METHOD AND DEVICE FOR MEASURING BODY DAMAGE

Description

[0001] The present invention relates to a method for measuring body damage, in particular hail damage such as dents, bumps and / or scratches, on motor vehicles, wherein a motor vehicle to be measured moves during the measurement relative to an illumination device and a measuring device comprising one or more area cameras and one or more 3D cameras, which are arranged on a stationary holding device.

[0002] Furthermore, the present invention relates to a device for measuring body damage, in particular hail damage such as dents, bumps and / or scratches, on motor vehicles, comprising a lighting device and a measuring device which are arranged on a stationary holding device in such a way that a movement of a motor vehicle to be measured relative to the lighting and measuring device is possible.

[0003] Methods and devices of the type in question have been used in practice for years. For example, DE 10 2007 013 883 A1 discloses a method and device for detecting and evaluating hail damage. The surface of a motor vehicle to be measured is digitally scanned using a laser scanner, based on pulse transit time, phase transit time, or light section triangulation. The laser scanner is moved mechanically relative to the motor vehicle to be measured, or the vehicle is moved at a constant speed relative to the stationary laser scanners using a feed device, e.g., using a towing mechanism or a sled-like structure.

[0004] The known device is disadvantageous in several respects. Firstly, measuring using laser scanners is time-consuming, which negatively impacts the device's throughput, and also involves complex evaluation of the measurement results. Secondly, to obtain precise measurement results, the known device requires a largely constant relative speed between the laser scanners and the vehicle. Accordingly, a complex mechanism is required, which either moves the entire measuring device, including the gantry and the attached laser scanners, or the vehicle itself.

[0005] DE 10 2014 106238 A1 discloses a method for determining the surface of an object, such as a vehicle body, using deflectometry. A known pattern is imaged on a surface of the object and recorded with a calibrated camera. The object is moved relative to the imaged pattern and the camera. In one embodiment, two cameras are provided, which simultaneously image the same area of ​​the surface with the imaged pattern, wherein the computing unit is configured to perform stereo deflectometry.

[0006] EP 1 837 623 A1 discloses a method for detecting a surface shape of an optically partially reflective surface, which works according to the principle of scanning reflectometry, in which at least one light pattern is imaged over the surface to be examined.

[0007] DE 10 2005 013614 describes a device for optical shape measurement and / or testing of objects, in which a binocular stereo method or photometric deflectometry, or a combination thereof, is used.

[0008] The present invention is therefore based on the object of designing and developing a method and device for measuring body damage of the type mentioned above in such a way that the aforementioned problems are largely avoided and precise surface measurements are possible in a simple manner. The device should have a simple design that allows for rapid assembly and flexible use in different environments.

[0009] According to the invention, the above object is achieved with regard to the method by the features of claim 1. The method in question is characterized in that an illumination structure consisting of adjacent light strips or lines is generated on the vehicle body by means of the illumination device, surface images of the vehicle body are created by means of the one or more area cameras by deflectometric detection of the illumination structure reflected by the vehicle body, depth images reflecting the shape of the vehicle are generated by means of the one or more 3D cameras, and defects on the body are identified in the surface images by means of a corresponding evaluation device and the identified defects are localized on the vehicle by projecting the surface images onto the depth images.

[0010] According to the invention, it has been recognized that damage detection can be significantly simplified with the highest level of accuracy by appropriately combining data from area-scan cameras, which use deflectometry to capture surface images of a lighting structure reflected from the vehicle body, with data from 3D cameras, which use distance measurements to capture depth images reflecting the shape of the vehicle. In a further embodiment of the invention, the proposed scanning principle utilizes the vehicle's own movement. This results in drastically increased compactness and transportability compared to the prior art described above, since complex mechanics for moving either the scanning system or the vehicle are no longer required. The device according to the present invention can therefore be easily transported and set up and used in a short time.Another significant advantage is that no prior vehicle data is required for scanning, as the 3D scan is performed by the scanning system itself. Furthermore, since a single pass of a vehicle through the scanning system is sufficient to fully inspect the vehicle, processing times are drastically reduced compared to the state of the art. The duration of the scanning process, including data processing and evaluation, is less than 5 minutes per vehicle when using the method according to the invention. This significantly accelerates the damage assessment process.

[0011] The dent scanning system according to the present invention is used in particular for the automatic detection and classification of hail damage to vehicle bodies. To record the damage, the vehicle slowly drives through or past the holding device equipped with lighting and cameras, while the area-scan cameras record the reflection of the lighting on the body. According to an advantageous embodiment, the size and position of each hail dent in the vehicle body is determined by subsequent evaluation of the images using image processing. From this, the repair costs can be calculated using recognized formulas. The system's primary use therefore lies in supporting vehicle experts at the hail experts' local hail stations. These hail stations are typically the contact points for customers who wish to have their hail damage inspected and processed under their insurance policy.

[0012] During the measurement of a motor vehicle, the position of the vehicle in the direction of travel relative to the measuring device is advantageously measured. The resulting vehicle position data is correlated with the recorded surface and depth images during data analysis, enabling the precise localization of identified defects on or in the vehicle body.

[0013] In a particularly advantageous manner, the recorded depth images are combined to create a 3D model of the vehicle, taking into account the respective position data obtained. Specifically, 3D data can be calculated from the individual depth images from the depth cameras, which are then combined using the position data to create a single, complete 3D model of the vehicle.

[0014] According to a further advantageous embodiment, a deformation image can be calculated from the recorded surface images for each of the area-scan cameras used as part of the data analysis. For this purpose, if necessary, after appropriate processing of the recorded individual images using a special image processing system that includes various filters, the recorded surface images can be shifted depending on the measured position of the vehicle and then added together.

[0015] Furthermore, it can be provided that a 3D deformation image of the entire vehicle body is generated as part of the data analysis. For this purpose, for example, the respective deformation images from the individual area-scan cameras can be combined using the 3D model of the vehicle.

[0016] According to a further advantageous embodiment, defects on the body can be classified according to type and size in the 3D deformation image using an image processing algorithm. The image processing algorithm used can, for example, be based on template matching, for which a set of different defects typically caused by hail damage must be learned in advance. Based on the learned defects, all identified defects can then be detected via image comparison. It can be provided that the image processing algorithm is designed such that it not only detects a defect (i.e., a classification), but also determines the size of the defect.

[0017] The problem in question is further solved by the features of claim 7. According to this, the device in question is characterized in that the lighting device has a lighting structure consisting of adjacent light strips or-lines are generated on the vehicle body, and that the measuring device comprises one or more area cameras and one or more 3D cameras, wherein the one or more area cameras are designed to create surface images of the vehicle body by deflectometric detection of the lighting structure reflected by the vehicle body, in which surface images defects on the body can be identified by means of a corresponding evaluation device, and wherein the one or more 3D cameras are designed to generate depth images reflecting the shape of the vehicle by means of distance determinations in such a way that by projecting the surface images onto the depth images, localization of the defects identified on the vehicle using the surface images is possible.

[0018] It should be noted that the previously discussed features of the method according to the invention can also have a device-related embodiment. A combination of these features with the features relating to the device claim is not only possible, but advantageous and expressly part of the disclosure.

[0019] Advantageously, both the lighting device and the measuring device are associated with a portal-like or arched support device. This support device is dimensioned to allow the passage of a motor vehicle.

[0020] According to an advantageous embodiment, both the number of area-scan cameras and 3D cameras used, as well as their arrangement on the holding device, are selected such that the entire vehicle body can be measured while the vehicle passes by or through the vehicle. This leads to a drastic reduction in processing times and thus to a significant increase in vehicle throughput.

[0021] With a view to flexible use of the device and rapid operational readiness of the device, one advantageous embodiment provides for the holding device to be constructed in a modular manner from individual segments or modules. This enables easy transport of the device in its dismantled state and thus supports flexible use of the device at different locations. The individual segments can be designed in such a way that they can be easily connected to one another using a plug-in, screw-in and / or locking mechanism. In a basic version, the segments can form a passage arch whose dimensions enable the measurement of all vehicles, from small cars to SUVs. In an extended version, additional segments can be provided which extend the passage arch in height and / or width so that larger vehicles, such as, for example,a small van or a minibus, can be measured.

[0022] According to a further advantageous embodiment, the device comprises a measuring system for determining the position of a motor vehicle to be measured relative to the measuring device as the motor vehicle passes the holding device or passes through the passage / archway. In a specific embodiment, the measuring system can have a distance meter, which, for example, comprises a laser distance sensor based on the time-of-flight method. Alternatively, the distance meter can also comprise a (consumer) depth camera operating according to the time-of-flight principle. With a view to positioning or arranging the distance meter in a way that enables motor vehicles to pass through smoothly, in both cases, i.e., when using a laser distance sensor as well as when using a depth camera, it can be provided that the distance meter is arranged in a drive-over threshold-like floor element.This floor element can be placed at a suitable distance, typically a few meters, in the direction of travel in front of the passage arch.

[0023] Additionally or alternatively, the measuring system can have a wheel camera system comprising a total of four cameras. These cameras can also be arranged on the holding device, although a separate arrangement is also conceivable. In any case, it is essential that the cameras are located at the height of the vehicle wheels of a motor vehicle to be measured and are arranged in pairs, so that there are two cameras on each side of a motor vehicle. The two camera pairs arranged in this way are designed such that they each recognize or detect the wheels of a motor vehicle and, using stereoscopy or stereovision, calculate the 3D position of all recognized or detected vehicle wheels. The 3D position of the motor vehicle can then be calculated from the 3D position of the vehicle wheels.

[0024] There are now various possibilities for advantageously embodying and developing the teaching of the present invention. For this purpose, reference is made, on the one hand, to the claims subordinate to claims 1 and 7 and, on the other hand, to the following explanation of preferred embodiments of the invention with reference to the drawing. In conjunction with the explanation of the preferred embodiments of the invention with reference to the drawing, generally preferred embodiments and developments of the teaching are also explained. The drawing shows: Fig. 1 in a schematic representation of an embodiment of an apparatus according to the invention for carrying out the method according to the invention, Fig. 2 in a schematic representation of an embodiment of a passage arch, Fig. 3 in a schematic representation of a plan view of the passage arch according to Fig. 2 , Fig. 4 in a schematic representation an interior view of a side module of the passage arch according to Fig. 2 , Fig. 5 in a schematic representation a cross-sectional view of a side module of the passage arch according to Fig. 2 , Fig. 6 a simplified flow chart to explain the data evaluation according to an embodiment of the method according to the invention, and Fig. 7 a simplified flow chart to explain the calculation of a deformation image according to an embodiment of the method according to the invention.

[0025] Fig. 1 shows, in a roughly schematic and not-to-scale view, an exemplary structure and the basic functional principle of an inventive device for carrying out an inventive method. The device according to the illustrated embodiment comprises a stationary holding device 1, a measuring system 2 for position determination, and an evaluation unit 3.

[0026] The holding device 1 is designed as a passage arch 4, which is dimensioned so that motor vehicles 5 of usual size to be measured, ie from small cars to SUVs, can pass through the passage arch 4, ie can drive under it, as in Fig. 1 by the arrow indicating the direction of movement of the motor vehicle 5. On the holding device 1, a (in Fig. 1 not shown) lighting device and a (in Fig. 1 A measuring device (also not shown) comprising one or more area cameras and one or more 3D cameras is arranged.

[0027] The measuring system 2 for position determination comprises in the Fig. 1 The illustrated embodiment includes a laser distance sensor 6, which, based on the time-of-flight method, continuously detects the relative distance of the vehicle 5 to be measured in the direction of travel from the measuring device as it passes through the holding device 1. To ensure that this measuring system 2, which is suitably positioned a few meters in front of the holding device 1 in the direction of travel, does not impede the passage of the vehicle 5, the laser distance sensor 6 is integrated into a drive-over element in the form of a speed bump 7.

[0028] It should be emphasized at this point that the use of a laser distance sensor 6, as described above and in Fig. 1 shown is merely an exemplary embodiment. Other variants for determining the position of the vehicle 5 while passing the cameras of the measuring device are also conceivable. For example, additional cameras could be provided to form a wheel camera system. In a specific embodiment, this wheel camera system could comprise a total of four cameras, which are arranged on the holding device 1 at the height of the vehicle wheels and stereoscopically detect the vehicle wheels in pairs on each side of the vehicle 5. From the 3D position of the vehicle wheels determined in this way, the relative position of the vehicle 5 with respect to the measuring device can then be determined.

[0029] As in Fig. 1 As indicated, both the measuring system 2 for determining the position of the vehicle 5 and the cameras of the measuring device arranged on the holding device 1 are connected to the evaluation unit 3 by means of corresponding cable connections. A wireless transmission of the measurement data to the evaluation unit 3 is also conceivable in principle.

[0030] Fig. 2 shows schematically an embodiment of a holding device 1 designed as a passage arch 4. The modular design of the holding device 1 is clearly visible, which enables quick and flexible assembly and disassembly of the device as well as uncomplicated transport. In the embodiment shown, the holding device 1 comprises a total of five modules, namely two flat side modules 8a arranged on the floor, a flat central module 8b, and two arched intermediate modules 8c, each of which is arranged between one of the side modules 8a and the central module 8b. The modules 8 are designed in such a way that they can be easily plugged together, with a pin formed on one of the modules 8 engaging in a corresponding groove formed on the other module 8. The plug-in connections produced in this way can be made by means of snap-in orSnap fasteners 13, which are also shown in the corresponding plan view according to . Fig. 3 can be seen. Although this method of connecting the modules 8 proves to be particularly suitable due to its stability and easy detachability, other mounting methods are also conceivable, for example, screw or snap-in connections.

[0031] The modules 8 are pre-assembled according to their intended use. This means that the respective cameras of the measuring device are already integrated into the modules 8 at locations suitable for the respective measurements. This can also be provided for the lighting device. If the intended use changes, individual modules 8 can be easily replaced or the setup can be expanded with additional modules 8. For example, the Fig. 2 The setup shown is for measuring normal cars (from small cars to SUVs). If the device is to be converted to measure small vans or minibuses, this can be easily achieved by inserting or adding two additional side modules 8a (to increase the clearance height) and an additional central module 8b (to increase the clearance width), which also incorporate the required cameras and lighting equipment as described above.

[0032] The lighting device arranged on the holding device 1 comprises several (in the Fig. 2-4 LEDs (not shown) that generate strip lighting on the inner side 9 of the passage arch 4. The lighting is designed to create a suitable reflection with a distinct structure on the vehicle body, which can be recorded by the area-scan cameras. Specifically, the lighting consists of a multitude of narrow, adjacent lines or stripes, resulting in alternating light and dark lines, preferably with sharp transitions. The line spacing is preferably constant, but can also vary.

[0033] For optimal scanning results, it is important that the reflection strips shine as homogeneously as possible. From a structural point of view, it can therefore be provided that the inner sides 9 of the modules 8 of the passage arch 4 have a matt-transparent layer, e.g. in the form of a Plexiglas plate or pane, which is backlit by means of suitable light sources. In particular, the layer or Plexiglas plate can be illuminated from behind with LED strips that are arranged at a certain distance from the layer or Plexiglas plate. The matt-transparent layer or Plexiglas plate can be filmed in such a way that those stripes or lines that are intended to appear dark in the lighting pattern to be generated are covered. In contrast, no cover is provided for the areas that are intended to appear as light stripes or lines. The distance of the illumination, e.g.The distance between the LEDs and the layer is chosen so that on the one hand as little light as possible is lost and on the other hand the individual light sources, e.g. the LEDs, of a light strip are not visible from the front, ie the light strip appears homogeneous.

[0034] It is essential for the most complete detection of a vehicle 5 during its passage through the passage arch 4 that the reflection is positioned as enclosingly as possible around the vehicle 5, so that the area cameras can see a reflection of the lighting in every body surface. Furthermore, with regard to ensuring optimal measurement results, it is advantageous if the reflection lines have a course that is as uninterrupted and kink-free as possible. Against this background, it is expedient that the intermediate modules 8c of the passage arch 4, as in the exemplary embodiment according to Fig. 2 , have a continuous curvature.

[0035] The Fig. 3-5 show different perspectives of the passage arch 4 according to Fig. 2 or sections thereof. In these figures it is clearly visible how the individual cameras of the measuring device are integrated into the modules 8 of the passage arch 4. For example, Fig. 5 , which represents a side module 8a of the passage arch 4 in cross-section, both in the upper and lower areas, an area camera 10 and a 3D camera 11 are provided. In the middle area and in the lower area, an additional camera 12 is provided, which, as described above, interact stereoscopically and form a wheel camera system for determining the position of the vehicle 5 during the passage. The cameras mentioned as well as the corresponding recesses in the interior paneling of the passage arch 4 are also particularly suitable in Fig. 2 clearly visible. A corresponding arrangement of cameras 10, 11, and 12 is located in the opposite side module 8a. To capture the top of the vehicle (i.e., in particular, the hood, roof, and trunk), an area-scan camera 10 and a 3D camera 11 are also provided in the central module 8b, installed in the same way. It is understood that, depending on the respective measurement situation, a different arrangement can be implemented, in which both the number of area-scan cameras 10 and 3D cameras 11, as well as the installation type, can deviate from the exemplary embodiment described above.

[0036] The area-scan cameras 10 are based on the measurement principle of deflectometry and generate surface images of the vehicle body by detecting the linear illumination structure reflected by the vehicle body. In one variant of the device, for example, area-scan cameras 10 of the type 'IDS UI-3140CP-C-HQ Rev. 2' are used. The 3D cameras 11, on the other hand, are used to generate depth images that reproduce the shape of the vehicle 5. In principle, commercially available consumer depth cameras can be used for this purpose, which, for example, operate according to the same or a similar functional principle as the 'Kinect for Xbox 360'.

[0037] The cameras 10 and 11 can either be pre-calibrated so that the measurement of vehicles can begin immediately after the scanning device has been set up. Alternatively, it can be provided that a one-time calibration process is carried out after the scanner has been set up. For this purpose, markers that can be detected by the cameras 10 and 11 are attached to a vehicle. The markers can, for example, be printed on reference plates, whereby the reference plates can be magnetized or can be attached to the vehicle body using suction cups at corresponding points on or at the vehicle body. The vehicle set up in this way then drives through the passage arch 4 twice at walking speed, with infrared images being taken by the 3D cameras 11 during the first pass and depth images being taken by the area cameras 10 during the second pass. In the two scans, the markers are detected by the 2D cameras 10 and 11.the 3D cameras 11 and from this the perspective transformation for the cameras and other relevant parameters are determined.

[0038] All cameras 10, 11, and 12 are directly connected to the evaluation unit 3. In a specific embodiment, the evaluation unit 3 can comprise a computer rack in which computer hardware for controlling the scanning system as well as evaluation computers for acquiring the camera and distance data and evaluating the data using image processing are installed. For ease of use, a touchscreen can be provided on the rack, which can be used to control the scanning system and view the measurement results.

[0039] While a vehicle 5 to be inspected moves slowly through the passage arch 4, the lighting device creates a reflection on the body that is distorted by the surface shape of the body. According to one embodiment, the vehicle body is continuously recorded by the area scan cameras 10 as the vehicle 5 passes through, such that the reflection is visible in the recorded surface images. Simultaneously, the depth or 3D cameras 11 record depth or 3D images of the body. The measuring system 2 for position determination, e.g., the laser distance sensor 6, supplies the current position of the vehicle 5 for each image taken by all cameras 10, 11. As described in detail below, defects in the body are identified in the surface images, if necessary after suitable preprocessing.By projecting the surface images onto the depth images, which is carried out on the basis of the respective data of the measuring system 2 for position determination, the identified defects are then localized on the vehicle 5.

[0040] Fig. 6 shows a simplified flowchart for explaining the data evaluation according to an embodiment of the method according to the invention. After the vehicle 5 has passed through the passage arch 4 once, a deformation image is first calculated as part of the data evaluation (step 601). This step is initially carried out separately for each area camera 10 by executing the image processing process according to Fig. 7 10 are scanned per area camera.

[0041] Accordingly, an edge filter is first applied to each image of the sequence recorded by an area-scan camera 10 so that only grayscale changes are visible (step 701). According to step 702, the edge-filtered image thus obtained is then smoothed with a blur filter, e.g., a Gaussian filter. To create a complete image of the vehicle 5, the processed camera images are shifted depending on the measured position of the vehicle 5 and added or combined with each other (step 703). Finally, according to step 704, bilateral filtering is performed, smoothing the fine line structure created by the reflection structure of the lighting device.

[0042] The deformation images thus generated by the individual area scan cameras 10 are then combined into a 3D deformation image using the 3D position data. Step 601 is thus completed.

[0043] In the subsequent step 602, the body parts in the deformation image are segmented using the 3D position data so that the scan result provides the size and position of each defect per body part.

[0044] According to step 603, 3D data is calculated from the individual depth images of the 3D cameras 11, which are then combined using the distance data to form a single, complete 3D model of the vehicle 5. The deformation images can be projected onto this 3D model, resulting in a complete spatial deformation image.

[0045] In step 604, defect detection takes place, with hail dents, in particular, being detected in the deformation image using an image processing algorithm based on template matching. For this purpose, a suitable set of typical hail dents is learned, which can then be used to detect all dents via image comparison. The algorithm is designed to not only recognize the type of specific defect, i.e., perform a classification, but also provide the size of a hail dent.

[0046] Finally, in step 605, the measurement results are visualized in a processed form. This can be implemented in various forms, e.g., as an exploded view of the body, in table format, or as an interactive 3D model. Based on the sizes and positions of the identified stops in the vehicle body, the corresponding repair costs can also be calculated and output or displayed using recognized damage types.

[0047] Within the scope of an advantageous embodiment, the scanning device specifically operates with ten area-scan cameras 10 and five 3D cameras 11, as well as an additional measuring system 2 for position determination, e.g., a distance sensor. It should be noted that a smaller number of cameras, for example, only five area-scan cameras 10, is also possible in principle. After recording the reflection and depth data, image preprocessing takes place, which includes the necessary perspective transformations or corrections of the individual recorded images and, if necessary, the application of further image processing steps. These further image processing steps can, for example, include a special direction-coding edge filter and / or an eigenvalue filter. The application of an eigenvalue filter generally has the advantage that dents can be made more visible in the subsequent overall image.

[0048] After completing the image preprocessing, as already described above, a 3D process is carried out by combining all recorded depth images into a complete 3D mesh or a complete depth map of the respective vehicle 5. In the next step, a dent or deformation image of the entire vehicle 5 is calculated based on this depth map by weighted addition of the preprocessed 2D images. In this dent or deformation image, the individual components of the vehicle body are segmented by detecting the component edges. Finally, dents are identified for each component in the dent or deformation image and classified based on learned dent patterns. Using recognized formulas, the repair costs, both component-specific and overall, can be calculated and output based on the identified and classified dents.

[0049] With regard to further advantageous embodiments of the device according to the invention or the method according to the invention, reference is made to the general part of the description and to the appended claims in order to avoid repetition.

[0050] Finally, it should be expressly noted that the above-described embodiments of the device and method according to the invention serve merely to explain the claimed teaching, but do not limit it to the embodiments. The invention is defined in the claims. Bezugszeichenliste

[0051] 1Holding device 2Measuring system for position determination 3Evaluation unit 4Arch 5Vehicle 6Laser distance sensor 7Threshold-like floor element 8aSide module 8bCentral module 8cIntermediate module 9Inside 10Area camera 113D camera 12Cameras of the wheel camera system 13Snap fastener

Claims

1. Method for measuring body damage, in particular hail damage, such as dents, bumps and / or scratches, on motor vehicles, wherein a motor vehicle (5) which is intended to be measured is moved during the measurement relative to an illumination device and a measurement device which comprises one or more area cameras (10) which are arranged on a fixed retention device (1), wherein by means of the illumination device an illumination structure comprising light strips or lines which are located beside each other is produced on the vehicle body, wherein surface images of the vehicle body are produced by means of the one or more area cameras (10) by means of deflectometric detection of the illumination structure reflected by the vehicle body, and wherein defects on the body are identified by means of a corresponding evaluation device (3) in the surface images, characterised in that the measurement device additionally comprises one or more 3D cameras arranged on the fixed retention device (1), wherein depth images which reproduce the shape of the vehicle are produced by means of the one or more 3D cameras (11) by means of distance determinations, and wherein by means of projection of the surface images on the depth images the identified defects on the vehicle (5) are located.

2. Method according to claim 1, wherein during the measurement of a motor vehicle (5) the position of the motor vehicle (5) is measured in the travel direction relative to the measurement device.

3. Method according to claim 1, wherein the recorded depth images taking into account the respective obtained position data of the motor vehicle (5) are combined in order to form a 3D model of the vehicle (5).

4. Method according to claim 3, wherein for each of the area cameras (10) a deformation image is produced by the surface images which are recorded in each case being displaced depending on the measured position of the vehicle (5) in each case and are then added together.

5. Method according to claim 4, wherein a 3D deformation image of the entire vehicle body is produced by the respective deformation images of the individual area camera (10) being combined using the 3D model of the vehicle (5).

6. Method according to claim 5, wherein in the 3D deformation image defects on the body are classified with regard to type and size by means of an image processing algorithm which is preferably based on template matching.

7. Apparatus for measuring body damage, in particular hail damage, such as dents, bumps and / or scratches, on motor vehicles, preferably for carrying out a method according to any one of claims 1 to 6, having an illumination device and a measurement device which are arranged on a fixed retention device (1) in such a manner that a movement of a motor vehicle (5) which is intended to be measured relative to the illumination and measurement device is enabled, wherein the illumination device produces an illumination structure comprising light strips or lines which are located beside each other on the vehicle body, and wherein the measurement device comprises one or more area cameras (10) which are configured to produce by means of deflectometric detection of the illumination structure reflected by the vehicle body surface images of the vehicle body, in which defects on the body can be identified by means of a corresponding evaluation device, characterised in that the measurement device additionally comprises one or more 3D cameras (11) which are configured to produce by means of distance determinations depth images which reproduce the shape of the vehicle in such a manner that by means of projection of the surface images on the depth images a location of the defects which are identified with reference to the surface images on the vehicle (5) is enabled.

8. Apparatus according to claim 7, wherein the illumination device and the measurement device are associated with a portal-like or archway-like retention device (1) which enables the passage of a motor vehicle (5).

9. Apparatus according to claim 7 or 8, wherein the number of area cameras (10) and 3D cameras (11) and the arrangement thereof on the retention device (1) are selected in such a manner that the entire vehicle body can be measured while the vehicle (5) is passing by or passing through.

10. Apparatus according to any one of claims 7 to 9, wherein the retention device (1) is constructed in a modular manner from individual segments or modules (8) which can preferably be connected to each other by means of a plug, screw and / or locking mechanism.

11. Apparatus according to any one of claims 7 to 10, further comprising a measurement system (2) for determining the position of a motor vehicle (5) which is intended to be measured relative to the measurement device during the passage of the motor vehicle (5) through the retention device (1).

12. Apparatus according to claim 11, wherein the measurement system (2) for position determination has a distance sensor (6) which is arranged in a threshold-like base element (7) which can be driven over, and / or wherein the measurement system (2) for position determination has four cameras (12), wherein at each side of a motor vehicle (5) which is intended to be measured two cameras (12) cooperate in pairs in each case and detect the two respective vehicle wheels and calculate by means of stereoscopy the 3D position of all the detected vehicle wheels.