DE102022002668A1 - Vehicle surface analysis device and method for vehicle surface analysis - Google Patents

Abstract

The invention relates to a vehicle surface analysis device, wherein an image camera arrangement has at least a first and a second image camera, which have essentially parallel image axes and a lateral distance and for a lateral relative movement to a vehicle and for detecting object points of the vehicle as image point data, at a first point in time t1 as a first image recording, at a second time t2 as a second image recording, at a third time t3 as a third image recording, each of the image cameras providing at least one of the image recordings, and wherein an evaluation unit is designed to include the first image recording a first time stamp, the second image recording with a second time stamp, the third image recording with a third time stamp, and to identify a reference object point of the vehicle based on the first image recording and to assign coordinates (x1, y1) to this identified reference object point, the identified reference object point based on the to recognize the second and third image recordings and to assign coordinates (x1', y1') or (x1", y1") to this reference object point, and using the time stamp, the camera distance, an angular position of further object points in the image recordings and the previously calculated relative speed to assign surface coordinates (xn, yn) to further object points and to assign depth coordinates (zn) to the surface coordinates (xn, yn) and to create a final image of the vehicle using the spatial coordinates (xn, yn, zn) obtained.

Description

The invention relates to a vehicle surface analysis device in which the surface of a vehicle can be analyzed using a stereo camera. The invention further relates to a method for vehicle surface analysis.

The state of the art describes technical solutions for vehicle surface analyzes in the form of both tactile and non-contact, non-destructive and optical methods and the associated devices.

The disadvantages of tactile methods are the low speed, the requirement for a specific working distance and the risk of damage to the vehicle surface to be detected.

In the case of non-contact methods, for example, in addition to ultrasound methods, laser triangulation methods, lidar measurements, time-of-flight recordings, purely image-recording methods are also known. Depending on the method, the disadvantages are in particular a low resolution, a small number of object points that can be detected or, in the case of image-recording methods, the requirements for the surfaces to be captured.

The typical structure of stereo cameras is also known here. At least two cameras are arranged at a known, clearly defined distance. Two photos are always taken from an overlapping detection area. Spatial information can be calculated from the different perspectives based on the parallax.

A major disadvantage of this method is the problem of recording the spatial coordinates of large object surfaces, such as vehicles, if the object to be analyzed protrudes beyond the image section to be recorded. It is only possible to a limited extent and with high computer-aided processing power to combine several recordings with partial sections into a complete recording. In addition, it is disadvantageous that a defined positional relationship between the object and the cameras is required. The alternative of recording only an entire image as a stereo image recording of large objects is also disadvantageous because of the low resolution. Another disadvantage of conventional stereo recordings is the susceptibility to interference, such as reflections or unfavorable light.

The object of the invention is to provide a device for detecting the spatial coordinates of the surface of vehicles, which can generate a fault-tolerant overall image of the vehicle surface, which is easy to use and inexpensive to produce.

The task is solved in relation to the vehicle analysis device by the features listed in claim 1 and in relation to the method by the features listed in claim 2. Preferred further developments result from the respective subclaims.

The vehicle surface analysis device according to the invention has an image camera arrangement and an evaluation unit as basic components.

The vehicle analysis device according to the invention creates a solution with which a high-resolution digital image of the three-dimensional surface of the vehicle can be generated from a vehicle that passes through the detection range of the image camera arrangement, whereby neither the speed of the movement nor the distance to the image camera arrangement need to be defined . This results in detail from the structure described below and the interaction of the image camera arrangement and evaluation unit as follows.

The image camera arrangement is constructed in the manner of a stereo camera and has at least a first image camera and a second image camera. The image cameras can be, for example, commercially available industrial cameras. The optical system and the resolution are adapted to the application.

Here, the first image camera has a first image camera detection area, which captures a vehicle at least in sections. The second image camera accordingly has a second image camera detection area, which also captures the vehicle at least in sections.

In addition, the image cameras have essentially parallel image axes, which are arranged at a lateral distance. The image axes are also referred to below as optical axes. The term lateral distance is understood to mean a lateral distance, with the image axes being arranged transversely, preferably orthogonally, to the extent of the lateral distance. In addition, the two image cameras are in a fixed positional relationship to one another.

The image cameras are designed to move laterally relative to the vehicle. The direction of movement of the vehicle deviates from the optical axes of the image cameras. Preferably, the direction of movement of the vehicle is orthogonal to the two parallel optical axes sen of the image cameras. A lateral relative movement is understood here to mean that the positional relationship between the vehicle and the image camera changes in an axis transverse to the optical axes of the image cameras and in their plane, this being preferably caused by a movement of the vehicle in front of a stationary image camera arrangement. Alternatively, an active movement of the image camera arrangement relative to a stationary vehicle is also possible.

The image capture areas of the image cameras are arranged so that the vehicle preferably travels completely through them. The measuring section is usually flat and optionally equipped with additional light sources. An enclosure can also advantageously be provided in order to minimize disruptive influences.

The image cameras are designed to capture object points of the vehicle as pixel data, which then form an image recording. The image camera is designed for recording at a first time as a first image recording, at a second time as a second image recording and at a third time as a third image recording.

According to the invention, each of the image cameras provides at least one of the image recordings. This also means that one of the at least three image recordings according to the invention is recorded by a different image camera than the other two of the at least three image recordings. For example, the first and second image recordings can be carried out by the first image camera and the third image recording can be carried out by the second image camera. However, it is also possible for only the first image to be captured by the first image camera and for the second and third images to be captured by the second image camera. The three image recordings that are essential to the invention for detecting the position and speed of the vehicle surface are also referred to below as the tracking-essential image recordings.

To calculate the final image capture, at least three image captures are required. Preferably, the three image recordings only capture sections of the vehicle surface and further image recordings of the vehicle are taken sequentially at a time interval so that the entire vehicle surface is recorded while the vehicle speed remains the same. The image cameras particularly preferably have a strip-shaped detection area transverse to the direction of movement of the relative movement, so that a large number of image recordings with a high resolution can be processed into the final image recording.

The evaluation unit is connected to the image camera arrangement and is designed to provide the first image recording with a first time stamp, the second image recording with a second time stamp and the third image recording with a third time stamp. For this purpose, the evaluation unit is preferably designed as a processor or computer that contains a corresponding image recognition module. Image recognition is optimized preferably on the hardware side by specialized graphics processors and on the software side by artificial intelligence algorithms, which look for certain easily recognizable features that can be recognized in all collected images.

Furthermore, the evaluation unit is designed to identify an object point of the vehicle based on the first image recording and to assign the coordinates (x1, y1) to this identified object point. This can advantageously be a striking, i.e. an easily recognizable and delimitable object point. The identified object point is hereinafter referred to as the reference object point.

It is then part of the solution according to the invention to recognize the identified reference object point of the vehicle based on the second image recording and to assign the coordinates (x1', y1') in the second image recording to this reference object point.

Furthermore, it corresponds to the solution according to the invention to recognize the identified reference object point of the vehicle based on the third image recording and to assign coordinates (x ₁ ", y ₁ ") to this reference object point in the third image recording.

The prominent reference object point is used to determine a 2D reference point, with the help of which the images can be put together in a position-referenced manner.

In addition, the evaluation unit is designed to determine a relative speed of the lateral relative movement of the vehicle by means of the time stamp, a distance between the first and the second image camera detection area, an angular position of the identified object point in the first, second and third image recording recordings.

This is based on the fact that the position of the identified reference object point is determined by an angular distance between the respective angular positions tion is marked, for example, in the first and second image recordings. The speed of the relative movement cannot be determined solely on the basis of this angular distance if the distance between the vehicle and the image cameras is not defined. Conversely, if the speed is not defined, the depth information cannot be determined. However, if the angular position of the identified reference object point in the third image recording is included in conjunction with the defined lateral distance between the two image cameras that is known for further calculation, the ratio of the lateral change in position of the identified reference object point between the images and its distance to the image cameras can be calculated. In addition to the time information available from the time stamps, the evaluation unit can calculate the relative speed. Based on the now known relative speed, the distance-time relationship is also known and can therefore be determined which position the vehicle occupies at which point in time. This means that at the same time the relative position for all object points is determined relative to one another between the image recordings.

Accordingly, the evaluation unit is designed to use the time stamp, the distance between the first and the second image camera detection area, an angular position of further object points in one of the first, second and third image recordings and the relative speed of these further object points to give the surface coordinates (x _n , y _n ) to assign. Depth coordinates (z _n ) are then assigned to the surface coordinates (x _n , y _n ) and a final image of the vehicle is created using the spatial coordinates (x _n , y _n , z _n ) obtained. In addition to the pure 2D image data, a spatial structure similar to a relief of the vehicle surface can be determined.

The vehicle surface analysis device according to the invention has in particular the following advantages.

A particular advantage is that high-quality 3D capture is possible even on problematic surfaces whose object points are sometimes not very distinctive and where reflections interfere with detection, as is the case with painted or glazed vehicle surface sections.

For this purpose, a reliable assignment of the positions of the object points between several image recordings can be achieved using the possible selection of prominent object points as features. It is advantageous here that such object points can be selected as reference object points that are particularly distinctive and are particularly suitable for feature tracking between the multiple image recordings. It is also possible that these prominent object points are less or not interesting as such for the final image recording, so that they only take on an auxiliary function for reliable feature tracking, hereinafter also referred to as feature tracking.

Based on this, the image points of the multiple recordings recorded by the object points can be assigned to one another even if their recording quality, viewed individually, is only low. In the final image recording, low recording qualities of an image point from one of the image recordings can be compensated for by higher recording qualities of the other image recordings. In this way, the pixels of the several recordings support each other.

Furthermore, a solution was advantageously found in which a comparatively large object such as a vehicle surface can be displayed in a final image recording in a high resolution using the stereo photogrammetry known from the prior art, without any complex positioning between the image cameras and the vehicle is required. Rather, the vehicle can simply be driven past the image cameras without having to set an exact distance.

It is also advantageously possible to generate further image recordings in addition to the three image recordings that are essential for tracking, so that image points from the additional image recordings are also available to form objects and can be put together to form a final image recording that is theoretically not limited in length.

This is also advantageously possible without the need for complex additional devices. Rather, the solution according to the invention can be based on a known and generally available stereo camera.

In this context, there is no need for an exact parallel alignment of the optical axes of the image cameras, since all that matters is determining the angular position and the lateral distance between the image cameras.

It is also advantageous that a vehicle surface analysis is made possible when vehicles are in motion. For example, a vehicle surface analysis can be carried out automatically in a workshop when a vehicle is in is brought into an inspection or repair area.

According to an advantageous development, the vehicle surface analysis device is characterized in that the evaluation unit is designed to assign brightness or color values (g _n ) to the spatial coordinates (x _n , y _n , z _n ) and by means of the data tuples (x _n , y _n , z _n , g _n ) to create the final image recording.

The final image recording can therefore advantageously contain both spatial coordinates and a visual image of the vehicle surface. For this purpose, the final image recording can preferably have several layers. Damage, deformation, paint repairs or other deviations from the standard can optionally be displayed directly using virtual reality technologies or processed using augmented reality to support repair decisions or procedures.

According to another development, the vehicle surface analysis device is characterized in that the optical axes of the image cameras are arranged in parallel.

The parallel arrangement advantageously simplifies the calculation process for determining the 3D relief of the vehicle surface.

According to another advantageous development, the vehicle surface analysis device is characterized in that the image axes of the image cameras are arranged orthogonally to the relative movement of the vehicle. The orthogonal arrangement offers the advantage that perspective distortions are kept as small as possible, which also leads to a reduction in the computing power and time required by the evaluation unit.

Another aspect of the invention relates to a method for vehicle surface analysis.

The description content of the device according to the invention applies in a corresponding manner to the method according to the invention, taking into account the following aspects.

1. a) Erfassen von Bildpunktdaten von Objektpunkten des Fahrzeugs mittels der Bildkameras zu einem ersten Zeitpunkt als eine erste Bildaufnahme und Übertragen an die Auswertungseinheit und Erzeugen eines ersten Zeitstempels.
2. b) Durchführen einer gleichförmigen lateralen Relativbewegung des Fahrzeugs.
3. c) Erfassen von Bildpunktdaten von Objektpunkten des Fahrzeugs mittels der Bildkameras zu einem zweiten Zeitpunkt als eine zweite Bildaufnahme und Übertragen an die Auswertungseinheit und Erzeugen eines zweiten Zeitstempels.
4. d) Fortsetzen der gleichförmigen lateralen Relativbewegung des Fahrzeugs.
5. e) Erfassen von Bildpunktdaten von Objektpunkten des Fahrzeugs mittels der Bildkameras zu einem dritten Zeitpunkt als eine dritte Bildaufnahme und Übertragen an die Auswertungseinheit und Erzeugen eines dritten Zeitstempels.

The method is carried out using the image camera arrangement and the evaluation unit and has the method steps described below.

1. a) Capturing pixel data from object points of the vehicle using the image cameras at a first time as a first image recording and transmitting it to the evaluation unit and generating a first time stamp.
2. b) Performing a uniform lateral relative movement of the vehicle.
3. c) capturing image point data from object points of the vehicle using the image cameras at a second time as a second image recording and transmitting it to the evaluation unit and generating a second time stamp.
4. d) Continuing the uniform lateral relative movement of the vehicle.
5. e) Capturing pixel data from object points of the vehicle using the image cameras at a third time as a third image recording and transmitting it to the evaluation unit and generating a third time stamp.

• f) Identifizieren eines Objektpunkts mittels der ersten Bildaufnahme als einen Referenzobjektpunkt und Zuordnen von Koordinatendaten (x, y).

In the group of method steps a) to e), the three image recordings according to the invention are created, with the vehicle executing the relative movement transversely to the optical axes of the image cameras. Preferably, the vehicle is driven past the image cameras to simplify the process. The more uniform the movement of the vehicle on the measuring route, the more accurate the measurement results are. However, from the large number of additional image recordings, an error correction for the speed on certain sections can be determined, which leads to an improvement in the result. As a result of the method steps, three image recordings are obtained, each with a time stamp, with two of the image recordings being recorded by one image camera and another image recording by the other image camera.

• f) identifying an object point using the first image recording as a reference object point and assigning coordinate data (x, y).

1. g) Erkennen des identifizierten Referenzobjektpunkts mittels der zweiten Bildaufnahme und zuordnen von Koordinatendaten (x', y') und erkennen des identifizierten Referenzobjektpunkts mittels der dritten Bildaufnahme und zuordnen von Koordinatendaten (x'', y'').

Advantageously, the image processing of the evaluation unit can select a prominent object point so that this can then also be reliably recognized in the two further recordings. In particular, it can also be an object point that is not of interest for the vehicle surface analysis as such. Optionally, several object points can also be identified as reference object points.

1. g) recognizing the identified reference object point using the second image recording and assigning coordinate data (x', y') and recognizing the identified reference object point using the third image recording and assigning coordinate data (x'', y'').

1. h) Bestimmen einer Relativgeschwindigkeit der lateralen Relativbewegung mittels
   • - der Zeitstempel,
   • - eines Abstands zwischen dem ersten und dem zweiten Bildkameraerfassungsbereich und
   • - einer Winkelposition des identifizierten Objektspunkts in der ersten, zweiten und dritten Bildaufnahme

This method step relates to the recognition of the reference object point identified in method step f) in one of the three image recordings in the other two images. This tracking of the feature is also known as feature tracking. According to the invention, the feature tracking is an aid in order to be able to establish a speed of the relative movement and thus a distance-time relationship between the three position-essential and optionally further image recordings in conjunction with the time stamps.

1. h) Determining a relative speed of the lateral relative movement using
   • - the timestamp,
   • - a distance between the first and the second image camera detection area and
   • - an angular position of the identified object point in the first, second and third image recordings

1. i) Zuordnen von Flächenkoordinaten (x_n, y_n) zu weiteren Objektpunkten und zuordnen von Tiefenkoordinaten (z_n) mittels,
   • - der Zeitstempel,
   • - des Abstands zwischen dem ersten und dem zweiten Bildkameraerfassungsbereich
   • - einer Winkelposition weiterer Objektspunkte in einer der ersten, zweiten und dritten Bildaufnahmeaufnahme
   • - und der Relativgeschwindigkeit

In this method step, the spatial distance of the identified reference object point is first calculated. This is done on the basis of the respective angular position in the image recordings and also on the basis of the known lateral distance between the image cameras. On this basis, the speed of the object can then be calculated based on the distance traveled and the time difference of the prominent reference object point between the image recordings.

1. i) assigning surface coordinates (x _n , y _n ) to further object points and assigning depth coordinates (z _n ) using,
   • - the timestamp,
   • - the distance between the first and the second image camera detection area
   • - an angular position of further object points in one of the first, second and third image recordings
   • - and the relative speed

1. j) Erzeugen einer finalen Bildaufnahme des Fahrzeugs mittels der erhaltenen Raumkoordinaten (x_n, y_n, z_n)

Due to the relative speed and the time information from the time stamps, the distance of the other object points can now also be determined, with the information of the other object points being expanded to include the depth values as a third dimension using triangulation. The spatial coordinate data is now available for all object points of interest.

1. j) Generating a final image of the vehicle using the spatial coordinates obtained (x _n , y _n , z _n )

In this final process step, the spatial coordinates of the individual object points are combined to form an overall image of the vehicle surface as a digital 3D object. This creates the final image of the vehicle.

The process therefore offers the advantage that a complete digital twin of the real vehicle surface can be created. This is done non-destructively and without tactile contact. The digital twin can now be used for electronic analysis of standard deviations and for visualization.

Furthermore, the advantages described in the vehicle surface analysis device according to the invention also apply in a corresponding manner to the method.

The method according to the invention can also be referred to as dynamic stereo photogrammetry, which according to the invention can advantageously also be carried out without defined distances between the vehicle surface and the image cameras.

In a further development, the method for vehicle surface analysis is characterized in that at least one further image recording and the generation of a further time stamp are carried out when the lateral relative movement continues.

This involves assigning area coordinates (x _n , y _n ) and depth coordinates (z _n ) to the other object points in the further image recording. The assignment of depth coordinates (z _n ) by means of the further time stamp, the distance between the first and the second image camera detection area, an angular position of the further object points in the further image recording and the relative speed also takes place accordingly. In method step j), a final image recording of the vehicle is generated using the spatial coordinates (xn, yn, zn) obtained, taking into account the further image recording.

This further development is based on the fact that the path-time relationship known from method steps a) to h) can be used to assign at least one further image recording beyond the three required tracking-essential image recordings. However, it is preferably a plurality of temporally sequential image recordings.

These one or more additional image recordings can be done both interpolatingly - i.e. between the first and the third image recording - and extrapolating before the first or after the third image recording.

It is advantageously even possible to use the three required tracking-essential image recordings only for the feature tracking and to provide the image points for the further object points from other, further image recordings.

This makes it possible to provide different conditions adapted to the respective requirements, such as exposure times, additional lighting sources, suitable recording angles and the like, for the tracking-essential image recordings and for the further image recordings. In this way, particularly reliable object point identification can be achieved in the tracking-essential image recordings on the one hand and, for example, particularly good color reproduction in the further image recordings on the other side.

According to a further advantageous development of the method for vehicle surface analysis, this is characterized in that brightness or color values (g _n ) from at least one image recording are assigned to the spatial coordinates (x _n , y _n , z _n ) and that the final image recording is done using the so obtained data tuple (x _n , y _n , z _n , g _n ) is created.

This further development provides a solution in which the final image capture contains information about the spatial structure of the vehicle surface as well as its visual appearance. This information can preferably be output through several layers of the final image recording.

• 1 schematische Darstellung der Fahrzeugoberflächenanalysevorrichtung beim Einfahren des Fahrzeugs in den Messbereich
• 2 schematische Darstellung der Fahrzeugoberflächenanalysevorrichtung beim Durchfahren des Fahrzeugs innerhalb des Messbereichs
• 3 schematische Darstellung der Fahrzeugoberflächenanalysevorrichtung beim Ausfahren des Fahrzeugs aus dem Messbereich

näher erläutert.The invention is illustrated as an exemplary embodiment using

• 1 Schematic representation of the vehicle surface analysis device as the vehicle enters the measuring area
• 2 Schematic representation of the vehicle surface analysis device as the vehicle drives through within the measuring range
• 3 Schematic representation of the vehicle surface analysis device as the vehicle leaves the measuring area

explained in more detail.

The same reference numbers in the different figures refer to the same features or components. The reference numbers are used in the description even if they are not shown in the relevant figure.

The 1 represents the schematic structure of an exemplary embodiment of the vehicle surface analysis device while the vehicle 5 enters the detection area.

The device consists of the image camera arrangement 1 and the evaluation unit 4. The image camera arrangement 1 with the first image camera 1.1 and the second image camera 1.2 is connected to the evaluation unit 4, so that the image recordings are transmitted to the evaluation unit 4.

The first image axis 3.1 of the first image camera 1.1 and the second image axis 3.2 of the second image camera 1.2 are arranged parallel to one another at a defined distance 3.3. The first image camera 1.1 has the first image camera detection area 2.1 and the second image camera 1.2 has the second image camera detection area 2.2.

To carry out the vehicle surface analysis method, the vehicle 5 moves laterally to the image axes 3.1, 3.2 of the image cameras 1.1, 1.2 into their image camera detection areas 2.1, 2.2. This is shown by the arrow pointing to the left. In the exemplary embodiment, a first image is recorded by the second image camera 1.2 at a first time t ₁ . The angular position of an object point that is identified by the evaluation unit 4 and hereinafter referred to as reference object point 6 is illustrated by an arrow emanating from the second image camera 1.2.

The 2 shows a time-advanced state at the second time t ₂ , at which the vehicle 5 is already moving further through the image recording area. At this second time t ₂ - in the exemplary embodiment also by the second image camera 1.2 - the second image recording of the vehicle 5 is created, transmitted to the evaluation unit 4 and provided with a time stamp. The arrow starting from the second image camera 1.2 illustrates the changed angular position of the reference object point 6 compared to the first image recording 1 .

The 3 shows a schematic representation of the vehicle 4 shortly before leaving the image recording area. Here, the third image recording is created at time t ₃ , transmitted to the evaluation unit 5 and also provided with a time stamp. In contrast to the first two image recordings, this third image recording in the exemplary embodiment is recorded by the first image camera 1.1, with the reference object point now being defined in its position by the angular position in the first image camera 1.1 and additionally by the lateral distance 3.3.

The three image recordings are then analyzed in the evaluation unit 3. During the evaluation, an object point, which here is the reference object point 6, is recognized and tracked as a special feature, so that the relative speed of the vehicle 5 can be determined.

Based on the time stamp and the calculated relative speed of the vehicle 5, which defines the distance-time relationship, image points of other object points recorded in the image recordings can be assigned to one another.

The stereo camera recordings can be used to create a relief of the vehicle surface, in which each object point is described as an image point with a number tuple (x _n , y _n , z _n ) as the spatial coordinate. The final image recording thus generated is output by the evaluation unit 5 as a digital image of the vehicle surface and can optionally be further processed electronically.

Reference symbols used

1

Image camera arrangement

1.1

first image camera

1.2

second image camera

2.1

first image camera detection area

2.2

second image camera detection area

3.1

first image axis

3.2

second image axis

3.3

lateral distance

4

Evaluation unit

5

vehicle

6

Reference object point

Claims (7)

Vehicle surface analysis device, comprising an image camera arrangement (1) and an evaluation unit (4), the image camera arrangement (1) having at least a first image camera (1.1) and a second image camera (1.2), the first image camera (1.1) having a first image camera detection area (2.1 ), which captures a vehicle (5) at least in sections, the second image camera (1.12) having a second image camera detection area (2.2) which captures the vehicle (4) at least in sections, the image cameras (1.1, 1.2) having essentially parallel image axes (3.1, 3.2), are arranged spaced apart from one another at a lateral distance (3.3) and are in a fixed positional relationship to one another, the image cameras (1.1, 1.2) being capable of a lateral relative movement to the vehicle (5) with a direction of movement that is from the image axes (3.1, 3.2) of the image cameras (1.1, 1.2), are designed, wherein the image cameras (1.1, 1.2) are designed to capture object points of the vehicle (5) as image point data, at a first time t ₁ as a first Image recording, at a second time t ₂ as a second image recording, at a third time t ₃ as a third image recording, each of the image cameras (1.1, 1.2) providing at least one of the image recordings, the evaluation unit (4) being connected to the image camera arrangement (1 ) is connected and designed to provide the first image recording with a first time stamp, the second image recording with a second time stamp, the third image recording with a third time stamp, the evaluation unit (4) being designed to provide a reference object point (6) of the vehicle (5 ) to identify based on the first image recording and to assign coordinates (x1, y1) to this identified reference object point, to recognize the identified reference object point of the vehicle (4) based on the second image recording and to assign coordinates (x1 ', y1') to this reference object point and the identified reference object point of the Vehicle (5) to recognize based on the third image recording and to assign coordinates (x ₁ ", y ₁ ") to this reference object point, the evaluation unit (3) being designed by means of the time stamp, the distance (3.3) between the first and the second image camera detection area (2.1, 2.2), an angular position of the identified reference object point (6) in the first, second and third image recording, to determine a relative speed of the lateral relative movement of the vehicle (5) and is further designed, by means of the time stamp, of the distance (3.3) between to assign surface coordinates (x _n , y _n ) to the first and second image camera detection areas (2.1, 2.2), an angular position of further object points in one of the first, second and third image recordings and the relative speed of these further object points and to the surface coordinates (x _n , y _n ) To assign depth coordinates (z _n ) and to create a final image of the vehicle (5) using the spatial coordinates (x _n , y _n , z _n ) obtained. Vehicle surface analysis device Claim 1 , characterized in that the Evaluation unit (4) is designed to assign brightness or color values (g _n ) to the spatial coordinates (x _n , y _n , z _n ) and to create the final image recording using the data tuples (x _n , y _n , z _n , g _n ). Vehicle surface analysis device Claim 1 and 2 , characterized in that the first image axis (3.1) of the first image camera (1.1) and the second image axis (3.2) of the second image camera (1. 2) are arranged parallel to one another. Vehicle surface analysis device according to one of the preceding claims, characterized in that the image axes (3.1, 3.2) of the image cameras (1.1, 1.2) are arranged orthogonally to the relative movement of the vehicle (5). Method for vehicle surface analysis, characterized in that by means of the image camera arrangement (1) and the evaluation unit (4) according to one of the preceding claims, has the following method steps: a) acquiring image point data from object points of the vehicle (5) using the image cameras (1.1, 1.2). a first time t ₁ as a first image recording and transmitting it to the evaluation unit (4) and generating a first time stamp, b) carrying out a uniform lateral relative movement of the vehicle (5), c) acquiring image point data of object points of the vehicle (5) by means of the Image cameras (1.1, 1.2) at a second time t ₂ as a second image recording and transmitting it to the evaluation unit (4) and generating a second time stamp, d) continuing the uniform lateral relative movement of the vehicle (5), e) acquiring image point data of object points of the vehicle (5) by means of the image cameras (1.1, 1.2) at a third time t ₃ as a third image recording and transmitted to the evaluation unit (3) and generating a third time stamp, f) identifying an object point as a reference object point (6) by means of the first Image recording and assigning coordinate data (x, y), g) recognizing the identified reference object point (6) using the second image recording and assigning coordinate data (x', y,') and recognizing the identified reference object point using the third image recording and assigning coordinate data ( x'', y,''), h) determining a relative speed of the lateral relative movement by means of - the time stamp, - a distance (3.3) between the first and the second image camera detection area (2.1, 2.2) and - an angular position of the identified reference object point (6 ) in the first, second and third image recordings, i) assigning surface coordinates (x _n , y _n ) to further object points and assigning depth coordinates (z _n ) using, - the time stamp, - the distance (3.3) between the first and the second image camera detection area (2.1, 2.2), - an angular position of further object points in one of the first, second and third image recording recordings, - and the relative speed, j) generating a final image recording of the vehicle (5) using the spatial coordinates obtained (x _n , y _n , z _n ) Method for vehicle surface analysis Claim 5 characterized in that when the lateral relative movement continues, at least one further image recording and generation of a further time stamp is carried out, that area coordinates (x _n , y _n ) and depth coordinates (z _n ) are assigned in the further image recording to the further ones Object points and an assignment of depth coordinates (z _n ) is carried out by means of the further time stamp, the distance (3.3) between the first and the second image camera detection area (2.1, 2.2), an angular position of the further object points in the further image recording and the relative speed, as well that in method step j), the final image recording of the vehicle (5) is generated, taking into account the further image recording using the spatial coordinates (x _n , y _n , z _n ) obtained. Method for vehicle surface analysis according to one of the preceding Claims 5 or 6 , characterized in that brightness or color values (g _n ) from at least one image recording are assigned to the spatial coordinates (x _n , y _n , z _n ) and the final image recording and by means of the data tuples (x _n , y _n , z _n , g _n ) is created.

Images