, Telraam starts distinguishing between cars versus large vehicles.
Phase 1: classification of pedestrians vs. two-wheelers vs. vehicles
Based on fullness value and axis ratio
Classification is based on the average observed fullness and axisratio of each observed object (that satisfies a set of criteria that helps filtering out any movement in the field of view that is likely not connected to road users).
These two parameters are distance (between the camera and the object) and speed independent, meaning that a car 10 meters or 5 meters from the camera driving 10 or 80 km/h will still have the same axisratio and fullness value.
-medium.png)
-medium.png)
To have a better grasp of what axisratio and fullness is, take a look at the figures above. The black rectangular shape is the circumscribed rectangle, while the width and the height of this rectangle are represented by the arrows. The area of the car or cyclists is represented here by the red shaded area. Fullness means how well this red shaded area fills up the space inside the black rectangle.
Axisratio is the ratio of the two arrows. It is easy to see, that the fullness of the car is significantly larger than the fullness of the cyclist (and it becomes even more so when we are looking at the objects from the typical height of a Telraam camera), while the axisratio of the cyclists (around 1, since width and height is more or less the same) is larger than the axisratio of the car (where width is twice the height, so axisratio = height/width = 0.5).
Global classificator matrix
Classification is done using these parameters by a global classifier, which is basically a two dimensional lookup table where we have (using an automated method) defined areas (in the axisratio fullness parameter space) where cars, cyclists, and pedestrians will typically fall.
Phase 2: classification of cars vs. large vehicles
After the distinction between pedestrians vs. two-wheelers vs. vehicles is done, it is still necessary to distinguish between cars and heavy vehicles within the class "vehicles". This is done based on a cut-off value. However, a calibration (period) is needed so that this cut-off value is optimally calculated for your location. For more explanation, see article "Telraam calibration ".
My data seems inconsistent
The distinction between large vehicles and cars is not always correct. The number of cars may be underestimated and the number of large vehicles overestimated, or vice versa. The difference between cyclists and pedestrians is always obvious either, especially if several pedestrians or cyclists are riding together. To learn more about the inconsistencies you may observe in your Telraam data, read the following article: "Potential inaccuracies with Telraam ".
More (detailled) info below
If you would like more information (including technical information) on the classification method, you can download and print the following article: "Vehicle classification".
The whole system is continuously evaluated and, as soon as possible, a new refinement of the data is established specifically for each Telraam.
However, we ask you to be careful with the use of the large vehicle count data by the Telraam. This data does not have the same accuracy as the Telraam car data.
