DE102012019234A1 - Method for detecting supporting unit mounted on trailer coupling, involves creating image of rear surroundings of motor vehicle, and selecting area of trailer coupling in camera image, where area of camera image is transformed - Google Patents

Abstract

The method involves creating an image of the rear surroundings of a motor vehicle, and selecting an area (3) of a trailer coupling (1) in the camera image. The area of the camera image is transformed into a binary image with two binary value pixels using a predetermined threshold value. The area transformed into the binary value is searched for a continuous supporting structure. A mounted supporting unit (2) is detected to locate the continuous supporting structure in the transformed area. Independent claims are also included for the following: (1) a device for performing the detection method; and (2) a trailer maneuvering assistant for maneuvering a locomotive set of a motor vehicle and a non-steerable trailer.

Description

The invention relates to a method for detecting a vehicle arranged at the rear of a vehicle carrier device, in particular a bicycle trailer, according to the preamble of claim 1, a corresponding device according to the preamble of claim 8, and a Anhängerrangierassistenten according to the preamble of claim 10.

In today's motor vehicles, driver assistants are increasingly being used to assist the driver of the vehicle in solving certain driving tasks or in mastering critical driving situations. Such driver assistance systems are, for example, parking aids or automatic parking systems, emergency braking systems, adaptive speed controls or lane change systems or lane keeping systems.

Also for teams, consisting of serving as a towing vehicle and a trailer coupled to the towing vehicle by means of a trailer hitch trailers exist meanwhile driver assistants who serve to support the driver of a team.

Thus, trailer or trailer stabilizers are known, which ensures a stabilization of the combination in the forward drive and counteracts a breakout of the trailer and a resulting spin of the team.

In general, maneuvering a team is difficult, with the difficulty of reversing increasing as the trailer tends to skid. In particular, for inexperienced people, the turning of the trailer of a team with a change in the steering angle of the towing vehicle during a reverse is not readily traceable, so that when parking a team often due to a driver unfavorably selected steering angle to a collision with an adjacent obstacle or comes to a buckling of the team. To aid the driver in maneuvering a team, especially in reverse, are Trailer Development Assistants in development, abbreviated to ARA.

In general, for the aforementioned driver assistants for teams, their functionality is only released if the trailer socket of the towing vehicle is occupied. In this case it is assumed that a trailer is coupled to the towing vehicle.

However, this conclusion, namely that when an occupant of the trailer socket a trailer is coupled to the vehicle, however, is not justified because carrier devices are in use, which are connected to the trailer hitch of a vehicle and which require an occupancy of the trailer socket to control their necessary lighting. Such carrier devices are arranged, for example, at the rear of the vehicle on the trailer hitch bike carrier, which have independent lighting and therefore occupy the trailer socket. Consequently, it can not be concluded from the assignment of the trailer socket on the coupling of a trailer and it must be determined whether a trailer is on the vehicle or whether the trailer hitch is covered with a carrier device, the determination should preferably be made automatically.

From the publication DE 10 2009 038 552 A1 For example, there is known a method and apparatus for detecting a trailer attached to a vehicle, wherein the trailer's articulation angle with respect to the trailer hitch is determined. Depending on the determined bending angle is detected whether a trailer is attached to the vehicle. In other words, changes in the ride of the vehicle, the bending angle, so a trailer is infected. A disadvantage is therefore that a change in the bending angle and thus a detection of the infected trailer is only possible when driving the vehicle.

Another way to detect a coupled trailer is in the document DE 10 2008 013 457 A1 in which a rear-facing camera captures a rearward image and compares that image with stored data representing a trailer. If the comparison is positive, then a trailer is attached to the vehicle. Critical here is the evaluation of the recorded image and the comparison with stored data, especially since trailers can have different shapes and dimensions.

• a) Prüfen, ob eine Anhängersteckdose des Fahrzeugs belegt ist;
• b) Prüfen, ob in einem zumindest näherungsweise durch die Position der gegebenenfalls montierten Trägereinrichtung begrenzten Entfernungsbereich von einem Referenzpunkt im Heckbereich des Fahrzeugs ein Objekt vorhanden ist; und
• c) Feststellen, dass der Anhänger angekuppelt ist, wenn in Schritt a) eine Belegung der Anhängersteckdose erkannt wird und wenn entweder in Schritt b) kein Objekt in dem Entfernungsbereich erkannt wird oder wenn in Schritt b) zwar ein Objekt in dem Entfernungsbereich erkannt, aber in einem zusätzlichem Schritt d) festgestellt wird, dass sich der Abstand zwischen dem Objekt und dem Referenzpunkt bei einer Geschwindigkeit des Fahrzeugs größer Null nach Maßgabe des Lenkwinkels oder des Lenkradwinkels des Fahrzeugs ändert.

From the publication DE 10 2004 057 220 A1 For example, another method of distinguishing between an attached trailer and a carrier mounted on the trailer hitch is known. The method comprises the following steps:

• a) Check whether a trailer socket of the vehicle is occupied;
• b) checking whether an object is present in a distance range bounded at least approximately by the position of the optionally mounted carrier device from a reference point in the rear region of the vehicle; and
• c) determining that the trailer is coupled, if in step a) an assignment of the Trailer socket is detected and if either in step b) no object is detected in the distance range or if in step b), although an object detected in the distance range, but in an additional step d) is determined that the distance between the object and the Reference point at a speed of the vehicle greater than zero in accordance with the steering angle or the steering wheel angle of the vehicle changes.

Again, it is necessary for the implementation of the method that the vehicle moves before a distinction between a trailer and a carrier device can be performed.

The invention is therefore based on the object of specifying a method and a device by means of which the presence of a carrier device mounted on the trailer coupling, such as a bicycle carrier, can be detected even when the vehicle is at a standstill.

This object is achieved by a method having the features of claim 1, by an apparatus having the features of claim 8, and by the use of the method in a Anhängerrangierassistent with the features of claim 10. Preferred embodiments of the method are the subject of the dependent claims.

• – Erstellen eines Bildes der rückwärtigen Umgebung des Kraftfahrzeugs,
• – Auswählen eines Bereichs um die Anhängerkupplung im Bild der Kamera,
• – Transformieren zumindest des ausgewählten Bereichs des Kamerabildes in ein binäres Bild mit Pixel eines ersten Binärwertes und Pixel eines zweiten Binärwertes unter Verwendung eines vorgegebenen Schwellenwertes,
• – Durchsuchen des in Binärwerte transformierten Bereichs nach mindestens einer zusammenhängenden Trägerstruktur der Trägereinrichtung, und
• – Detektion des Vorhandenseins einer montierten Trägereinrichtung beim Auffinden mindestens einer zusammenhängenden Trägerstruktur im ausgewählten transformierten Bereich.

The inventive method for detecting a mounted at the rear of a motor vehicle by means of a trailer hitch support means, wherein the motor vehicle is equipped with a rear-facing camera, comprises the following steps:

• - Creating an image of the rear environment of the motor vehicle,
• Selecting an area around the hitch in the image of the camera,
• Transforming at least the selected area of the camera image into a binary image having pixels of a first binary value and pixels of a second binary value using a predetermined threshold value,
• - Browsing the binaries transformed area for at least one contiguous support structure of the carrier device, and
• - Detecting the presence of a mounted carrier device when finding at least one contiguous support structure in the selected transformed area.

Since the load of the carrier device has to be dissipated to the trailer hitch, a coherent carrier structure of the carrier device must be able to be found in a predetermined region of the trailer hitch if a carrier device, such as a bicycle carrier, is mounted on the trailer hitch. If no coherent structure can be found in the area around the trailer hitch, the logical conclusion follows that no carrier device is mounted. If a coherent support structure is found, this implies the statement that a support device is mounted. This statement is independent of an assignment of the trailer socket and can also be taken at standstill of the vehicle. The black-and-white transformation of at least the area of the image in the vicinity of the trailer coupling makes it possible to find the coherent support structure simply as a black or white contiguous pixel structure.

More preferably, the area to be examined around the trailer coupling of the image of the camera has a rectangular shape, so that the transformed area to be examined forms a pixel matrix in binary values, for example with the values zero and one or with the values black and white.

Preferably, in the case of displaying the contiguous support structure of the carrier means by pixels of the first binary value, first the columns of the binarized transformed region in terms of the columnwise contiguous occurrence frequency of pixels of the first binary value and then the rows of the transformed region after line coincidences of the column-contiguous one Examining the pixels of the first binary value or examining first the lines of the transformed area after the line-connected frequency of occurrence of the pixels of the first binary value, and then the columns after column-like matches in the occurrence of the line-wise contiguous frequency of the pixels of the first binary value.

Preferably, in the case of representing the contiguous support structure of the carrier means by pixels of the second binary value, first the columns of the binarized transformed region in terms of the columnally contiguous frequency of occurrence of the pixels of the second binary value and then the rows of the transformed region after line coincidences of the column-contiguous Occurrence of the pixels of the time binary value is examined or the rows of the transformed area are first line-by-line according to the number of occurrences of the pixels of the second binary value and then the columns according to column-like matches in the occurrence of the lines contiguous frequency of pixels of the second binary value.

Further preferably, the line-connected frequency of occurrence of the pixels of the first and the second binary value is determined by a predetermined frequency limit AS, so that a number of pixels of the first or second binary value in a column or a row is not considered as coherent, if the number is less than the predetermined frequency limit and the number is considered contiguous if the frequency is equal to or greater than the predetermined frequency limit AS.

More preferably, the line-by-line masking of the columnar contiguous frequency or the columnar coverage of the line-wise contiguous frequency for adjacent rows or adjacent columns must be greater than or equal to a predetermined coverage threshold AZ to form a contiguous support structure.

Preferably, the values of black and white or zero and one are used as binary values.

Thus, for example, in the case of representing the binary values of black and white, and in the case of the column view and the display of the black pixel array, a column i has a contiguous number X> AS of black pixels, then in adjacent columns (i) 1) and i also each have a contiguous number of pixels, the arrangements in the columns (i-1) and i being able to be shifted from each other. Further, the number of contiguous pixels of the column (i-1) need not correspond to that of the column i. However, the number of rows of overlapping contiguous black pixels of columns (i-1) and i must be greater than the predetermined coverage threshold AZ to speak of a contiguous support structure.

More preferably, a given direct environment of the trailer coupling in the search of the transformed area can be disregarded. This is due to the fact that the structure of a bicycle carrier can be such that just right on the trailer hitch many braces are attached. To simplify the search for contiguous support structures, the direct environment of the trailer coupling can be disregarded when searching the transformed area. This means searching for coherent substructures left and right of the trailer hitch. If the number of contiguous sections on the left and right of the trailer coupling then exceeds a respectively predetermined value W1 or Wr in the binary image of the transformed area, then a carrier structure and thus a carrier device are found.

• – eine Einrichtung zur Transformation zumindest eines vorgegebenen Bereichs des Bildes um die Anhängerkupplung in ein binäres Bild, und
• – eine Einrichtung zum Durchsuchen des transformierten Bildbereichs nach mindestens einer zusammenhängenden Trägerstruktur.

The device according to the invention for carrying out the above-described method for detecting a mounted at the rear of a motor vehicle on a trailer hitch carrier device, with a backward camera, in particular a digital camera, for generating a rear image and an evaluation device for evaluating the recorded image comprises :

• Means for transforming at least a predetermined area of the image around the hitch into a binary image, and
• - Means for searching the transformed image area for at least one contiguous support structure.

The Anhängerrangierassistent invention of a motor vehicle for maneuvering a team consisting of a motor vehicle and a non-steerable trailer, has a device by means of which the occupancy of the trailer socket of the motor vehicle can be checked, and has a device described above for detecting a at the rear of a motor vehicle a trailer hitch mounted on carrier means. In this case, an activation of the Anhängerrangierfunktionen only take place when an assignment of the trailer socket has been detected and found that no support device is present at the rear of the vehicle.

A preferred embodiment of the invention will be explained below with reference to the drawings. It shows

1 the image taken by the rear camera of the vehicle of a bicycle carrier,

2 the area of 1 after the black and white transfer for further processing.

1 shows by way of example the image of a reversing camera, the one on a trailer hitch 1 of a motor vehicle mounted bicycle carrier 2 shows, which has its own lighting device including brake lights, as a loaded bike carrier 2 can cover the rear lights of the vehicle. Therefore, the bike carrier points 2 a corresponding cable with plug for power and signal supply, which is connected to the trailer socket of the vehicle. It is therefore obvious that the fact of Occupancy of the trailer socket is not sufficient for determining whether a bicycle carrier 2 or a trailer occupies the trailer socket. In particular, in a Anhängerrangierassistenten must be clear before activating the Rangierfunktionen whether a trailer is actually present and coupled to the vehicle.

The image of the bike carrier 2 When using standard video cameras, the reversing camera is usually a color image consisting of corresponding color channels, for example R, G and B. The color image can be converted into a gray scale image (not shown) in the usual way. However, it is also possible to use only one color channel or a combination of color channels for the evaluation. In the following, it is assumed that the color image of the rear camera has been converted into a gray value image, that is to say only the luminance component of the color image is considered in the further consideration.

As 1 shows, the image of the rear view camera is distorted. It is possible to equalize the image before further evaluation. However, since this does not lead to a gain in knowledge for the detection of a bicycle carrier, the further evaluation can be based on the distorted image, ie an equalization of the image is not necessary, which contributes to reduced computing capacities of the method.

Further shows 1 an area 3 around the trailer hitch 1 which is selected for further consideration. The size of the area depends on this 3 according to the type and shape of the bike carrier 2 , Furthermore, the area depends 3 also from the resolution and the field of view of the rear view camera.

The restriction of the investigation of the image of the reversing camera on the area 3 the trailer hitch 1 is due to the fact that the bodies of bicycle carriers 2 to hold the bikes can all be different. Thus, a carrier system may have straight carrier elements, while the carrier system of the example of 1 having curved support elements. Furthermore, the carrier may already be equipped with bicycles or not. Thus, for a consideration, that portion of the mounted bicycle carrier is 2 interesting, which experiences the least changes under changeable conditions. This is the area 3 around the trailer hitch 1 because the total load of the bike carrier 2 at this point, and all struts and rods ultimately converge at that point.

2 shows the area essential for the consideration 3 the trailer hitch 1 the picture of the 1 , which is already a black and white picture. The black and white image emerges from the 1 By subtracting the threshold from the derived gray value image such that all values above a predetermined threshold Sg are assigned the value White and the value Black is assigned to all values that are equal to or smaller than the threshold value Sg.

It is also possible, first, the area 3 of the 1 select only this area 3 undergo a gray value conversion, so that then only the range 3 after the gray value conversion into the black and white picture of the 2 is converted by means of a suitably chosen threshold value.

It serves the conversion of the area 3 in a black and white image the purpose, the evaluation "pixels could become a bike carrier 1 belong "to simplify. The transformation of the gray values into black and white values takes place as described above, namely that a predetermined, suitable gray value is defined as the threshold value Sg and gray values which are above or below the threshold are set to white or black.

2 therefore shows a coherent support structure 4 in black as well as some black pixels 5 on the in 1 shown background of the area 3 are attributed. This in 2 shown area 3 is now searched in the manner outlined below to around the trailer hitch 1 the coherent support structure 4 to determine.

The assumption of continuity is made, which means that a bicycle carrier has a continuous structure. By this is meant that the rods and connectors of the bicycle carrier are all directly connected to each other, which is true at least in the area of the trailer hitch, since the total load is taken up at this point and all struts and bars must lead to this point. This in turn means that in the black and white image, the black or white values must be continuous.

If, in the creation of the black-and-white image, pixels do not correspond to the assumption of continuity, they are filtered during the examination of the black-and-white image, which is to be illustrated by means of an example.

It is assumed that the background is structured such that isolated pixels are assigned to the black value, the rest of the background assumes white values. The structure of the bike carrier is shown with black pixels.

After the assumption of the continuity belong the isolated black pixels of the underground therefore not to the bicycle rack. Due to the assumed continuity, a certain number of black pixels must belong to the bicycle carrier. All black pixels that are scattered on the ground can therefore be set to the value of the pixel environment, which means white here. The same is true with white pixels on the bicycle rack. These are also interpreted as "noise" and adapted to the pixel environment.

Due to sun and shade or different types of bicycle carriers, the thickness of the bars may vary and appear thicker or thinner in the black and white image. Therefore, the number of contiguous pixels can also vary.

Now there is a column-line adjustment of the black and white image. Which order is used does not matter. In the following, first the columns and then the lines are compared, assuming that the bicycle carrier is represented by black pixels.

Each column consists of a certain number of black and white pixels. After the assumption made above, the black pixels belong to the bike carrier. In the event that consecutive black pixels are present in a column, this number must exceed a predetermined frequency threshold AS. If this condition is fulfilled, the row entries j of the examined column i are stored.

After the scan is done for all columns, the rows are examined.

Due to the assumption of continuity, the rows of black pixels from column i must be close to the rows of black pixels from column i-1 in order to belong to a bicycle carrier. Here, too, a match threshold AZ is defined which describes by how many lines the contiguous black pixels from column i must at least overlap with column i-1.

Thus, the rows of each column having pixel values of the bicycle rack are compared with each other. If these values are close together, these sections describe a bike carrier.

Since the structure of a bicycle carrier can be designed so that many braces are mounted directly on the trailer hitch, the pixels directly in the vicinity of the trailer hitch can be omitted in a further embodiment when viewed. This means that the number of sections left and right of the trailer hitch is considered. If the number of sections left and right of the trailer hitch exceeds a certain value Wl and Wr in the SW picture, then there is a bicycle carrier.

Using the example of the column line alignment, the algorithm is reproduced below. The alternative form of row-column alignment results in an identical way by interchanging columns and rows. Furthermore, the algorithm is analogously applicable to white pixels if the support structure is represented by white pixels.

Initially, the captured image of the rear camera is considered to decide whether to examine a single color component, multiple color components, or a combination of color components.

Thereafter, a conversion of the entire image into a gray-scale image and then the selection of the area to be examined in the vicinity of the trailer hitch can take place. It is also possible to first determine the area to be examined around the trailer coupling and then to perform a transfer of the image of the area to be examined in a gray scale image. Both approaches are identical in their result.

In the next step, a black-and-white image of the gray-scale image of the area to be examined is created with the aid of a threshold Sg, in which all gray values with a value smaller than the threshold Sg have the value black and all gray values with a value greater than or equal to the threshold value White is assigned. This assignment can also be reversed. It is only necessary to know if the suspected carrier structure has the value black or white. In other words, this step is used to separate the substrate and support structure.

The area to be examined, the trailer hitch 3 is after the black-and-white transformation in the form of a matrix with the columns i = 1 to n and the rows j = 1 to m before. It must now be investigated whether at least one coherent structure, ie a support structure, is present in the area to be investigated. For this purpose, the relationship or coincidence in the line-wise direction is examined first for column-like relationships and then starting in front of the column-related components.

To determine if in the area to be examined 3 In the next step, columns i = 1 to n are examined for whether the number of contiguous black pixels in the current column i is larger or equal to a predetermined frequency threshold AS. If so, these lines j _{i of} the column i are stored with black pixels.

Next, it must be examined whether the black contiguous pixel areas of the columns also coincide in the line-wise direction. For this purpose, the contiguous regions of adjacent columns are checked and the largest and the smallest common line j _i and j _{i-1 of} two adjacent columns i and i-1 are determined, which are referred to as g _{i_max} and g _{i_min} . The difference (g _{i_max} -g _{i_min} ) of the largest and the smallest common row is formed, and this difference is compared with a predetermined match _threshold AZ. If this difference is greater than or equal to the coincidence threshold value AZ, then this statement is stored in a list L with the column entry i-1.

If the difference of the continuous positive column entries in the list L is greater than a further predetermined threshold W, then a support structure is found, so that the statement can be made that a support device is mounted at the rear of the vehicle.

Furthermore, as already mentioned above, the direct environment 6 the trailer hitch omitted in the consideration. For this purpose, the columns in list L, which concern the direct environment of the trailer coupling, are not taken into account.

It is then further investigated whether the difference of the elements in the list L (i _{range hitch right} - i ₁ ) is greater than or equal to a left threshold Wl. It is further checked whether the difference of the elements in the list L (i _n - i _{area trailer hitch left} ) is greater than or equal to a right threshold Wl. If this is the case for both sides, ie with the logical AND connection, a bicycle carrier is mounted at the rear of the vehicle.

LIST OF REFERENCE NUMBERS

1

trailer hitch

2

bike rack

3

Area

4

support structure

5

sough

6

direct environment of the trailer hitch

7

left substructure

8th

right substructure

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102009038552 A1 [0008]
• DE 102008013457 A1 [0009]
• DE 102004057220 A1 [0010]

Claims (10)

Method for detecting a rear of a motor vehicle on a trailer hitch ( 1 ) mounted carrier device ( 2 ), wherein the motor vehicle is equipped with a rear-facing camera, characterized by the steps: - Creating an image of the rear surroundings of the motor vehicle, selecting an area ( 3 ) around the trailer hitch ( 1 ) in the image of the camera, transform at least the area ( 3 ) of the camera image into a binary image having pixels of a first binary value and pixels of a second binary value using a predetermined threshold value (Sg), searching the area transformed into a binary value ( 3 ) according to at least one contiguous support structure ( 4 ), and detection of a mounted carrier device when finding at least one coherent carrier structure ( 4 ) in the binarized range ( 3 ). A method according to claim 1, characterized in that in the case of the representation of the contiguous support structure ( 3 ) first through the pixels of the first binary value, the columns of the domain transformed into binary values ( 3 ) in terms of the column-related frequency of occurrence of pixels of the first binary value and then the lines of the transformed region ( 3 ) are searched for line matches of the column-wise contiguous occurrence of the pixels of the first binary value, or first the lines of the transformed area ( 3 ) are examined for the line-by-line frequency of occurrence of the pixels of the first binary value, and then the columns for column-like matches in the occurrence of the row-wise contiguous frequency of the pixels of the first binary value. A method according to claim 1, characterized in that in the case of the representation of the contiguous support structure ( 3 ) through pixels of the second binary value first the columns of the black-and-white transformed region ( 3 ) in terms of the column-related frequency of occurrence of the pixels of the second binary value and then the lines of the transformed region ( 3 ) are searched for line matches of the columnar occurrences of the pixels of the second binary value, or first the lines of the transformed area ( 3 ) are examined for the line-by-line frequency of occurrence of the pixels of the second binary value, and then the columns for column-like matches in the occurrence of the line-by-line related frequency of the pixels of the second binary value. Method according to one of claims 2 or 3, characterized in that the contiguous frequency of the occurrence of the pixels of the first or second binary value by a predetermined frequency threshold (AS) is set, so that a number of pixels of the first and second binary value in a Column or row is not considered contiguous if the count is less than the predetermined frequency threshold and considered contiguous if the frequency is equal to or greater than the given frequency threshold. A method according to claim 4, characterized in that the line-by-line coverage of the column-wise contiguous frequency or the columnar coverage of the row-wise contiguous frequency for adjacent rows or adjacent columns must be greater than or equal to a predetermined coverage threshold to form a contiguous support structure. Method according to one of the preceding claims, characterized in that the binary values are realized by the values black and white or zero and one. Method according to one of the preceding claims, characterized in that a predetermined environment ( 6 ) of the trailer hitch ( 1 ) when searching the transformed area ( 3 ) is not considered. Device for carrying out the method for detecting a rear of a motor vehicle on a trailer hitch ( 1 ) mounted carrier device ( 2 ) according to one of the preceding claims, with a rear-facing camera for generating a rear image and an evaluation device for evaluating the recorded image, characterized in that the device further comprises: means for transforming at least one predetermined area ( 3 ) of the image around the hitch in a binary image, and means for searching the transformed image area for contiguous support structures ( 4 ). Apparatus according to claim 8, characterized in that a digital camera is used. Anhängerrangierassistent a motor vehicle for maneuvering a team consisting of the motor vehicle and a non-steerable Trailer, wherein the Anhängerrangierassistent has a device by means of which the occupancy of the trailer socket of the motor vehicle can be checked, characterized in that the Anhängerrangierassistent has a device according to claim 8 and activation of the Anhängerrangierfunktionen can only take place when occupancy of the trailer socket is detected and no carrier device was recognized at the rear of the vehicle.

Images