Side view camera detecting wheels

#This application claims priority to European Patent Application No. 20 176 003.0, filed on May 22, 2020, the entirety of which is incorporated herein by reference.

#The present disclosed subject matter relates to a method for detecting wheels of a vehicle on a road, which wheels are visible on a side of the vehicle.

#Detecting and counting the number of wheels visible on a vehicle side allows, e.g., to deduce the presence of axles and their number and is an important task in ITS (Intelligence Transportation Systems) and vehicle tolling applications when vehicles shall be monitored, directed or tolled depending on the number of wheels or axles. For example, different toll rates may apply for vehicles with different numbers of wheels/axles such as passenger cars, light and heavy trucks. Accurately detecting and determining the number of wheels/axles of a vehicle is critical to avoid false charges.

#The disclosed subject matter will now be described by means of exemplary embodiments thereof with reference to the enclosed drawings, in which show:

#FIG. 1 an apparatus performing the method of the disclosed subject matter in a schematic perspective view;

#FIG. 2a a source 2D image plane with a perspective-distorted source 2D image recorded with the apparatus of FIG. 1;

#FIG. 1 shows an apparatus 1 for detecting wheels 2 of a vehicle 3 on a road 4. The wheels 2 of the vehicle 3 are visible on the left or right lateral side 5 of the vehicle 3. The road 4 has a direction of travel 6. During the detection of its wheels 2 the vehicle 3 is usually passing a section 7 of the road 4 in the direction 6, but the vehicle 3 could also be at rest within the section 7.

#The apparatus 1 comprises a vehicle classification sensor 8 mounted, e.g., above or besides the road 4 on a gantry 9 spanning the road 4. The position of the road 4 and the position PVCS of the vehicle classification sensor 8 within a given coordinate system 10 are known.

#The vehicle classification sensor 8 can be a stereoscopic camera, a radar scanner, a laser scanner or generally any sensor which is capable of recording a 3D representation of the vehicle 3 in the section 7 from its sensor position PVCS. If the classification sensor 8 is a laser scanner, it projects, e.g., a fan 11 of light rays onto the road 4. From reflections of the light rays of the fan 11 the shape of the vehicle 3 can be recorded in the sensor 8, e.g., by time-of-flight or interference measurements on the projected and reflected light rays. When only moving vehicles 3 shall be scanned the light ray fan 11 can be kept stationary, and the pass of the vehicle 3 yields a sequence of scan lines 12 the entirety of which form a 3D representation 13 of at least a part of the vehicle 3 in the coordinate system 10, which part comprises a portion of the vehicle side 5. When also stationary vehicles 3 shall be scanned, the light ray fan 11 can be swept, e.g., in the direction of travel 6, to scan a vehicle 3 at rest in the section 7. The 3D representation 13 may be recorded by a reconstruction of several subsequent measurements performed for several vehicle positions and/or several subsequent measurement times, e.g., by transforming, merging, stitching, interpolating or extrapolating points measured by the vehicle classification sensor 8.