EP2462005A1

EP2462005A1 - Method and device for distance-based debouncing of light characteristic changes

Info

Publication number: EP2462005A1
Application number: EP10745192A
Authority: EP
Inventors: Tobias Ehlgen; Petko Faber
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2009-08-07
Filing date: 2010-08-05
Publication date: 2012-06-13
Also published as: WO2011015625A1; CN102470793B; CN102470793A

Abstract

A method and device for debouncing light characteristic changes (105, 110, 111, 112) (for example street lamps, headlights of other vehicles) which are determined for a vehicle (102) from image data from a camera (31), wherein light characteristic changes are identified in a continuous sequence of images and a temporal distance between the light characteristic changes is ascertained, wherein a spatial distance between the light characteristic changes is ascertained with reference to the speed (37) of the vehicle and from the temporal distance between the light characteristic changes, and wherein the debouncing is based on the spatial distance.

Description

description

title

Method and device for distance-based debouncing of

Light characteristic change

State of the art

The invention relates to a method and a device for debouncing light characteristic changes (eg street lights, headlights of other vehicles), which are determined from image data of a camera for a vehicle, light characteristic changes are detected in a continuous image sequence and a time interval between the light characteristic changes is determined wherein, based on the speed of the vehicle and the time interval between the light characteristic changes, a spatial distance between the light characteristic change is determined and wherein the

Entprellung based on the spatial distance, according to the preamble of the independent claims.

For some time now, systems have been available on the market that actively control the lighting of a vehicle (headlamp). The aim of this measure is, firstly, to provide the driver with an optimal view of the lane in front of him or the relevant traffic situation and, second, (subordinate) the

Obstructing the visibility of other road users or persons (pedestrians, cyclists) or objects (buildings) as far as possible. In particular, systems that avoid dazzling oncoming vehicles and automatically switch between high beam and low beam have long been known.

Less well known are systems that adapt the light distribution especially to the current driving situation, or to the busy area. For the former For example, the active headlamp range control, or the cornering light as an example, for the second special light distributions such as motorway light or

City light. The light distribution of the city lights is characterized by a large, wide illumination characteristic, which makes it easier to see in intersections and illuminate the sidewalks, especially in urban (or inhabited)

Areas, more and more road users. Such a headlamp, which enables a variable luminous characteristic by appropriate control, is e.g. in WO 2006/116960 Al. In order to control such headlights correctly and automatically, the vehicle must have suitable means to determine the current driving situation or the whereabouts of the vehicle. For various purposes, such systems already exist which detect objects at night with the aid of a camera (e.g., Dual Purpose Camera (DPC) or Multi Purpose Camera (MPC)).

In US 6861809 B2, a detection of whether a vehicle is in a city area is carried out on the basis of the light intensity, which is measured by means of a special intensity sensor. At a high

Ambient light intensity assumes that the vehicle is located in an illuminated area and is therefore advisable in the city and an activation of the city lights. In particular, these ambient light sensors detect whether the light has an invisible AC voltage characteristic resulting from the mains frequency. This is described in claim 12. This method fails when LED street lights are used because they have their own or no clocking.

Furthermore, it is known by means of a speed threshold, d. H. when exceeding or falling below a certain speed with the vehicle to turn on or off a specific light distribution. In particular, when exceeding a certain speed is a possible

City light mode deactivated as described in claim 14.

In column 12, lines 17-21 of the Scripture is briefly the use of a

Mentioned imaging sensor, with the street lights are detected and the city light is activated when a certain number of street lamps be recognized within a certain time. This doctrine ignores whether a vehicle travels a long distance in this time span, and thus has already traveled a considerable distance in the city before the city light is activated. The mentioned at this point

Consideration of the vehicle speed refers to the previously described speed threshold, the overriding of which generally turns the city light off and standard light on.

Definitions

Known light distributions have, for example, the self-explanatory

Designations motorway light, bad weather light or city light. Under

Standard light is understood in the present case in most cases low beam or high beam.

A street lighting unit is a light source that illuminates the street and / or can be perceived by the camera of the vehicle. For the present invention, it must fulfill the purpose of being characteristic of an illuminated area. Typically street lamps or lanterns are meant, in particular those that have a

Illuminate a street or a street to a greater length and not only punctually at dangerous situations. Excluded are lights that do not primarily serve this purpose, especially those that are mobile, such. B. headlights of other vehicles. Also, light sources (active or passive) may be suitable for this purpose, such as. As traffic lights, other signal lights or reflection of

Describe. For example A frequent occurrence of traffic lights also indicates an illuminated and built-up area.

A light characteristic change is the change of a current light situation (light change), i. E. a change in strength, distribution, color, frequency,

Modulation (eg AC supply voltage) of the incident from the environment on the vehicle, or the detecting sensor (light sensor, camera) incident light. The characteristic can be determined by certain changes of light, for example by the change in the incidence of light when passing an oncoming vehicle, or the light change already described when passing by stationary objects, such as street lamps. Also, a characteristic may be characterized by more than one light change,

For example, a certain number of light changes within a certain time, or a sequence of different light colors.

A possible characteristic may be a light-dark change, such as. in the

Exit from a lighted tunnel; Another characteristic may be a dark-light-dark change, as in the passage past a

Street lamp. Such passing by can also be interpreted as 2 characteristic changes: dark-bright and then light-dark.

The term "stationarity" of objects is understood to mean their positional stability, that is to say the stationarity of the object is given if the object is a stationary object.The term "debouncing" is used here in such a way that it is correct

Hysteresis would have to. In Wikipedia, this is defined as follows: "Hysteresis characterizes a system whose output depends not only on the type of input, but also on the history of the input. The system proves this

Path dependence. "

Light characteristics change depending on occurrence of previous

Light characteristics change and states of a system affected thereby (e.g., current headlamp setting).

Classically, a debounce is known in a switch, which can generate several unwanted bounce pulses during the on and off. Against this, e.g. a Schmitt trigger, which has a hysteresis functionality, remedy by waiting after a last pulse for a defined time, if another impulse occurs. If not, the shift is considered complete and completed and will be issued / reported.

Another type of debouncing may be waiting for a certain number of

Events before an action is taken and debouncing is terminated.

Similarly, after a light characteristic change a certain time can be waited for example, to detect another (or n more) lantern (s), or a message (eg. For activating a special light distribution of a Headlamps). This "waiting for time" is well known in the art, therefore, the present invention is concerned with waiting and debarking by path or speed, as explained further below.

The terms territory and area are to be used synonymously in the present case, in particular a city area designates an urban area, equivalent to an area of an urban area or an inhabited area of an urban one

Surroundings, but not limited to city limits, or municipalities, but to localities or populated areas in general.

Disclosure of the invention

The method described or the device with the features of the independent claims, on the other hand, have the advantage that it can be more accurately assessed whether an area is illuminated and that these

Information is determined faster. Thus, it is proposed to consider a region as illuminated when at least two street lighting units occur within a certain distance. The average distance from street lamps in Germany is about 25 to 30m. So z. Legs

Threshold of 40 to 50m are chosen to the spacing of two

Classify street lights as typical of illuminated areas. A classification can only take place if more than two

Street lights are detected within a certain distance or based on the number of detected street lights, a probability for the existence of a lighted area is determined. Thus, firstly, the detection of the illumination of an area after a certain number of street lights and thus a certain distance traveled is determined and not only after a certain time when the lighting of an area may not be recognized until due to high speed already a considerable piece in the illuminated area was driven into it. Second, the lighting of an area can be detected even if the vehicle is traveling too slowly to detect a certain number of lanterns within a certain time. Conversely, if the street lights no longer appear at this distance from each other, the rating of the area can be considered to be withdrawn. Extracting the information to one

Street lighting unit can be seen in the image data is with

appropriate pattern recognition or other appropriate methods

Perform methods that are not described in detail in the context of this invention.

Advantageously, the spatial distance of the

Street lighting units are already determined from the image data of a single image. For this purpose, simplified geometric methods can be used, which measure the distance of the street lighting units. The evaluation of several images is not absolutely necessary here, but an iterative or additional evaluation of a sequence of images is also possible. Typically, current systems sometimes detect three lanterns in a single image in advance. The only image available can be data from a stereo camera, i. H. Strictly speaking, two or more pictures taken at the same time and of which one

Depth information of the objects (lanterns) can be deducted.

Advantageously, if, for example, several lanterns can not be seen in a single image, a continuous image sequence can be used to recognize the street lights, the time interval to the appearance of the previous street lighting unit being determined. Since the first detection time of a lantern of the local events, as well

Environmental influences, such. B. weathering, may be dependent, the detection is not always at the same distance from the (in front of the) vehicle instead. Thus, for example, the detection of the second (or further) lantern could be delayed such that the evaluation of the threshold value of the distance is not achieved and a region is incorrectly classified as lit or not lit. In order to solve this problem, the evaluation of the image information can take place in such a way that a lantern is only recognized as being detected at a certain point in the image or at a certain distance from the vehicle, although optical recognition would usually be possible earlier. As a result, the distance can be defined exactly in which a Lantern must lie in front of the vehicle and optionally from which distance in front of the at least one further lantern, the evaluation of whether an area is lit, takes place. This method can alternatively or additionally be performed by means of a temporally or rearwardly looking camera. In particular, the change between the illuminated and non-illuminated areas can be made plausible by means of the rear-facing camera together with the forward-facing camera or its image data.

It is advantageous to determine the spatial distance of the

Street lighting units from the determining time interval in the continuous image sequence, in which the speed of the vehicle is used to calculate this. In the simplest case, using the formula Distance between two lanterns = vehicle speed * time difference of

Appearance of the two lanterns this result determined. Thus, the distance traveled is the decisive evaluation variable. Optionally, in addition, a plausibility of the distance via other methods, such. B. geometric methods of

Image analysis, to be performed.

Advantageously, debouncing may take place, allowing more accurate detection of an illuminated area. In this case, in a first step, a first street lighting unit is detected, in a second step, a second, or depending on applicative request also an nth,

Detected street lighting unit, determined in a third step, the temporal or spatial distance of at least two street lighting units and tested in a fourth step, if this distance is below a fixed threshold. Alternatively, from the occurrence of the first

Street lighting unit also start a time or distance counter, which only checks on reaching the time or distance threshold, if at least one (or n-1) further street lighting unit (s) has occurred. The debouncing described is useful to a back and forth of the city lights too avoid, which could give the impression of a flickering or misleadingly a flare.

Comparable debouncing would make sense even if in the oncoming traffic between high beam and low beam back and forth would switch back and forth would happen directly after passing an oncoming vehicle, without checking if there are other subsequent vehicles, otherwise Toggling or flickering occurred.

It is conceivable, generally in light changes, regardless of their origin, to use the path as a size for debouncing for the light control function, instead of time.

Advantageously, this debouncing even when resetting the

Assessing whether an area is lit up, take place, d. H. when moving out of a lighted in an unlit area. This would be after the

Occurrence of a last street lighting unit waited a certain time or a certain distance and if no further (or nth)

Street lighting unit occurs within this distance, the evaluation performed as a non-illuminated area. When Entprellung by way can advantageously be quickly switched back to normal light, especially at high speed and thus a ride with city light and suboptimal illumination for out-of-town as small and short as possible.

Furthermore, a debouncing of the detection of a lighted area can also be carried out with other parameters than debouncing the

Detection of a non-illuminated area, eg. For example, an illuminated area can be classified after only two street lamps, but a non-illuminated area only after several missing street lamps. Advantageously, the change of said threshold or said

Distance dependent on environmental conditions, this threshold can be adjusted variably even while driving. Thus, the typical distance from street lights may change depending on the country you are driving, or situations like turning off every other lantern due to

Energy-saving measures in some communities. Last In this case, for example, the time as additional

Environmental parameters are taken into consideration, for example, from 22:00 to 06:00 clock the threshold for the lantern distance is doubled. If necessary, however, a limit can then be reached, according to which the area must be assessed objectively as no longer sufficiently illuminated, so that driving in urban areas with standard light should also be achieved.

It is advantageous that the information as to whether the area currently being used is an area in the city can be deduced from the degree of illumination of the roads, since there are appropriate regulations for such stationary lighting and many areas in many countries accordingly are expanded. Depending on the degree of illumination or the number of detected lanterns, a probability could be derived instead, which is an area in the city.

For example, the occurrence of a lantern may be a likelihood of

30% and from the appearance of three lanterns a probability of 100% for the health in a district speak.

Advantageously, in addition to the relationship between the illumination of a region and the condition in the city, further indications or

Parameters are taken into account in order to consolidate the thesis that the vehicle is located in the city. To do so, signals from other sensors can be evaluated and evaluated in a corresponding evaluation logic or software. In particular, the map data, z. B. one

Navigation systems are ideal for easily reading information about an area in the city from its database. In this way, a correct condition of the vehicle could be determined even if in some (possibly the map known) regions the

Street lighting units (temporarily) are switched off.

On the other hand, along with the described procedure can

NichtVorhandensein, or a failure of the navigation system or a missing update of the map data can be compensated.

Furthermore, for example, the following criteria can speak for the fact that the vehicle is located in an illuminated area or an area in the city or whether activating the starting light is advised: the current Traffic density, the information of a brightness sensor, the illumination by non-stationary street lighting units, in particular the flowing traffic, the detection of pedestrians, lane and

Curve detection systems, information exchange via Car to X (C2X) or a driving profile of the vehicle, or the driving style of the driver. Information about the location or the traffic situation can be transmitted by other vehicles or, for example, traffic signs (C2X). The driving profile is expressed in addition to the speed, for example, in the steering angle, or the yaw rate from, for example, if this is large, so strong steering angles

This is an indication of a built-up area.

It is advantageous if only the street lighting units are taken into account, which are also characteristic of a local area. Thus, some exclusions of stationary lighting are conceivable, in particular the lighting of intersections, as they occur on national roads, railroad crossings, exits or junctions or

Advertising lighting. Likewise, it may be useful to consider only the street lights on a street side, in particular those on the lane side for performing the method. Optionally, by extended methods, a recognition of lanterns of cross streets take place, which are then not taken into account.

Advantageous is the application of the method of light control wherein a specific light distribution (eg city light) is performed in an area that is rated as in-town and activation of the standard light or other light distribution or deactivation of e.g. Town lights in an area outside, or when changing to an area out of town.

Optionally, instead or additionally, a sliding

Headlamp leveling, whereby the range of headlamps in a local area is reduced and increased in a non-urban area, possibly additionally taking into account other parameters such as oncoming traffic. Deactivating, for example, city lights can lead to the activation of standard light or other light distributions, for example those that were active prior to the activation of the city light or another that appears appropriate on the basis of the current driving situation. About the light control In addition, alternatively or additionally, the information about an area in the city for activating or deactivating other systems in

Vehicle or outside of the vehicle, such as biasing occupant protection systems, such as one

Pedestrian impact system as a collision with a pedestrian in one

Area is more likely to be in the city than outside.

It is advantageous that on the detection of a certain number of lanterns within a certain time or distance, a so-called pre-or post-function serves to verify the evaluation more accurate. In this pre- or post-run, or a further (external) Entprellung, it is checked whether the result of the evaluation, for example, whether an area is lit, whether an area is a city area or whether a specific light distribution should be activated during the pre - or follow-up phase exist, or remains valid. This makes it possible to assume with a higher degree of certainty that the rating is correct.

It is advantageous if the behavior of the vehicle does not change with respect to its light characteristic during very slow journeys. In a very slow ride, the driver would perceive a changing light characteristic, in particular a fade, as surprising and inappropriate. Therefore, it makes sense that at very slow driving the light characteristic is either not, or not switched in the direction of superimposing light characteristic. Depending on the application, if the speed threshold is exceeded again, the usual light behavior can be set. The

Deburring between the appearance of the lanterns on the basis of the path instead of the time to perform offers the advantage, in particular for slow cruises, that the city light may also be maintained for a long time. Advantageously, standard light can then generally be maintained, or on

Standard light will be switched if a specific

Speed threshold is exceeded. A typical threshold is, for example, 80 km / h, where it is assumed that the advantages of the city light can not be meaningfully used at such high speeds and a light characteristic is more suitable, the fast driving means facilitate optimal and wider (rather than wider) illumination. As a result, an activation of the city lights can be avoided even if it is a road, especially a highway outside town. For example, in some countries, the highways are illuminated. A distinction between an area in the city and such a lighted expressway may possibly be possible via the lantern distance.

Another embodiment of the method may be used to determine whether an area is illuminated in which there is a separate vehicle (102) by means of image data at least one camera (31) for the vehicle (102), for the purpose of light control (34 ) of the vehicle, wherein at least two (205) stationary street lighting units (105) are detected on the basis of the image data and an area is then rated as illuminated (206) if these at least two street lighting units (105) are spaced apart from each other within defined limits (107, 204).

Another specialization is characterized in that the spatial distance is determined from the image data of a single image.

Another alternative specialization is characterized in that an occurrence of further street lighting units (105, 111) in one

continuous image sequence is detected and the time interval to the occurrence of the previous street lighting unit (110) is determined.

Another specialization is characterized in that based on the speed of the vehicle (37) from the time interval of the spatial distance of these street lighting units is determined. Another specialization is characterized in that

Falling below a time threshold or a spatial distance threshold (204, 205) between the occurrence of a first (110) and a further (111) street lighting unit, an area is rated as illuminated (206). A further specialization is characterized in that, when a temporal threshold or a spatial distance threshold (107, 204) is exceeded, an area is evaluated as not illuminated from the occurrence of the first or last street lighting unit (112) without occurrence of a further street lighting unit (207, 113). becomes.

Another specialization is characterized in that the time threshold or the spatial distance threshold (204) due to

Environmental conditions is adjusted variably.

Another specialization is characterized by the assumption that the illuminated area is a city or area within a locality. Another specialization is characterized in that, in order to determine whether it is an area inside or outside a place linking the information whether the area is illuminated, is carried out with other sources of information. Another specialization is characterized in that only the

Street lighting units (105), which are characteristic of an inner-city area.

Another specialization is characterized in that the city light is activated when the area is used as a city center (111) and standard light when the

Area was determined as out of town (113, 114).

A further specialization is characterized in that an assessment with respect to the illuminated area or area in the city, or an activation of the city light (111) or standard light (114) takes place only when the evaluated

State for a certain time or distance (108, 211).

Another specialization is characterized in that

Slow cruising below a set speed threshold activated city light is not deactivated, although the area has since been designated as out of town.

Another specialization is characterized in that

Exceeding a defined speed threshold with the vehicle, the area is determined as not lit, and / or as out-of-town, or / and the city light is deactivated.

Another embodiment of the device can be used to determine whether an area is illuminated, in which a separate vehicle (102) is located, with first means or an interface for receiving image data (31) and second means for detecting stationary

Street lighting units (32) or receiving this information are provided, said second means (32, 33) at least two stationary street lighting units in at least one image, or a sequence of images, recognize the image data of the first means and depending on the spatial

Evaluate distance (107) and the result or control (34) of

Headlamps (35) transmitted via an interface.

drawings

Embodiments of the invention are illustrated in the drawings and explained in more detail in the following description.

Show it

FIG. 1 shows the entry and exit of a vehicle in a lighted area, FIG. 2 shows a flowchart of an embodiment of the method,

Figure 3 is a block diagram of a possible system arrangement with a camera.

FIG. 1 shows a scene on a road 101, on which a vehicle 102 drives from below upwards into an illuminated area, which is characterized by lanterns 105. At the beginning, the vehicle travels with standard light 103, while the system travels from a distance 106 ahead of the first Lantern detects this at the time of the passage of the point 110. Also at a distance 106, this second lantern is detected at the point 111 in front of the second lantern, and at this point the city light 104 is switched on. After driving through the illuminated area, the system recognizes at point 112 the last lantern for the time being, although it does not yet know at this point that it is the last lantern. After no further lantern has been detected in a debounce distance 107, the area is no longer evaluated as illuminated at point 113. After a follow-up time or a follow-up path 108, the city light 104 is switched to the standard light 103 again at the point 114.

FIG. 2 shows the sequence of an embodiment of the method, wherein in an input situation 201 the area is assessed as not yet lit or is not evaluated at all. In step 202 initial values are set, for example the distance traveled or the path "X" between two lanterns to zero and the counter "n" for the number of lanterns is set to 1. In step 203, it is checked if the nth lantern has been detected. If this is not the case, it is checked in a step 204 whether the distance X has already exceeded a certain threshold. This threshold has already been described and gives a typical distance of two or n lanterns plus optionally one

Safety offset on. As long as this threshold has not been exceeded, step 203 continues with the detection of the nth lantern. If an nth lantern has been detected, in step 205 the lantern counter is incremented by one and a check is made as to whether the required number of detected lanterns has been reached. If this is not yet the case, the increment of the counter is retained and the procedure continues with step 204. If the number of lanterns has been reached, then in step 206, an evaluation is made on an illuminated area. If the threshold has been exceeded in step 204, detection is performed on a non-illuminated area. Up to this point, FIG. 2 illustrates a possible realization of the basic concept of the invention.

In the following, a possible forward or backward function is described: The evaluation results of steps 206 and 207 are buffer-stored in a memory in a step 210. Then, from a step 211 a certain time or distance corresponding to the desired forward or Wait time waited for in step 212 a comparison between the stored and the currently available evaluation results

perform. If these results are the same, then in a step 213 the evaluation is carried out as urban area or non-urban area or the activation of the city light or the standard light. In case of inequality of

Results are carried out according to other evaluations, for example maintaining the current rating or the current light distribution. FIG. 3 is a schematic drawing of a multi or dual purpose

Camera 36, which receives the image data by means of a camera or optics 31 and by means of vehicle and object detection algorithms 32 extracts the lanterns from the image data. In a signal evaluation unit 33, the evaluation is now carried out with respect to the number of lanterns within a certain path or time. For this purpose, input signals 37, such. B. the

Vehicle speed or time taken into account. One

Headlight control unit (ASF control unit) 34 now controls the headlights 35, depending on the information as to whether the vehicle is located in the city. The components of the device may also be grouped separately, in particular, e.g. the camera optics 31 and outsourced over a

Interface be connected to the evaluation controller 36.

Claims

claims

1. Procedure

for debouncing at least two light characteristic changes (105, 110, 111, 112),

determined from the image data of at least one camera (31) for a vehicle (102),

- Wherein the at least two light characteristic changes are detected in a continuous image sequence of the at least one camera

- And a time interval between the at least two

Light characteristic changes is determined,

characterized in that

by the speed of the vehicle (37)

- From the time interval between the at least two

Light characteristic change

a spatial distance (107) of the at least two light characteristic changes is determined,

- The Entprellung based on the spatial distance (107).

2. The method according to claim 1, characterized in that

- When the spatial distance between a first (110) and another (111) or n-th light characteristic change

exceeds or falls short of a spatial distance threshold (204, 205),

- Then the Entprellung terminated and a rating (206) is performed.

3. The method according to claim 1 or 2, characterized in that

- The light characteristic changes caused by the passage of the vehicle to stationary objects or light sources (105).

4. The method according to claim 2 or 3, characterized in that

- the assessment assesses whether the area in which the vehicle is located

- is lit and / or

- it concerns a city area or area within a town and / or - City light is activated when the area is designated as inner city (111) and standard light when the area is designated as outlying (113, 114).

5. The method according to claim 2 to 4, characterized in that

- The spatial distance threshold (204) is variably adjusted due to environmental conditions.

6. The method according to claim 1 or 2, characterized in that

- at a known distance of the objects or light sources,

by the speed of the vehicle (37)

- the speed or stationarity (v = 0) of the objects or light sources is determined.

7. The method according to claim 1 or 2, characterized in that

the light characteristic changes are caused by moving objects or light sources and

- Is determined by the speeds of the objects or light sources of / their spatial (r) distance.

8. The method according to claim 1 or 2, characterized in that

- In addition to or instead of the determination of the spatial distance of the at least two light characteristic changes and / or debouncing the path between the Lichtcharakterisitkwechseln

- Debouncing based on the distance traveled by the own vehicle (37), takes place after the light characteristic change.

9. The method according to claim 8, characterized in that

the light characteristic changes are caused by oncoming vehicles,

- and after debouncing based on the distance traveled by one's own

Vehicle (37)

- after a last light characteristic change

- A change in the headlight control takes place.

10. The method according to claim 4 and 8, characterized in that an assessment with respect to the illuminated area or area in the city, or activation of the city light (111) or standard light (114) only then takes place,

- if the evaluated state lasts a certain distance traveled (108, 211) long.

11. The method according to any one of the preceding claims, characterized in that

- the debouncing is based on the speed instead of or in addition to the distance.

12. The method according to any one of the preceding claims, characterized in that

- the rating is communicated to other systems (ABS, ESP, airbag (control unit), etc.) for further evaluation.

13. Device

for debouncing at least two light characteristic changes (105, 110, 111,

112)

with first means or an interface for receiving image data (31) for a vehicle (102) and

second means provided for detecting light characteristic changes (32) or receiving this information,

characterized in that

- These second means (32, 33) at least two light characteristic change

in at least one image, or a sequence of images, which recognize image data of the first means and

by the speed of the vehicle (37)

- From the time interval between the at least two

Light characteristic change

determine a spatial distance (107) of the at least two light characteristic changes,

- The Entprellung based on the spatial distance (107).