DE102011016217A1 - Method for warning driver of passenger car before rear end collision, involves projecting path of image, and warning driver of vehicle based on image with projected path, where path is determined based on velocity and steering angle - Google Patents

Abstract

The method involves taking an image of a surrounding area (4) behind a motor vehicle (1) using an optical camera (3) of a camera system (2). A path of the captured image is determined based on velocity (V) and steering angle (theta) of the vehicle, where the path is a distance traveled by the vehicle. The path of the image is projected, and a driver of the vehicle is warned based on the image with the projected path. A warning relevant region is defined along the projected path of the image, where width of the region corresponds to width of the vehicle in the image. The optical camera is designed as a complementary metal-oxide-semiconductor (CMOS) camera or a charge coupled device (CCD) camera. An independent claim is also included for a camera system for a motor vehicle.

Description

The invention relates to a method for warning a driver of a motor vehicle of a danger of collision by means of a camera system of the motor vehicle. An image of a surrounding area behind the motor vehicle is taken, namely by means of a camera of the camera system. The driver is then warned based on the captured image. The invention also relates to a camera system which is designed to carry out such a method, as well as to a motor vehicle with such a camera system.

In the present case, the interest is directed in particular - but not exclusively - to a camera system with a rearward camera, by means of which the driver can be warned of the danger of a rear-end collision from behind. Such camera systems are already state of the art. As is known, a camera is used which detects the surroundings behind the motor vehicle. Based on the captured images, the risk of collision is then estimated and the driver is warned or not. Namely, other vehicles traveling behind the own motor vehicle can be recognized in the captured images, and the risk of a collision can be estimated depending on a relative position of a vehicle recognized in the image with respect to the own motor vehicle.

As is known, road markings-in particular longitudinal markings-can be detected in the recorded image. Then, in the warning of the driver, the relative position of another vehicle recognized in the image with respect to the detected road markings in the image can also be taken into account. At this state of the art, however, is to be regarded as a disadvantage of the fact that road markings are not always available and in the absence of road markings warning the driver can be done only insufficient.

It may also happen that in the recorded image, although another vehicle is detected, which is particularly close to its own motor vehicle, but driving on an adjacent lane. In this case, a warning is not mandatory; However, the camera system must be able to detect such a situation even in the absence of the road markings. So unnecessary warnings should also be prevented.

It is an object of the invention to provide a solution, as in a method of the type mentioned above, the driver particularly reliable using the camera system before dangerous driving situations - especially from the risk of rear-end collision from behind - can be warned.

This object is achieved by a method with the features according to claim 1, by a camera system with the features according to claim 13 and by a motor vehicle having the features of claim 14. Advantageous embodiments of the invention are the subject of the dependent claims, the description and the figures.

In a method according to the invention, a driver of a motor vehicle is warned of a risk of collision - in particular of the risk of rear-end collision from behind - with the aid of a camera system of the motor vehicle. An optical camera of the camera system captures an image of an environmental area behind the motor vehicle. The camera system warns the driver based on the captured image. The camera system determines a path traveled by the motor vehicle as a function of a speed and of a steering angle of the motor vehicle. The trajectory is then projected onto the captured image, and the driver is warned of the image with the projected path.

In other words, a core idea of the invention thus consists in determining a traversed path-that is, a trajectory traveled-of the motor vehicle, as a function of at least two specific parameters of the motor vehicle: the speed, on the one hand, and the steering angle, on the other. The steering angle can be measured by means of a suitable steering angle sensor; If such a sensor is not present, then the steering angle can also be determined from measured values of a yaw rate sensor by means of a difference method, or else from measured values of another existing direction sensor. The determined path is then linked to the image of the surrounding area behind the motor vehicle, so that the camera system can compare the positions of imaged objects with respect to the covered path of the own motor vehicle. The driver is warned based on the projection of the traveled path on the captured image. In this way, the risk of collision or the risk of an accident can be estimated particularly reliably and precisely, and the driver can be reliably warned of the risk of collision. In particular, vehicles can be detected that actually follow their own motor vehicle and thus represent a potential danger. The risk of a collision with other vehicles can also be estimated with the method according to the invention particularly true to reality if no markings - especially no longitudinal markings - on the road are applied. Not least, the two parameters speed and steering angle enable a particularly precise and highly accurate determination of the distance covered.

The optical camera is preferably a video camera, which at predetermined time intervals - for example, at intervals of 33 ms - can record frames (frames). The path traveled can also be determined continuously, so that, for example, the distance traveled is updated and projected for each recorded image. In this way, it is possible to check the risk of collision continuously or continuously, so that the driver can be warned quickly. The speed and the steering angle may also be updated at times different from the timings of the image capture - the image rate of the image sensor may be different from the frequency of the measurement of the speed and the steering angle.

The camera can be arranged in a rear region of the motor vehicle, for example on a tailgate or a rear bumper or on a rear window of the motor vehicle.

The camera can be a CMOS camera (Complementary Metal Oxide Semiconductor) or a CCD camera (Charge Coupled Device).

In the recorded image, an external vehicle object, in particular an adjacent vehicle, can be detected. The risk of collision can then be estimated as a function of a relative position of the object with respect to the projected path in the image. This means that the driver can be warned in relation to the projected path depending on this relative position of the detected object. This embodiment has the advantage that false warnings or unnecessary warnings can be prevented. Namely, another vehicle located outside the traveled path presents no or less danger of collision than a vehicle on the covered path.

It can be defined in the captured image a relevant area for the warning, namely along the projected path. This relevant area thus extends parallel to the projected path and may have such a width which corresponds at least to the width of the motor vehicle in the image. Then the warning of the driver can take place on the assumption that, on the one hand, the object (adjacent vehicle) is within the relevant area - for example along the path traveled - and, on the other hand, a distance between the object and the motor vehicle is below a predetermined limit value. The warning is preferably already made when the object is only partially within the relevant area or projects into the relevant area. Especially then there is the risk of rear-end collision from behind, and the driver can be warned against this collision hazard. In this way, the driver is informed that a sudden deceleration of the own motor vehicle can lead to a rear-end collision and thus to a collision. The risk of an accident is thus reduced to a minimum. The system can also be used to prepare the vehicle for an impending accident, e.g. For example, the system may trigger the pretension of a seat belt if the system recognizes that an accident is imminent and unavoidable.

The driver can be warned by the camera system in various ways: For example, an audible warning can be issued, by means of which the driver is warned acoustically. Additionally or alternatively, a haptic device - such as a vibration actuator - can be controlled, by means of which the driver is warned by haptic means. Additionally or alternatively, the recorded image can be displayed together with the projected path on an optical display device - such as a display and / or a rear-view mirror - so that the driver is visually warned of a possible collision hazard. With the above methods, the driver can be warned particularly reliable; In particular, the display of the image on the display device can be combined with the acoustic and / or haptic warning, so that the driver on the one hand gets an image of the environment behind the motor vehicle displayed on the display and on the other hand also warned by an audible and / or haptic warning ,

As already stated, the distance covered is determined as a function of the speed and the steering angle of the motor vehicle. In principle, the most varied methods for determining the distance traveled are conceivable here, and the present invention is not limited to a specific method. However, it proves to be particularly advantageous if the covered path of the motor vehicle is determined according to the following description:
During the driving of the motor vehicle, a plurality of values of the speed as well as associated values of the steering angle of the motor vehicle can be recorded at predetermined time intervals. Thus, it is possible to continuously record discrete values of the speed on the one hand and of the steering angle on the other hand, namely while the motor vehicle is traveling. Then there is a temporal sequence or sequence of value pairs from values of speed on the one hand and values of the steering angle on the other. Advantageously, the path traveled can be calculated in sections; It is possible to calculate sections of the path traveled in each case as a function of two adjacent value pairs, ie, one after the other in temporal succession. Preferably, the individual sections of the path are calculated exclusively as a function of the respective adjacent value pairs. By "breaking down" the path covered into individual sections, this path can be determined with little technical effort. Namely, only the individual sections need to be calculated, namely in each case as a function of two values of the speed and of two associated values of the steering angle. Such calculation of the traveled path is also particularly accurate. Namely, the relationship holds that the larger the frequency of recording the values of the speed and the steering angle, the higher the resolution of the detected path.

During the drive of the motor vehicle, therefore, a chronological sequence of values of the speed and associated values of the steering angle can be recorded. Each value of the speed is thus assigned a value of the steering angle in time. This means that at a certain time, a value of the steering angle and a value of the speed are detected. If a new value pair is recorded from a new value of the speed and a new value of the steering angle, then a new portion of the traveled path can be calculated between the current position of the motor vehicle on the one hand and the past (last) position of the motor vehicle on the other hand Value pair was recorded. As part of the calculation of the new section, this past position of the motor vehicle with respect to the current position is determined, and these two points are connected together, so that a new section of the path covered is created.

In principle, the time intervals, in which the values of the speed and the steering angle are recorded, can be arbitrary. These time intervals can also vary, so that the values are recorded at irregular time intervals. As already stated, however, it proves to be advantageous if these values are recorded at a relatively high frequency. In that case, the path traveled can namely be determined with a particularly high resolution and thus also with particular precision. In a preferred embodiment, the values of the speed and of the steering angle are recorded with a frequency, in particular a constant frequency, which is equal to the frequency of the image acquisition by the camera. With each frame is then an updated path for assessing the driving situation available, and the driver can be warned very quickly and early.

The values of the speed and the steering angle can be recorded, for example, at intervals from a value range of 20 ms to 40 ms, in particular at intervals of 33 ms (intervals between the individual images of the camera).

The computational effort when calculating individual sections of the path can then be reduced to a minimum if the individual sections are calculated as ring-segment-shaped sections, ie sections of a ring. The respective radius can be determined as a function of the respective adjacent values of the steering angle. Here, for example, an average value of the steering angle can be calculated from the two successive values of the steering angle. From this mean value of the steering angle then the respective radius and thus also the curvature of the section can be calculated. The position of the center - that is the center - of the ring relative to the motor vehicle may depend on known vehicle data - such as a distance between the front and rear vehicle axles and a distance between the left and right wheels of the same axle - as well as the average the steering angle are determined. In principle, any methods can be used which allow the calculation of the steering radius from the mean value of the steering angle. In order to further reduce computational effort, in one embodiment, rather than as annular segments, the sections may also be calculated as linear segments. Here, the direction of the linear portions may be calculated on the basis of the steering angle of the motor vehicle, while the length of the linear portions may be calculated on the basis of the speed.

If the radius and thus the curvature of the single section of the path is known, then only the length of this section needs to be calculated. For the calculation of the length, a speed of the motor vehicle which is constant for this section is preferably assumed. This speed can be calculated as an average of the values of the speed of the adjacent value pairs. The length of each section can thus be calculated without much computational effort, namely also taking into account the time interval in which the two adjacent value pairs were recorded.

When determining the distance traveled, position data of the motor vehicle can also be used be taken into account, which are provided by means of a navigation system of the motor vehicle. For example, the position data of the motor vehicle for the calculation of said steering radius can be taken into account in determining the curvature of a single section of the path. By taking into account the position data, the path traveled can be determined even more accurately or the path can be made plausible with the position data of the motor vehicle.

In addition to the above-described function of warning the driver of rear-end collisions from the rear, the camera system can also take on additional functions: For example, it is possible to use the image with the projected path to detect an own vehicle overtaking another vehicle. An other vehicle recognized in the captured image may be classified as a passing vehicle if the detected vehicle is adjacent to the projected path and the distance between the two vehicles is decreased. In this case, a warning or information may be issued that an overtaking vehicle has been detected. The driver is thus informed about this overtaking vehicle and the risk of a collision is minimal.

The camera system can also be used as a lane departure warning. Based on the captured image including the projected path can be seen that the motor vehicle leaves its lane. This can be recognized by the course of the path covered without much effort. In this embodiment, the information may also be taken into account if the driver has activated a turn signal or not. If leaving the lane without an activated turn signal is detected, the driver can be warned.

It is thus also possible to detect an overtaking maneuver of the motor vehicle on the basis of the image with the projected path. In this case, the path traveled also has a specific course.

The invention also relates to a camera system for a motor vehicle, namely for warning a driver of the motor vehicle of a risk of collision. The camera system comprises a camera which is designed to record an image of a surrounding area behind the motor vehicle. Means are provided for alerting the driver based on the captured image. The camera system can determine a path traveled by the motor vehicle as a function of a speed and of a steering angle of the motor vehicle. In the camera system, the path is projected onto the recorded image. The camera system can warn the driver based on the image with the projected path.

A motor vehicle according to the invention includes a camera system according to the invention.

The preferred embodiments presented with reference to the method according to the invention and their advantages apply correspondingly to the camera system according to the invention and to the motor vehicle according to the invention.

Further features of the invention will become apparent from the claims, the figures and the description of the figures. All the features and feature combinations mentioned above in the description as well as the features and feature combinations mentioned below in the description of the figures and / or shown alone in the figures can be used not only in the respectively indicated combination but also in other combinations or alone.

The invention will now be described with reference to individual preferred embodiments, as well as with reference to the accompanying drawings.

Show it:

1 a schematic representation of a motor vehicle with a camera system according to an embodiment of the invention;

2 in a schematic representation of a side view of the motor vehicle according to 1 , wherein behind the motor vehicle another vehicle is located;

3 an exemplary image of a surrounding area behind the motor vehicle;

4 a schematic plan view of the motor vehicle, the determination of a path covered according to a method according to an embodiment of the invention is explained in more detail;

5 in abstract representation, two different positions of the motor vehicle, wherein the calculation of a single section of the path is explained in detail;

6 a schematic plan view of the motor vehicle, the determination of the path covered is explained in more detail; and

7 an exemplary image of the surrounding area behind the motor vehicle with a projected path covered the motor vehicle

An in 1 illustrated motor vehicle 1 According to one embodiment of the invention is a passenger car. The car 1 includes a camera system 2 which is used to warn a driver of the motor vehicle 1 is formed before dangerous driving situations or against a risk of collision. The camera system 2 includes an optical camera 3 which is, for example, a CMOS camera or a CCD camera. The camera 3 is an optical device that detects light in the visible spectral range and thus optical images of the environment of the motor vehicle 1 can capture. The camera 3 is in the rear of the vehicle 1 arranged and aligned in the vehicle longitudinal direction to the rear. The camera 3 can, for example, on a rear bumper of the motor vehicle 1 or be arranged on a tailgate.

The camera 3 captures images of a surrounding area 4 behind the motor vehicle 1 , The captured images are then sent to a computing device 5 of the camera system 2 transmitted, which processes the captured images. The computing device 5 also receives data with the respective instantaneous values of a speed v of the motor vehicle, as well as a current steering angle O of the motor vehicle 1 ,

This information can be the computing device 5 for example, on a communication bus of the motor vehicle 1 tapping, such as the CAN bus. The computing device 5 can with an optical display device 6 and / or with an acoustic output device 7 and / or with a haptic device 8th be electrically coupled. The display device 6 may be, for example, a display, which is arranged in the field of vision of the driver. The haptic device 8th For example, it may be a vibration actuator by means of which the driver can be warned by haptic means. The acoustic output device 7 can be a speaker.

Optionally, the computing device 5 also receive position data P, which by means of a navigation system of the motor vehicle 1 to be provided. These position data P characterize the respective current global or geographical position of the motor vehicle 1 ,

As already stated, the camera system is used 2 to warn the driver of dangerous driving situations, namely in particular from rear-end collisions from behind. In 2 is a driving situation with the motor vehicle 1 shown. Shown is a side view of the motor vehicle 1 with the camera arranged on it 3 which the surrounding area 4 behind the motor vehicle 1 detected. The car 1 is on a street 9 and moves forward, namely according to the arrow 10 , Behind the own motor vehicle 1 there is another vehicle 11 which is a potential hazard to the motor vehicle 1 represents. The other vehicle 11 is within the detection range of the camera 3 and is thus shown in the recorded images.

In 3 is an exemplary picture 12 shown as it is through the camera 3 is recorded. How out 3 is apparent in the picture 12 the other vehicle 11 ' pictured, also on the street 9 ' is driven. A special challenge now is based on the pictures taken 12 determine if the additional vehicle 11 actually a danger to your own motor vehicle 1 represents or not and whether the driver of the motor vehicle 1 actually be warned of a risk of rear-end collision or not. This is in the captured image 12 searched for a dangerous area, ie after a relevant area for the warning of the driver area.

In order to determine the said relevant area, the computing device calculates 5 first one by the motor vehicle 1 covered path 13 as he exemplifies in

4 is shown. The path covered 13 is a trajectory along which the motor vehicle 1 was driven. To calculate the path 13 detects the computing device 5 continuously discrete values of the speed v on the one hand and the steering angle O on the other hand. Thus, a time sequence of value pairs of values of the speed v on the one hand and values of the steering angle O on the other hand is recorded. The respective value pairs, for example, all 33 ms, namely according to the frequency of the image acquisition by the camera 3 , Each value pair is a position of the motor vehicle 1 assigned. In 4 are three past positions 14 of the motor vehicle 1 shown as well as a current position 15 , In the past positions 14 In each case, a value pair was recorded from a value of the speed v _-1 , v _-2 ,..., v _-n and a value of the steering angle Θ _-1 , Θ _-2 ,..., Θ _-n .

In 4 t denotes the time intervals in which the value pairs are recorded.

For the determination of the traveled path 13 is thus a temporal sequence of value pairs v, Θ available. These values can now be past positions 14 in relation to the current position 15 of the motor vehicle 1 be calculated. This becomes the path 13 calculated in sections, ie the path 13 will be in a variety of sections 13a . 13b . 13c divided, which are calculated separately from each other. Are the past positions 14 in relation to the current position 15 known, so can these positions 14 to a complete path 13 be summarized.

Referring to 5 will now be determining the past position 14 of the motor vehicle 1 at a time t _-1 with respect to the current position 15 at the current time t ₀ explained in more detail. At the current time t ₀ , the value pair is detected from the current value of the speed v ₀ and the current value of the steering angle Θ ₀ , while at time t _-1 the value pair of the value of the speed v _-1 and the value of the steering angle Θ _-1 was recorded. The section 13c of the path 13 is calculated as a ring-segment shaped section, which minimizes the computational effort. The annular section 13c is uniquely defined by its length d ₀ and by a radius r ₀ . The radius r ₀ is calculated as a function of a mean value of the steering angle Θ, which is calculated from the two values Θ ₀ and Θ _-1 . From the mean value of the steering angle O, the radius r ₀ can be calculated, which is the steering radius of the motor vehicle 1 corresponds to the given steering angle. In this case, other vehicle parameters can be taken into account, namely in particular a distance between the two axes and a distance between two wheels of the same axis. Methods for calculating the steering radius r ₀ as a function of the steering angle O are known in principle; Any method can be used here.

If the radius r _{0 is} known, then only the length d _{0 of} the section still needs 13c to be calculated. This length d ₀ is calculated as a function of an average of the values of the velocity v ₀ and v _-1 . This mean can be easily compared with the time interval t between the two positions 14 . 15 be multiplied.

• – (x₀, y₀): die aktuelle Position des Kraftfahrzeugs 1; und
• – (x_–1, y_–1), (x_–2, y_–2), (x_–n, y_–n): die vergangenen Positionen 14 des Kraftfahrzeugs 1 zu jeweiligen Zeitpunkten t_–1 bzw. t_–2 bzw. t_–n

As in 6 In this way, all past positions can be represented 14 in relation to the current position 15 of the motor vehicle 1 be calculated. In a coordinate system of the motor vehicle 1 describe:

• - (x ₀ , y ₀ ): the current position of the motor vehicle 1 ; and
• - (x _-1 , y _-1 ), (x _-2 , y _-2 ), (x _-n , y _-n ): the past positions 14 of the motor vehicle 1 at respective times t _-1 and t _-2 and t _-n, respectively

Are the past positions 14 in the coordinate system of the motor vehicle 1 known, so can the entire path 13 in the recorded image 12 the camera 3 be projected, as in 6 based on the arrow 16 is indicated. In 6 is an example picture on the right side 12 the camera 3 represented, namely with the projected path covered 13 ' , In this projection, the coordinate values (x, y) of the past positions become 14 in coordinate values (u, z) of the image 12 converted, so the path 13 ' is superimposed realistically. This will be the position of the camera 3 on the motor vehicle 1 considered.

In 7 is an exemplary picture 12 shown as it is with the camera 3 was recorded. Pictured is the other vehicle 11 ' as well as the path covered 13 ' of the motor vehicle 1 who is now on the picture 12 is projected. Depending on the projected path 13 ' Now the above range is determined, which is relevant for the warning of the driver. This area extends parallel along the projected path 13 ' and is by two boundary lines 18 . 19 laterally limited. The relevant area has such a width 17 which is at least the width of the motor vehicle 1 in the picture corresponds and, for example, a little larger than the width of the motor vehicle 1 is. The actual path covered 13 ' is located in the middle between the two boundary lines 18 . 19 , which according to the width 17 spaced apart from each other. The area between the boundary lines 18 . 19 So the area within the width 17 , thus defines the above-mentioned relevant area or dangerous area, which is relevant for the warning of the driver. The computing device 5 Now check if the other motor vehicle 11 ' located in this relevant area or not. The warning can then be issued when the motor vehicle 11 ' within this range - that is, within the width 17 - is located and the distance between the two motor vehicles 1 . 11 is less than a predetermined limit.

So the camera system serves 2 to warn the driver of rear-end collisions from behind. The camera system 2 However, it can also be used for other functions: for example, overtaking maneuvers of one's own motor vehicle 1 and / or overtaking vehicles are detected. In addition or alternatively, using the camera system 2 leaving a lane by the motor vehicle 1 be detected, namely on the basis of the image 12 with the projected path 13 ,

Claims (14)

Method for warning a driver of a motor vehicle ( 1 ) against a danger of collision by means of a camera system ( 2 ) of the motor vehicle ( 1 ) by - taking a picture ( 12 ) of a surrounding area ( 4 ) behind the motor vehicle ( 1 ) by means of a camera ( 3 ) of the camera system ( 2 ) and Warning the driver based on the captured image ( 12 ), characterized by - determining by the motor vehicle ( 1 ) traveled path ( 13 ) as a function of a speed (v) and of a steering angle (Θ) of the motor vehicle ( 1 ), - project the path ( 13 ) on the recorded image ( 12 ) and - warning the driver based on the image ( 12 ) with the projected path ( 13 ' ). Method according to Claim 1, characterized by - detecting an object external of the vehicle ( 11 ), in particular of a neighboring vehicle, in the image ( 12 ) and - warning the driver in dependence on a relative position of the object ( 11 ) in relation to the path ( 13 ' ) in the picture ( 12 ). Method according to claim 1 or 2, characterized in that along the projected path ( 13 ' ) in the picture ( 12 ) a warning area ( 18 . 19 ) with such a width ( 17 ) is defined, the at least one width of the motor vehicle ( 1 ) in the picture ( 12 ) and the warning of the driver then takes place when the object ( 11 ) within the relevant area ( 18 . 19 ) and a distance of the object ( 11 ) to the motor vehicle ( 1 ) is below a predetermined limit. Method according to one of the preceding claims, characterized in that warning the driver includes: - emitting an audible warning signal and / or - driving a haptic device ( 8th ) and / or - displaying the image ( 12 ) with the projected path ( 13 ' ) on an optical display device ( 6 ). Method according to one of the preceding claims, characterized in that the step of determining the path traveled ( 13 ) comprises the following steps: - recording values of the speed (v) and associated values of the steering angle (Θ) of the motor vehicle ( 1 ) at predetermined time intervals (t), so that there is a time sequence of value pairs (v, Θ) of values of the speed (v) on the one hand and values of the steering angle (Θ) on the other hand, and - calculating the distance traveled ( 13 ) in sections, respectively, in particular exclusively, as a function of two adjacent value pairs (v, Θ). Method according to Claim 5, characterized in that the values of the speed (v) and of the steering angle (Θ) are recorded at a frequency which is equal to a frequency of the image recording of the camera ( 3 ). Method according to claim 5 or 6, characterized in that the individual sections ( 13a . 13b . 13c ) of the path ( 13 ) are calculated as ring-segment-shaped sections whose respective radius (r) is determined as a function of the values of the steering angle of the adjacent value pairs (v, Θ). Method according to one of claims 5 to 7, characterized in that for the calculation of a length (d) of a single section ( 13a . 13b . 13c ) one for this section ( 13a . 13b . 13c ) constant speed (v) of the motor vehicle ( 1 ), which is calculated as the mean value from the values of the velocity (v) of the neighboring value pairs (v, Θ). Method according to one of the preceding claims, characterized in that the determination of the traveled path ( 13 ) also taking into account by means of a navigation system of the motor vehicle ( 1 ) position data (P) is provided. Method according to one of the preceding claims, characterized in that based on the image ( 12 ) with the projected path ( 13 ' ) a motor vehicle ( 1 ) Recognizing adjacent vehicle is detected. Method according to one of the preceding claims, characterized in that based on the image ( 12 ) with the projected path ( 13 ' ) is detected that the motor vehicle ( 1 ) leaves a lane. Method according to one of the preceding claims, characterized in that based on the image ( 12 ) with the projected path ( 13 ) an overtaking maneuver of the motor vehicle ( 1 ) is detected. Camera system ( 2 ) for a motor vehicle ( 1 ), for warning a driver of the motor vehicle ( 1 ) against a collision hazard, with: - a camera ( 3 ) to take a picture ( 12 ) of a surrounding area ( 4 ) behind the motor vehicle ( 1 ) and - means ( 5 ) to warn the driver based on the captured image ( 12 ), characterized in that the camera system ( 2 ) is designed to - one by the motor vehicle ( 1 ) path ( 13 ) as a function of a speed (v) and of a steering angle (Θ) of the motor vehicle ( 1 ) to investigate, - the path ( 13 ) on the recorded image ( 12 ) and - the driver on the basis of the image ( 12 ) with the projected path ( 13 ' ) to warn. Motor vehicle ( 1 ) with a camera system ( 2 ) according to claim 13.

Images