Method and device for camera-based determination of the distance of a moving object in the vicinity of a motor vehicle

DE102019210300B4Active Publication Date: 2024-11-28VOLKSWAGEN AG
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
DE102019210300
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Filing Date
2019-07-11
Publication Date
2024-11-28
Estimated Expiration
2039-07-11

Smart Images

  • Figure 00000000_0001_ABST
    Figure 00000000_0001_ABST
  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Method for camera-based determination of a distance (b_1,b_2) of a moving object (60) in the environment (30) of a motor vehicle (50), wherein a first environment image (10) of the environment (30) at a first time (t1) and a second environment image (11) of the environment (30) at a subsequent second time (t2) are captured by means of a camera (2), wherein the moving object (60) is detected in the captured environment images (10,11), and wherein at least one stationary object (61) is detected in the captured environment images (10,11) and selected in the first environment image (10) depending on a distinguished geometric relationship relative to the moving object (60), wherein an image size ratio (k) between an image size (x'_1) of the moving object (60) and an image size (y'_1) of the stationary object (61) in the first environment image (10) is determined, where an image size (y'_2) of the stationary object (61) is determined in the second environment image (11), where distances (a_1,a_2) of the stationary object (61) at the first time (t1) and at the second time (t2) and an object size (y) of the stationary object (61) are determined based on the determined image sizes (y'_1,y'_2) of the stationary object (61) and a vehicle speed (v_Fzg), and where distances (b_1,b_2) of the moving object (60) at the first time (t1) and at the second time (t2) are determined and provided based on the determined image size ratio (k), the determined distances (a_1,a_2) and the determined object size (y) of the stationary object (61).
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a method and a device for determining the distance of a moving object in the vicinity of a motor vehicle using a camera. The invention further relates to a motor vehicle.

[0002] Modern vehicles use cameras to capture their surroundings. For spatial capture, multiple cameras are needed to create a stereoscopic image of the environment.

[0003] From CN 106 197 382 B, a method for dynamically determining the distance to a target using a single vehicle camera is known. Starting with several individual images captured by the vehicle camera, the image size of an object in each image is determined. Using these determined image sizes and a distance calculated from the vehicle speed, a distance to the object is calculated. However, the method can only be applied to stationary objects.

[0004] From EP 2 202 601 A2, a reference navigation method is known that is a coordinate-less method. For navigation, a 3D scanner is integrated into a first vehicle and a second vehicle. The scanner of the first vehicle records the generated images of the route. This route and image data is stored and transmitted to the second vehicle(s) immediately or at a later time. The transmission can be direct, for example by radio, or indirect by transferring an information carrier, such as a CD. The second vehicle, in turn, captures image data of its surroundings via its scanner. This data is compared and correlated with the image data of the first vehicle.

[0005] From EP 3 451 021 A1, a measuring device with scan functionality for optical measurement of an environment is known, wherein the measuring device has a sensor with an arrangement of microcells as a receiving surface and direction-dependent active sub-areas of the receiver are defined depending on the transmission direction of the transmission radiation in order to adjust the active receiver surface to a varying imaging position of the received radiation.

[0006] From DE 10 2006 057 552 A1, a system for determining the distance and lateral position of a vehicle is known. The system comprises a camera and a processor. The camera is configured to view an area of ​​interest in which a vehicle is located and generates an electronic image of the area. The processor communicates electrically with the camera to receive the electronic image and analyzes the image by executing a pattern-matching algorithm based on a sample image. A value based on the algorithm is determined, indicating the probability that certain features of the electronic image actually correspond to the vehicle.

[0007] The invention is based on the objective of creating a method and a device for camera-based determination of the distance of a moving object in the vicinity of a motor vehicle.

[0008] The problem is solved according to the invention by a method with the features of claim 1 and a device with the features of claim 7. Advantageous embodiments of the invention are set forth in the dependent claims.

[0009] In particular, a method for camera-based determination of the distance of a moving object in the vicinity of a motor vehicle is provided, wherein a first image of the environment is captured at a first time and a second image of the environment is captured at a subsequent second time using a camera, wherein the moving object is detected in the captured images of the environment, and wherein at least one stationary object is detected in the captured images of the environment and selected in the first image of the environment depending on a distinguished geometric relationship relative to the moving object, wherein an image size ratio between an image size of the moving object and an image size of the stationary object in the first image of the environment is determined, and wherein an image size of the stationary object in the second image of the environment is determined.wherein distances of the stationary object at the first and second time points and an object size of the stationary object are determined based on the specified image sizes of the stationary object and a vehicle speed, and wherein distances of the moving object at the first and second time points are determined and provided based on the specified image size ratio, the specified distances and the specified object size of the stationary object.

[0010] Furthermore, a device for camera-based determination of the distance of a moving object in the vicinity of a motor vehicle is provided, comprising a camera for capturing the environment and an evaluation unit, wherein the camera is configured to capture a first image of the environment at a first time and a second image of the environment at a subsequent second time, and wherein the evaluation unit is configured to recognize the moving object in the captured images of the environment and furthermore to recognize at least one stationary object in the captured images of the environment and, depending on a distinguished geometric relationship relative to the moving object in the first image of the environment, to select an image size ratio between an image size of the moving object and an image size of the stationary object in the first image of the environment.and furthermore, to determine an image size of the stationary object in the second environment image, to determine distances of the stationary object at the first and second time points and an object size of the stationary object based on the determined image sizes of the stationary object and a vehicle speed, and to determine and provide distances of the moving object at the first and second time points based on the determined image size ratio, the determined distances and the determined object size of the stationary object.

[0011] The method and the device enable the determination of the distance to a moving object solely based on two images of the surrounding environment captured by a single camera at successive times. It is assumed that the camera, particularly in conjunction with a motor vehicle, is moved relative to its surroundings between the first and second time points. The moving object, for example, a vehicle traveling ahead, is detected in the captured images. Furthermore, at least one stationary object is detected in the captured images and selected based on a specific geometric relationship relative to the moving object in the first image. This detection and selection process includes, in particular, the identification of multiple stationary objects in the vicinity of the vehicle.After several stationary objects have been detected in the environment, a suitable stationary object is selected depending on a distinguished geometric relationship relative to the moving object in the first environment image.

[0012] The following process steps are performed with the selected stationary object. An image size ratio is determined between an image size, that is, a number of image elements, for example, in the horizontal or vertical direction (i.e., with respect to a width or height) of the moving object, and an image size of the stationary object in the first environment image. In particular, one image dimension (i.e., a width measured in image elements or a height measured in image elements) is considered separately. However, it is also possible to consider both image dimensions and, for example, calculate an average of the image size ratios determined for each width and height. If the size of an image element is known, the number of image elements can also be converted into a corresponding physical quantity (size of the image on the camera sensor).Subsequently, an image size of the stationary object in the second environment image is determined in the same way.

[0013] Based on the specified image sizes of the stationary object and a vehicle speed, which is determined, for example, by the vehicle's odometry system via the evaluation unit, the distances of the stationary object to the vehicle or camera at the first and second time points, as well as the size of the stationary object, are determined. In the final step, the distances of the moving object at the first and second time points are determined and provided based on the specified image size ratio, the specified distances of the stationary object, and the specified size of the stationary object.

[0014] In other words, the selected stationary object is used as a reference object for which distances to the camera or the vehicle at two points in time are determined using known imaging conditions of the camera and a known distance traveled. Since the stationary object was selected based on a specific geometric relationship relative to the moving object in the first environment image, the corresponding distances of the moving object at the first and second points in time can be determined from the determined image size ratio of the moving and stationary objects in the first environment image and the determined distances of the stationary object, taking the specific geometric relationship into account.

[0015] The advantage of the invention is that the distance to a moving object can be determined using a single camera. Since only one camera is required, costs and installation space can be saved.

[0016] The camera is specifically a monocular or single camera, meaning a camera that captures the surroundings from only one direction or with a single lens. The camera is typically mounted on a motor vehicle and captures an area in front of or behind the vehicle. During the capture of the first and second images of the surroundings, the camera moves along with the vehicle, so that at least the distance between the vehicle / camera and the stationary object changes, and at least the image size of the stationary object changes. If the camera captures an area behind the vehicle, this can, for example, be used to monitor reversing or to detect another vehicle approaching from behind before an overtaking maneuver.

[0017] The detection of moving and stationary objects in the captured environment images is performed using well-known pattern recognition methods. Machine learning techniques, for example, can be employed. It may be possible to specify which types of objects should be detected or searched for in the captured environment images. For example, the specification could be that the moving object is a vehicle traveling ahead.

[0018] Selecting the stationary object based on a specific geometric relationship relative to the moving object in the first environment mapping is intended to include, in particular, selecting a stationary object that is favorably positioned relative to the moving object. Specifically, the specific geometric relationship here refers to the condition that the stationary object is located in the immediate vicinity of the moving object due to its proximity, or that it at least partially touches the moving object. The underlying rationale is that, in the case of immediate proximity or contact, the size of the moving object and the stationary object are comparable, at least in this immediate vicinity or at points of contact, so that the dimensions of the moving object can be inferred from the dimensions of the stationary object.In the case of contact or close proximity, the distance, i.e., the distance to the camera, of the moving and stationary objects is the same. A point of contact between the vehicle and the road surface, which is always present, is particularly suitable for this purpose. Due to the excellent geometric relationship, the determined image sizes, and thus also the actual sizes of the objects, are therefore at least partially directly comparable.

[0019] The process steps for determining the distances of the moving object are carried out primarily by means of the evaluation unit. Parts of the evaluation unit can be implemented individually or collectively as a combination of hardware and software, for example, as program code executed on a microcontroller or microprocessor. However, it is also possible for parts to be implemented individually or collectively as an application-specific integrated circuit (ASIC).

[0020] The procedure can be repeated cyclically. In particular, the procedure can be repeated for pairs of environment images captured at different, especially adjacent, times. In this way, a distance to the moving object can be determined repeatedly. With repeated execution, a change in relative speed can be detected via a change in distance (and a relative speed determinable from it). This allows conclusions to be drawn about braking or acceleration of the moving object (e.g., a vehicle ahead), depending on the speed or change in speed of the vehicle, to which a driver assistance system can react accordingly.

[0021] It may be possible to execute the procedure for multiple moving objects based on the first and second captured environment images. In this case, the procedure steps are performed separately for each moving object. However, it is also possible that the same selected stationary object is used for multiple moving objects.

[0022] In one embodiment, the size of the moving object is determined based on the specified image size ratio and the size of the stationary object. A subsequent distance to the moving object is then determined solely based on this initial size and the size of the moving object as captured in a further image of the surrounding environment. This allows for the determination of the distance to the moving object in subsequent images after the initial size has been determined. This approach saves resources, as the remaining process steps are no longer necessary.

[0023] In one embodiment, the distances and / or the size of the stationary object are determined based on known optical properties of the camera, assuming a fixed-focus lens with a known image distance. This allows for a particularly simple and computationally efficient determination of the distances and size of the stationary object.

[0024] The same principle can also be applied to cameras with adjustable focus, whereby the image distances (sometimes also referred to as image widths) are determined or calculated from captured camera data.

[0025] In one embodiment, the desired geometric relationship is defined by at least one point of contact between image elements of the moving object and image elements of the stationary object in the first environment image. If the moving object is, for example, a vehicle traveling ahead, points of contact between the vehicle and a road surface, such as contact lines of the rear wheels with the road surface, can be identified in the first environment image and used to define the geometric relationship. In other words, it can be assumed that the actual dimensions at the points of contact between the vehicle and the road are the same, since they are equidistant from the camera. The stationary object could, for example, be a road marking derived from these points of contact.A specific or estimated lane width of a roadway is used.

[0026] In one embodiment, the relative speed of the moving object to the motor vehicle is determined and provided based on the specified distances of the moving object and a time difference between the first and second points in time. This also allows for the provision of status information regarding the dynamic state of the moving object.

[0027] In a further training course, it is planned that the absolute velocity of the moving object will be determined and provided. This is achieved by adding the determined relative velocity and the vehicle speed of the motor vehicle. The vehicle speed can be queried, for example, using the evaluation unit of a vehicle odometry system. This also provides an absolute value for the dynamic state of the moving object. Furthermore, the acceleration or deceleration of the moving object can be calculated from several absolute velocities determined at successive times.

[0028] Features for the design of the device result from the description of embodiments of the method. The advantages of the device are the same in each case as in the embodiments of the method.

[0029] Furthermore, a motor vehicle is created, comprising at least one device according to any of the described embodiments.

[0030] The invention is explained in more detail below with reference to preferred embodiments and the figures. These show: Fig. 1 a schematic representation of an embodiment of the device for camera-based determination of the distance of a moving object in the vicinity of a motor vehicle; Fig. 2 a schematic representation of an exemplary environment at the first time and at the second time to illustrate the invention.

[0031] In Fig. Figure 1 shows a schematic representation of an embodiment of the device 1 for camera-based determination of the distance of a moving object in the vicinity of a motor vehicle 50. The device 1 is arranged in the motor vehicle 50. The device 1 comprises a single camera 2 and an evaluation unit 3.

[0032] Parts of the evaluation unit 3 can be configured individually or collectively as a combination of hardware and software, for example, as program code executed on a microcontroller or microprocessor. However, it is also possible for parts to be configured individually or collectively as an application-specific integrated circuit (ASIC). The evaluation unit 3 comprises, in particular, a computing unit 4, e.g., in the form of a microprocessor, and a memory 5 in which the computing unit 4 can perform arithmetic operations.

[0033] Camera 2 captures a first image 10 of the surroundings at a first time point and a second image 11 of the surroundings at a subsequent second time point, and transmits the captured images 10 and 11 to the evaluation unit 3. Between the two time points, the vehicle 50, and consequently camera 2, is moved. It can be specifically provided that camera 2 captures a stream of images of the surroundings in the form of a video, and that the first image 10 and the second image 11 are extracted as individual frames from the video stream.

[0034] The evaluation unit 3 detects the moving object, from which the distance is to be determined, in the captured environment images 10, 11. The moving object could, for example, be a vehicle traveling ahead. Furthermore, the evaluation unit 3 detects stationary objects in the captured environment images 10, 11 and selects one of these stationary objects based on a specific geometric relationship relative to the moving object in the first environment image 10. In particular, the specific geometric relationship could be the same image plane in the first environment image 11, that is, the same distance to camera 2 for at least parts of the moving object and the stationary object. For example, image elements of a vehicle traveling ahead, such as rear tires, could be adjacent to image elements of the stationary object, such as a roadway or lane markings.From the context (ahead vehicle + road + rear tire) it can then be concluded that the moving object and the stationary object in this image plane, i.e. at the location of the adjacent image elements, have the same distance to the camera or to the vehicle 50.

[0035] The evaluation unit 3 then determines an image size ratio between the image size of the moving object and the image size of the stationary object in the first environment image 10. For this purpose, a number of image elements (pixels) in the environment image 10 is determined for a height and / or a width of the objects. From the two values ​​for the number of image elements, the image size ratio of the two objects to each other in the first environment image 10 can be determined.

[0036] In the second environment image 11, the evaluation unit 3 also determines an image size of the stationary object.

[0037] The evaluation unit 3 determines the distances of the stationary object at the first and second time points and the size of the stationary object based on the image sizes of the stationary object determined in the environment images 10, 11 and a vehicle speed v_Fzg. The vehicle speed v_Fzg is supplied to the evaluation unit 3, for example, by a vehicle odometry system 51 of the motor vehicle 50 and / or queried from it.

[0038] In the final step, the evaluation unit determines three distances b_1 and b_2 of the moving object at the first and second time points, based on the specified image aspect ratio, the specified distances, and the specified size of the stationary object. These determined distances b_1 and b_2 are then made available for further processing and output, for example, in the form of a digital data packet.

[0039] It may be provided that the evaluation unit 3 determines an object size of the moving object based on the determined image size ratio and the determined object size of the stationary object, whereby a further distance b_3 of the moving object is determined at a later time exclusively on the basis of the determined object size and an image size of the moving object determined in a captured further environment image.

[0040] Fig. Figure 2 shows a schematic representation of an exemplary environment 30 at the first time point t1 and at the subsequent second time point t2 to illustrate the invention. On the left side of the Fig. Figure 2 shows a top view of the surroundings 30, on the right side of the Fig. Figure 2 shows a schematic representation of the captured environment images 10, 11.

[0041] The environment 30 of the motor vehicle 50 comprises a moving object 60 and a stationary object 61. The moving object 60 has a size x and the stationary object 61 a size y, where the sizes x and y, in the example shown, refer to the width of objects 60 and 61 as viewed from the motor vehicle 50. For clarity, only half the width is shown in the top view. Additionally, a tree 62 is shown as a stationary object; however, this serves only to illustrate the change in image sizes. In principle, the tree 62 could also be selected as a stationary object.

[0042] The following section explains the individual steps of the procedure in more detail. It is assumed that camera 2 is positioned in the center of the vehicle 50 and that camera 2 captures the surroundings 30 in the direction of travel 52.

[0043] The steps described below are performed in evaluation unit 3 ( Fig. 1) carried out.

[0044] In one step, the moving object 60, in this case a vehicle driving ahead, is detected in the first captured environment image 10. An image size x'_1 of the moving object 60 is determined, whereby a number of image elements that the moving object 60 occupies in the first environment image 10 in a width (alternatively in a height) is determined.

[0045] In a subsequent step, stationary objects are detected in the first environment image 10, and a stationary object 61 is selected in the first environment image 10 based on a suitable geometric relationship to the moving object 60. In the example shown, the geometric relationship is, for instance, one or more points of contact of the rear wheels 63 of the preceding vehicle on the lane 31. In the example shown, a portion of the lane 31, or its width at this position, is selected as the stationary object 61, which is defined by the distance between the corresponding lane markings 32. At the points of contact, the dimensions of the moving object 60 and the stationary object 61 in the first environment image 10 can be directly compared, since the distance a_1, b_1 to the camera 2 is the same. In this step, an image size y'_1 of the stationary object 61 is also determined.The image size is determined by the number of image elements (pixels) that the stationary object 61 occupies in the first environment image 10 with respect to a width (alternatively height). If the size of an image element is known, the number of image elements can then be converted into a corresponding physical quantity (size of the image on the camera sensor).

[0046] Subsequently, an image size ratio k between the moving object 60 and the stationary object 61 in the first environment image 10 is determined. This is done by comparing the respective number of image elements: k = x'_1 / y'_1

[0047] After a known time period Δt (difference between the first time and the second time) and a known distance traveled Δa = v_Fzg * Δt in direction 52, the following steps are carried out using the evaluation unit 3 ( Fig. 1) carried out.

[0048] In the second captured environment image 11, an image size y'_2 of the stationary object 61, i.e. the lane or its width at the corresponding position, is determined.

[0049] Subsequently, the distances a_1, a_2 and an object size y of the stationary object 61 are determined. This assumes a fixed-focus optic for camera 2 with known properties.

[0050] According to geometric optics, the following applies: y'_1 / y = a'_1 / a_1 and y'_2 / y = a'_2 / a_2

[0051] Using the vehicle speed v_Fzg of the motor vehicle 50 (or the camera 2), the distance traveled can be determined with a known Δt: Δa = v_Fzg * Δt = a_1 - a_2

[0052] From this, and via an approximation of a fixed-focus optic with a known image distance: a' = a'_1 = b'_1≈a'_2≈b'_2 follows: y'_1*a_1 = y'_2(a_1⋅Δa)

[0053] After rearranging, this results in: a_1 = y'_2 * Δa / (y'_2 - y'_2) as well as a_2 = a_1 - Δa y = (a_1 / a')*y'_1 = (a_2 / a')*y'_2

[0054] In a subsequent step, the evaluation unit 3 calculates ( Fig. 1) the size x of the moving object 60 via the known image size ratio k: x = k * y

[0055] The distances b_1 and b_2 can then be determined from this: b_1 = a_1 b_2 = (x / x'_2)*a'

[0056] It may be provided that a relative velocity v_Obj_rel of the moving object 60 to the motor vehicle 50 is determined and provided on the basis of the determined distances b_1, b_2 of the moving object 60 and the time difference Δt between the first time t1 and the second time t2: v_Obj_rel = (b_2 - b_1) / Δt

[0057] Furthermore, it may be provided that an absolute speed v_Obj_abs of the moving object 60 is determined and provided, whereby the determined relative speed v_Obj_rel and a vehicle speed v_Fzg of the motor vehicle 50 are added together: v_Obj_abs = v_Obj_rel + v_Fzg

[0058] The method and the device make it possible to easily determine the distance to a moving object using a single camera. It can be provided that the method is repeated cyclically for further time points. Furthermore, it can be provided that the method is carried out for other moving objects in the vicinity of the motor vehicle 50. Reference symbol list 1 Device 2 cameras 3 Evaluation unit 4 Computer equipment 5 storage 10 First environment map 11 second environment image 30 environment 31 lanes 32 Lane markings 50 motor vehicles 51 Vehicle odometry 52 Direction of travel 60 moving object 61 stationary object 62 Tree 63 rear wheel t1 first time point t2 second time point x size (moving object) y size (stationary object) x'_1 Image size (moving object, first time point) x'_2 Image size (moving object, second point in time) y'_1 Image size (stationary object, first time point) y'_2 Image size (stationary object, second time point) k Image size ratio a_1 distance (stationary object, first time point) a_2 distance (stationary object, second time point) b_1 distance (moving object, first time point) b_2 distance (moving object, second time point) b_3 further distance (moving object, further time) v_Fzg Vehicle speed v_Obj_rel Relative velocity (moving object) v_Obj_abs Absolute velocity (moving object) Δt time difference Δa distance traveled

Claims

[1] Method for camera-based determination of a distance (b_1, b_2) of a moving object (60) in the environment (30) of a motor vehicle (50), wherein a first environment image (10) of the environment (30) at a first time (t1) and a second environment image (11) of the environment (30) at a subsequent second time (t2) are captured by means of a camera (2), wherein the moving object (60) is detected in the captured environmental images (10, 11), and wherein at least one stationary object (61) is detected in the captured environmental images (10, 11) and selected depending on a distinct geometric relationship relative to the moving object (60) in the first environmental image (10), wherein an image size ratio (k) between an image size (x'_1) of the moving object (60) and an image size (y'_1) of the stationary object (61) in the first environment image (10) is determined, wherein an image size (y'_2) of the stationary object (61) is determined in the second environment image (11), wherein distances (a_1,a_2) of the stationary object (61) at the first time (t1) and at the second time (t2) and an object size (y) of the stationary object (61) are determined on the basis of the determined image sizes (y'_1,y'_2) of the stationary object (61) and a vehicle speed (v_Vzg), and wherein distances (b_1,b_2) of the moving object (60) at the first time (t1) and at the second time (t2) are determined and provided on the basis of the determined image size ratio (k), the determined distances (a_1,a_2) and the determined object size (y) of the stationary object (61). [2] Method according to claim 1, characterized bythat an object size (x) of the moving object (60) is determined on the basis of the determined image size ratio (k) and the determined object size (y) of the stationary object (61), wherein a further distance (b_3) of the moving object (60) is determined at a later further point in time exclusively on the basis of the determined object size (x) and an image size of the moving object (60) determined in a captured further environmental image. [3] Method according to claim 1 or 2, characterized by that the distances (b_1,b_2) and / or the object size (x) of the stationary object (61) are determined on the basis of known optical properties of the camera (2) assuming a fixed focus optics with a known image distance. [4] Method according to one of the preceding claims, characterized bythat the distinguished geometric relationship is or will be defined by at least one point of contact between image elements of the moving object (60) and image elements of the stationary object in the first environment image (10). [5] Method according to one of the preceding claims, characterized by that a relative speed (v_Obj_rel) of the moving object (60) to the motor vehicle (50) is determined and provided on the basis of the determined distances (b_1,b_2) of the moving object (60) and a time difference (Δt) between the first time (t1) and the second time (t2). [6] Method according to claim 5, characterized by that an absolute speed (v_Obj_abs) of the moving object (60) is determined and provided, wherein for this purpose the determined relative speed (v_Obj_rel) and a vehicle speed (v_Fzg) of the motor vehicle (50) are added together. [7] Device (1) for camera-based determination of a distance (b_1,b_2) of a moving object (60) in the environment (30) of a motor vehicle (50), comprising: a camera (2) for detecting the surroundings (30), and an evaluation device (3), wherein the camera (2) is designed to capture a first environment image (10) of the environment (30) at a first time (t1) and a second environment image (11) of the environment (30) at a subsequent second time (t2), and wherein the evaluation device (3) is designed to detect the moving object (60) in the captured environmental images (10, 11) and further to detect at least one stationary object (61) in the captured environmental images (10, 11) and to select it as a function of a distinguished geometric relationship relative to the moving object (60) in the first environmental image (10), to determine an image size ratio (k) between an image size (x'_1) of the moving object (60) and an image size (y'_1) of the stationary object (61) in the first environmental image (10), and further to determine an image size (y'_2) of the stationary object (61) in the second surroundings image (11), to determine distances (a_1, a_2) of the stationary object (61) at the first time (t1) and at the second time (t2) and an object size (y) of the stationary object (61) on the basis of the determined image sizes (y'_1, y'_2) of the stationary object (61) and a vehicle speed (v_Vzg), and To determine and provide distances (b_1,b_2) of the moving object (60) at the first time (t1) and at the second time (t2) on the basis of the determined image size ratio (k), the determined distances (a_1,a_2) and the determined object size (y) of the stationary object (61). [8] Device (1) according to claim 7, characterized byin that the evaluation device (3) is further designed to determine an object size (x) of the moving object (60) on the basis of the determined image size ratio (k) and the determined object size (y) of the stationary object (60), and to determine a further distance (b_3) of the moving object (60) at a later further point in time exclusively on the basis of the determined object size (x) and an image size of the moving object (60) determined in a captured further environmental image. [9] Device (1) according to claim 7 or 8, characterized by that the evaluation device (3) is further designed to determine a relative speed (v_Obj_rel) of the moving object (60) to the motor vehicle (50) on the basis of the determined distances (b_1,b_2) of the moving object (60) and a time difference (Δt) between the first time (t1) and the second time (t2). [10] Motor vehicle (50) comprising at least one device (1) according to one of claims 7 to 9.

Citation Information

Patent Citations

  • A Method for Dynamically Measuring the Distance of On-vehicle Single Camera Targets

    CN106197382B

  • system and method for measuring the distance to a vehicle in front

    DE102006057552A1

  • Navigation for unmanned vehicles in particular

    EP2202601A2

  • Measuring device with scan functionality and adjustable receiving areas of the receiver

    EP3451021A1

  • CN000106197382B