DE102017100669A1 - Method for detecting an environmental region of a motor vehicle with adaptation of a region of interest as a function of a trailer, computing device, camera system and motor vehicle - Google Patents

Abstract

The invention relates to a method for detecting an environmental region (8) of a motor vehicle (1) in which at least one image (10) recorded by a camera (4) is received, wherein the at least one image (10) describes the environmental region (8) in which image (10) a region of interest (19) is determined and within the region of interest (19) objects are detected, wherein if it is recognized that a trailer (9) is connected to the motor vehicle (1), in the at least one image (10) is determined a trailer area (11), which describes the trailer (9), and the region of interest (19) on the basis of the trailer area (11) is determined.

Description

The present invention relates to a method for detecting an environmental region of a motor vehicle in which at least one image captured by a camera is received, wherein the at least one image describes the environmental region in which image a region of interest is determined and within the region of interest objects be recorded. In addition, the present invention relates to a computing device for a camera system. Moreover, the present invention relates to a camera system with such a computing device. Finally, the present invention relates to a motor vehicle with such a camera system.

The interest here is directed to methods with which a surrounding area of a motor vehicle can be detected by means of at least one camera. In this case, images of the surrounding area can be provided with the at least one camera. By means of corresponding object recognition algorithms or methods of machine vision, objects can then be detected in these images. It is also known from the prior art that in the images regions of interest are determined and they are used for further processing. Thus, for example, areas that are not of interest can not be considered. Depending on the application, these areas, which are not taken into account, for example, describe the motor vehicle, the sky or the like. Thus, memory requirements can be saved and computing power can be increased.

If a trailer is connected to the motor vehicle, the region of interest determined in the image may be blocked by the trailer. For example, the region of interest may include areas in the image that show components of the trailer, such as the drawbar or the hitch. This may result in affecting a machine vision method or an object recognition algorithm.

In the context of methods for detecting objects by means of ultrasonic sensors, it is known from the prior art that an ultrasonic sensor, which is assigned to a trailer device, is not operated in the transmission mode, but only in the receiving mode. Thus, for example, it can be prevented that a trailer hitch is erroneously recognized as an object. Such a method is for example in the WO 2015/193060 A1 described.

It is an object of the present invention to provide a solution as to how a surrounding area of a motor vehicle can be reliably detected by means of a camera even in the case of a trailer connected to the motor vehicle.

This object is achieved by a method by a computing device, by a camera system and by a motor vehicle with the features according to the respective independent claims. Advantageous developments of the present invention are the subject of the dependent claims.

According to one embodiment of a method for detecting an environmental region of a motor vehicle, at least one image captured by a camera is received. The picture especially describes the surrounding area. In particular, a region of interest is determined in the image. Furthermore, objects are preferably acquired in the region of interest. It is preferably provided that if it is detected that a trailer is connected to the motor vehicle, in which at least one image preferably a trailer area is determined, which describes in particular the trailer. Furthermore, the region of interest is preferably determined by the trailer area.

An inventive method is used to detect an environmental area of a motor vehicle. In this case, at least one image recorded by a camera is received, wherein the at least one image describes the surrounding area. Furthermore, a region of interest is determined in the image, and objects are detected within the region of interest. If it is detected that a trailer is connected to the motor vehicle, a trailer area which describes the trailer is determined in the at least one image. Furthermore, the region of interest is determined by the trailer area.

By means of the method, the surrounding area of the motor vehicle is to be detected. The method can be carried out with a computing device of a camera system. This computing device can be connected for data transmission with at least one camera of the camera system. Thus, the at least one image, which is taken with the camera and which describes the surrounding area, can be transmitted to the computing device. It is preferably provided that a plurality of temporally successive images are taken with the camera and transmitted to the computing device. The computing device can then be used to determine a region of interest (Rol region of interest) within the at least one image. This region of Interest can be determined so that the areas describing the lane include. In addition, the region of interest may be determined so as not to include those areas which describe the motor vehicle itself or the sky. In particular, the region of interest describes a section of the at least one image. In this region of interest then objects can be detected.

According to an essential aspect of the present invention, it is provided, if it is recognized that the trailer is connected to the motor vehicle, in which at least one image of the trailer area is determined. In this case, the image may originate in particular from a camera of the motor vehicle, which is arranged at a rear region of the motor vehicle or by means of which a surrounding area in the direction of travel is received behind the motor vehicle. The trailer area determined within the image shows at least part of the trailer or area blocked by the trailer. Once the trailer area has been defined within the image, it can be taken into account in determining the region of interest. Thus, for example, the trailer can be prevented from being taken into account by an object recognition algorithm or a method of machine vision. In addition, it can be prevented that the trailer or parts thereof are erroneously recognized as an object. This can reduce the false positive rate. Overall, it can thus be achieved that objects in the surrounding area are reliably detected even when the trailer is connected to the motor vehicle.

Preferably, the region of interest is determined such that the region of interest is different from the pendant region. In other words, the region of interest may exclude the trailer area. In this case, it may be provided, for example, that initially an original region of interest is determined, which is usually used when there is no trailer on the motor vehicle. Following this, the trailer area associated with the trailer can be determined. To determine the region of interest, then the parts of the original region of interest that are different from the tag region can be selected. It can also be provided that first the trailer area is determined and then based on the trailer area the region of interest is determined. Thus, as a whole, it can be reliably prevented that the trailer or parts thereof are taken into account in the detection of the surrounding area. In this case, it can be considered in particular that objects can not be located in the area occupied by the trailer.

In one embodiment, at least two temporally successive images are received, the at least two images are each assigned to a plurality of pixels, a motion vector is determined for each of the pixels, and the pixels are assigned to the trailer region depending on their motion vector. It is provided in particular that the temporally successive images are taken by the camera while the motor vehicle is being moved. The temporally successive images which are received are each divided into several picture elements. In this case, each of the picture elements comprises at least one picture element of the picture. In this case, the corresponding picture elements of the temporally successive pictures can be compared with one another. This makes it possible to determine a motion vector for each of the picture elements. In particular, the motion vector describes a movement of the object, which is imaged by the picture element, between the at least two temporally successive pictures. The motion vector may be determined, for example, by an optical flow method. When a motion vector has been determined for each of the picture elements, it may be decided whether or not the respective picture elements are assigned to the trailer area. This allows a reliable determination of the trailer area.

Furthermore, it is advantageous if those picture elements are assigned to the trailer area whose motion vectors describe a lower movement speed and / or a different direction of movement compared to the other motion vectors. In this case, the knowledge is taken into account that the motion vectors of those picture elements which describe the trailer describe no or very little movement. This is due to the fact that the trailer is moved together with the motor vehicle. Thus, the picture elements describing the tag can be easily distinguished from other picture elements describing, for example, objects in the surrounding area. Thus, for example, picture elements which describe the roadway can be distinguished from picture elements which describe the trailer. It can also be checked whether the motion vectors of the pixels have a different direction of movement compared to the other motion vectors. Even such outliers among the motion vectors can then be assigned to the trailer area.

In this case, it is provided in particular that continuous successive images are received and the trailer area is continuously updated on the basis of the images. As already explained, it is provided in particular that the images are taken during the movement of the motor vehicle. Based on the temporally successive images can then be determined for each of the pixels of the motion vector. This makes it possible to continuously update or refine the trailer area. With the trailer area in the pictures can be precisely determined. In this way it is also possible that the trailer area for different trailer or trailer types that can be attached to the motor vehicle can be determined.

Furthermore, it is advantageous if in the at least one image a location area is determined which describes the trailer with a predetermined probability, and the trailer area is determined on the basis of the location area. The image received by the computing device describes the surrounding area of the motor vehicle. The mounting position of the camera on the motor vehicle and the detection range of the camera are known. Thus it can be determined in which area of the image the trailer can be present. In this case, it can be considered, for example, that the trailer is connected to the trailer coupling, which is located centrally on the rear area of the motor vehicle. Thus, the location area in which the trailer is with a high probability can be taken into account in the determination of the trailer area. This allows a reliable determination of the trailer area.

In a further embodiment, a trailer angle between the motor vehicle and the trailer is continuously determined and the trailer area is determined in dependence on the trailer angle. In particular, the trailer angle describes an angle between the longitudinal axis of the motor vehicle and the longitudinal axis of the trailer. The trailer angle can be determined for example with a corresponding sensor, which is arranged in the trailer hitch of the motor vehicle or the drawbar of the trailer. Furthermore, the trailer can be determined with a corresponding environmental sensor of the motor vehicle. In this case, it may further be provided that the movement of the motor vehicle and in particular a steering angle or rate of rotation of the motor vehicle are detected continuously. In this way, the trailer angle can be estimated. If the trailer angle is known, it can be determined where the trailer is relative to the motor vehicle. This can also be used to determine where the image of the trailer is located within the image. This can be taken into account when determining the trailer area. This allows a precise determination of the trailer area.

In one embodiment, to detect that the trailer is connected to the motor vehicle, a connection signal is received. The computing device can receive this connection signal, for example via a data bus of the motor vehicle. This connection signal can be sent out if the trailer is electrically or mechanically connected to the motor vehicle. Thus, it can be recognized in a permissible manner that the trailer is coupled to the motor vehicle.

Furthermore, it is advantageous if the objects in the region of interest are detected by means of an object recognition algorithm. It may also be provided that a method of machine vision is used to detect the objects in the region of interest. Thus, for example, further road users in the vicinity of the motor vehicle can be reliably detected. In particular, it becomes possible to detect pedestrians in the surrounding area. In the same way static objects or obstacles can be detected. In the region of interest, the roadway or the road surface can be detected. This can be used to calibrate the settings of the camera. By determining the region of interest outside the trailer area, reliable detection of the objects can be enabled. In addition, the memory requirement can be reduced and the computational effort can be reduced.

A computer direction according to the invention for a camera system is designed for performing a method according to the invention and the advantageous embodiments thereof. The computing device can be, for example, a programmable computer, such as a digital signal processor, microcontroller or the like. Accordingly, a computer program may be provided which is stored, for example, on a storage medium, wherein the program is programmed to carry out the method described here when it is executed on the computing device.

An inventive camera system for a motor vehicle comprises a computing device according to the invention and at least one camera. With the camera images of the surrounding area can be provided and transmitted to the computing device. The camera system preferably comprises a plurality of cameras. These cameras can then be distributed to the motor vehicle.

A motor vehicle according to the invention comprises a camera system according to the invention. The motor vehicle can be designed as a passenger car or as a commercial vehicle.

The preferred embodiments presented with reference to the method according to the invention and their advantages apply correspondingly to the computing device according to the invention, 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. The features and feature combinations mentioned above in the description, as well as the features and combinations of features mentioned below in the description of the figures and / or shown alone in the figures, can be used not only in the respectively specified combination but also in other combinations or in isolation, without the frame to leave the invention. Thus, embodiments of the invention are to be regarded as encompassed and disclosed, which are not explicitly shown and explained in the figures, but which emerge and can be produced by separated combinations of features from the embodiments explained. Embodiments and combinations of features are also to be regarded as disclosed, which thus do not have all the features of an originally formulated independent claim. Moreover, embodiments and combinations of features, in particular by the embodiments set out above, are to be regarded as disclosed, which go beyond or deviate from the combinations of features set out in the back references of the claims.

The invention will now be described with reference to preferred embodiments and with reference to the accompanying drawings.

• 1 ein Kraftfahrzeug gemäß einer Ausführungsform der Erfindung, welches ein Kamerasystem aufweist, wobei das Kraftfahrzeug mit einem Anhänger verbunden ist;
• 2 ein Bild, welches von einer Kamera des Kamerasystems bereitgestellt wird, wobei das Bild in eine Mehrzahl von Bildelementen eingeteilt ist, wobei jedem der Bildelemente ein Bewegungsvektor zugeordnet ist;
• 3 ein weiteres Bild, welches von einer Kamera bereitgestellt wird, wobei in dem Bild eine Region von Interesse definiert ist; und
• 4 das Bild von 3, bei welchem eine Region von Interesse gemäß einer Ausführungsform der Erfindung definiert ist.

Showing:

• 1 a motor vehicle according to an embodiment of the invention, which has a camera system, wherein the motor vehicle is connected to a trailer;
• 2 an image provided by a camera of the camera system, the image being divided into a plurality of pixels, each of the pixels being associated with a motion vector;
• 3 another image provided by a camera, wherein in the image a region of interest is defined; and
• 4 the picture of 3 in which a region of interest is defined according to an embodiment of the invention.

In the figures, identical or functionally identical elements are provided with the same reference numerals.

1 shows a motor vehicle 1 according to an embodiment of the present invention in a plan view. The car 1 is designed in the present case as a passenger car. The car 1 is with a trailer 9 connected. The car 1 make up together with the trailer 9 a team.

The car 1 includes a camera system 2 , The camera system 2 in turn comprises a computing device 3 , for example, by an electronic control unit of the motor vehicle 1 can be formed. In addition, the camera system includes 2 at least one camera 4 , In the present embodiment, the camera system includes 2 four cameras 4 distributed to the motor vehicle 1 are arranged. Present is one of the cameras 4 in a rear area 5 arranged one of the cameras 4 is in a front area 7 of the motor vehicle 1 arranged and the remaining two cameras 4 are in a respective page area 6 , in particular in an area of the side mirrors, arranged. The number and arrangement of the cameras 4 of the camera system 2 is to be understood in the present case purely by way of example.

With the cameras 4 can be a surrounding area 8th of the motor vehicle 1 be recorded. The four cameras 4 are preferably of identical design. In particular, with the cameras 4 images 10 or a picture sequence are provided which the environment area 8th describe. These pictures 10 can from the cameras 4 to the computing device 3 be transmitted. By means of the computing device 3 can display device of the motor vehicle, not shown here 1 be controlled, giving the driver the pictures 10 the cameras 4 can be displayed. Based on the pictures 10 can with the computing device 3 Objects are detected. The camera system 2 thus serves to support the driver of the motor vehicle 1 while driving the motor vehicle 1 , The camera system 2 For example, it may be a so-called electronic rearview mirror or a parking assist system or other system.

2 shows a picture 10 , which with one of the cameras 4 of the camera system 2 provided. In particular, the picture becomes 10 with the camera 4 provided, which is located at the rear area 5 of the motor vehicle 1 located. This shows the picture 10 at least part of the trailer 9 , Within the picture 10 should now be a trailer area 11 be recognized, which the trailer 9 assigned. For this purpose, in the picture 10 a plurality of picture elements 12 Are defined. The picture elements 12 are in the present case rectangular or square executed. For each of the picture elements 12 becomes a motion vector 13 certainly. This is done with the computing device 3 temporally consecutive pictures 10 from the camera 4 receive. At this time, the picture elements 12 of the temporally successive pictures become 10 compared to each other. From this can then the respective motion vector 13 be determined.

Following this, the respective motion vectors 13 the picture elements 12 compared with each other. Present are picture elements 12 to recognize which of a first group 14 can be assigned. These picture elements 12 the first group 14 describe here a roadway 18 on which the motor vehicle 1 and the trailer 9 move. The motion vectors 13 of these picture elements 12 describe essentially the same speed of movement and the same direction of movement. Furthermore, there are picture elements 12 present, which is a second group 15 can be assigned. The motion vectors 13 These picture elements 12 differ in particular on the basis of their direction of movement of the motion vectors 13 the picture elements 12 the first group 14 , In addition, picture elements 12 a third group 16 be assigned. The motion vectors 13 of these picture elements 12 are characterized by the fact that these compared to the picture elements 12 the first group 14 and the second group 15 describe a lower speed of movement.

Due to the assignment of the picture elements 12 to the groups 14 . 15 . 16 can now the trailer area 11 be determined, which the trailer 9 describes. It may initially be provided that the picture elements 12 the first group 14 the roadway 18 be assigned. The picture elements 12 the second group 15 and the third group 16 can first contact the trailer 9 be assigned. It is provided in particular that continuously successive images 10 be received and thus the trailer area 11 is updated continuously.

In addition, it is preferably provided that a residence area 17 it is determined which the trailer 9 in the picture 10 shows with a high probability. In other words, this residence area describes 17 the area in which the trailer 9 with a high probability in the picture 10 is shown. This can be taken into account, for example, that the trailer 9 connected to a trailer hitch, which is centered on the rear area 5 of the motor vehicle 1 located. In addition, it is preferably provided that a follower angle α is determined continuously. The trailer angle α describes the angle between a longitudinal axis of the motor vehicle 1 and a longitudinal axis of the trailer 9. For this purpose, a corresponding sensor can be used. In addition, also the steering angle of the motor vehicle 1 be determined continuously and from this the trailer angle α can be determined. Based on the trailer angle α can be determined where the trailer 9 relative to the motor vehicle 1 located. The information about the area of residence 17 and / or the trailer angle α can be used to assign the picture elements 12 to the trailer area 11 to correct or refine. For example, it can thus be determined that only the picture elements 12 the third group 16 the trailer area 11 are assigned.

3 shows a picture 10 which with the camera 4 is provided according to another embodiment. Here is inside the picture 10 a region of interest 19 certainly. Within this region of interest 19 For example, objects can then be detected and / or methods of machine vision applied.

The region is of interest here 19 determined so that this is the image of the road 18 includes. In addition, the region closes of interest 19 Regions of the image 10, which the motor vehicle 1 describe. In the present case, the region is of interest 18 but also the trailer area 11 with a.

In comparison shows 4 the picture 10 according to 3 in a further embodiment. This was the trailer area 11 in determining the region of interest 19 considered. The region of interest 19 This is determined so that this from the trailer area 11 is different or the trailer area 11 excludes. On the basis of this region of interest 19 then object recognition algorithms or machine vision techniques can be applied. Thus, for example, can be prevented that the trailer 9 erroneously recognized as an object. The information concerning the particular region of interest 19 It is also possible to transmit other algorithms or functional devices containing the images 10 the camera 4 use.

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

• WO 2015/193060 A1 [0004]

Claims (12)

Method for detecting an environmental region (8) of a motor vehicle (1) in which at least one image (10) recorded by a camera (4) is received, wherein the at least one image (10) describes the surrounding region (8) in the image (10) a region of interest (19) is determined and within the region of interest (19) objects are detected, characterized in that if it is recognized that a trailer (9) is connected to the motor vehicle (1) in which at least one image (10) is determined a trailer area (11), which describes the trailer (9), and the region of interest (19) on the basis of the trailer area (11) is determined. Method according to Claim 1 characterized in that the region of interest (19) is determined such that the region of interest (19) is different from the trailer region (11). Method according to Claim 1 or 2 , characterized in that at least two temporally successive images (10) are received, to which at least two images (10) each a plurality of pixels (12) is assigned, for each of the pixels (12) a motion vector (13) is determined and the Picture elements (12) are assigned to the trailer area (11) as a function of their motion vector (13). Method according to Claim 3 , characterized in that those picture elements (12) are assigned to the trailer area (11) whose motion vectors (13) describe a lower speed of movement and / or a different direction of movement compared to the other motion vectors (13). Method according to one of the preceding claims, characterized in that continuous successive images (10) are received and the trailer area (11) is continuously updated on the basis of the images (10). Method according to one of the preceding claims, characterized in that in the at least one image (10) a residence area (17) is determined which describes the trailer (9) with a predetermined probability, and the trailer area (11) is determined on the basis of the location area (17 ) is determined. Method according to one of the preceding claims, characterized in that continuously a trailer angle (a) between the motor vehicle (1) and the trailer (9) is determined and the trailer area (11) in dependence on the trailer angle (a) is determined. Method according to one of the preceding claims, characterized in that for detecting that the trailer (9) is connected to the motor vehicle (1), a connection signal is received. Method according to one of the preceding claims, characterized in that within the region of interest (19) objects are detected by means of an object recognition algorithm and / or a method of machine vision. Computing device (3) for a camera system (2), which is designed to carry out a method according to one of the preceding claims. Camera system (2) for a motor vehicle (1) with a computing device (3) according to Claim 10 and with at least one camera (4). Motor vehicle (1) with a camera system (2) according to Claim 11 ,

Images