Detection of individual free marked target areas from images of a camera system of a mobile vehicle

Abstract

The present invention relates to a method and device for detecting individual, free marked target areas (46), such as parking areas, landing sites, docking points, etc., from images (R-4) of a camera system (2-i) of a mobile vehicle (10) by a machine learning system, as well as a method for training the machine learning system. The method for segmenting and identifying the individual free marked target regions (46) includes the following steps: - capturing an image (R-4) of the surroundings of at least one mobile means (10) by a camera system (2-i); - segmenting (S-4) each free marked target area (46) from at least one image (R-4) by a trained machine learning system according to any one of the preceding claims; and - outputting to the control unit (5) first segments (35) corresponding to the respective free marked target areas (46) and at least one further class of second segments (30; 32) of the captured periphery of the mobile means (10).

Description

[Technical field]

[0001] The present invention uses a machine learning system to extract individual Vacant The present invention relates to a method and apparatus for detecting marked target areas, such as parking areas, landing sites, docking sites, etc., and a method for training a machine learning device. [Background technology]

[0002] EP2486513B1 discloses a driver assistance system for a motor vehicle with a camera for recognizing road markings: a device for recognizing the parking / stopping mode of the vehicle is provided, and a control device receives image data of the surroundings of the vehicle from the camera in the parking / stopping mode of the vehicle and evaluates road markings indicating parking / stopping prohibition, the control device controls a signaling means to issue a warning when detecting a road marking indicating parking / stopping prohibition around the vehicle to be stopped or parked. Instead of one camera, multiple cameras with multiple viewing directions (e.g. to the side of the vehicle) and mounting positions (e.g. in the exterior mirrors) or image acquisition units can also be used.

[0003] DE102018214915A1 discloses a system and a method for a vehicle to recognize a parking space in the direction of travel. The space must have at least a predefined length and / or at least a predefined width, which corresponds to the length and width of the vehicle. The sensor is configured to be able to determine the outer contour of the boundary. The sensor can be an individual sensor, for example a camera, or a plurality of sensors, for example a plurality of cameras, or a plurality of types of sensors, for example a camera and a radar sensor.

[0004] The map is configured such that an outer contour of a boundary can be assigned to a location within the map. A placement module is configured to place a rectangle having a predefined length and a predefined width in a space between the boundaries, and if the rectangle can be placed in the space, the identified space can be considered suitable for accommodating a vehicle. The boundary can also include road markings.

[0005] These road markings are, for example, restricted areas, pedestrian crossings, parking space boundaries, etc. In this regard, in some embodiments, these road markings can be taken into account in a different way than the outer contour: for example, a space that is large enough from the outer contour but not large enough considering the parking space boundaries requires the driver of the vehicle to decide to "ignore the parking space boundaries."

[0006] EP3731138A1 discloses an image processing device and an on-board camera system that uses edge recognition to recognize edge marks at the ends of lines that indicate parking spaces. Based on the recognized edge marks, a provisional parking space frame is set and verified. In this way, Vacant Parking spaces can be identified. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] EP2486513B1 [Patent Document 2] DE102018214915A1 [Patent Document 3] EP3731138A1 Summary of the Invention [Problem to be solved by the invention]

[0008] The object of the present invention is to provide a method for detecting a VacantThe object of the present invention is to provide an improved solution for identifying target regions. [Means for solving the problem]

[0009] The solution is the subject of the independent claims. Advantageous developments are the subject of the dependent patent claims.

[0010] The starting point for developing the solution were the problems visible in possible concepts in the field of automated parking.

[0011] One possibility is based on parking lines and Vacant The aim of this study is to use an algorithm for detecting parking spaces based on spatially independent detection. In a first step, a top-view image is created using static camera calibration and flat world assumptions to transform curved lines (fish-eye domain) into straight lines (virtual pinhole camera). On this image, traditional image processing (e.g. Hough transform) is performed to recognize parking lines for model-based parking space candidate detection. Additionally, a second algorithm (e.g. occupied grid ) by Vacant Extract spatial awareness. Using this second piece of information, Vacant Parking spaces can be identified.

[0012] The key prerequisites for this possible concept are a correct calibration of the cameras and a flat perimeter. An incorrectly calibrated camera, e.g. due to uneven loading of the vehicle, or an uneven perimeter (parking on a slope), will result in a poor quality top view and therefore a poor quality of the parking space recognition. In addition, Vacant Spatial recognition typically relies on point clouds based on sparse objects, which provide very little information for uniform surfaces, e.g. pavement, and are only possible by using rigid assumptions.

[0013] Thus, within the scope of the present invention, a novel solution concept is pursued.

[0014] From the image of the camera system for capturing the surroundings of the mobile means (10), individual Vacant Segmentation of target areas marked, e.g. parking areas for vehicles, landing points for planes and drones, docking points for spacecraft and robots Station and a method for training a machine learning system to perform classification, comprising: semantic Segment Station In this category, we assume the following: Station Data When a learning input image is input, the machine learning system generates a learning target segment. Station Data The training data is used to adjust the parameters of the machine learning system so as to generate output data similar to the training target segment. Station Data This adjustment can be performed by minimizing a function that mathematically describes the difference between the target segment for training. Station Data is an individual Vacant Each of the first segments includes a first segment (e.g., a location and extent or size of the segment) corresponding to a marked target area, wherein each of the first segments includes a single Vacant The training target segment must correspond exactly to the marked target region (i.e., its instance). Marked means, for example, that the target region is indicated by a border or that the target region is visually recognizable in some other way. Station Data includes at least one second segment corresponding to at least one further class of captured surroundings of the vehicle. In many cases, there are multiple second segments corresponding to different classes. The machine learning system includes, for example, an artificial neural network whose parameters (e.g., weights) are adjusted by the training data.

[0015] In one embodiment, the training target segment Station Dataindicates a type of the first segment. If the target area corresponds to a parking space, the type of the parking space can be, for example, a parking space for the disabled, a parking space for women only, an electric vehicle parking space, a parking space for children, etc.

[0016] In one implementation variation, the second segment includes objects in the vicinity of the vehicle and indicates at least one class information of each object. semantic ) Example of segment class information: Buildings, signs, plants, cars, motorbikes, towed vehicles, containers, people, animals, etc. Road markings can also be objects: for example, lane lines, parking lot lines 52, road arrows, road traffic symbols, stop lines, etc.

[0017] In one embodiment, the camera system includes a camera with a fisheye lens. The fisheye lens camera has a capture angle of more than 180°, for example 195°. In this case, the training input images are images of the fisheye lens camera. These can be used without rectification ("rectification" = correction of the optical imaging properties of the fisheye lens). In this case, the training target segment of the fisheye map is Station Data This embodiment uses a fisheye lens image as a base for segmentation. Station However, it offers the advantage of being more robust than a rotating camera system.

[0018] The (computer-implemented) segmentation method according to the present invention Station and collect individual images from the camera system of the vehicle. Vacant A method for identifying a marked target area includes the following steps: - capturing an image of the surroundings of at least one vehicle by means of a camera system; VacantThe marked target region is segmented from at least one image by a trained machine learning system according to any one of the preceding claims. Station Steps to do; - (Identified) Individual Vacant outputting a first segment corresponding to the marked target area and a second segment of at least one further class of the captured periphery of the mobile means to a control unit, the control unit being adapted to determine based on the received segments: Vacant The movement of the mobile means to the target area is supported or performed autonomously, i.e. fully automatically. For this purpose, the segments can be translated into information about the geometry of the real surroundings, for example via a precisely calibrated camera system. It can also be assumed that the surrounding surface on which the mobile means is present is essentially planar. Many camera systems offer special options for three-dimensional reconstruction of the captured surroundings, for example stereo methods by triangulation using two individual cameras with overlapping viewing areas.

[0019] In one embodiment, the camera system is positioned to capture a 360° view of the vehicle; composition The camera includes multiple (fisheye) cameras.

[0020] According to one embodiment, the control unit determines the distance of the vehicle based on the output segments. Vacant Support for navigation to the target area can be implemented by visual displays. For example, the surroundings of the means of travel can be visualized, whereby: Vacant The target area can be visually highlighted, and the visualization can be displayed on a display of the vehicle.

[0021] In one embodiment, the control unit determines the location of the vehicle based on the output segments. VacantSupport for movement towards the target area is carried out acoustically, haptically or the movement of the means of transport is partly effected by control measures (for example steering support or braking support in a parking process).

[0022] According to one embodiment, the control unit determines the distance of the vehicle based on the output segments. Vacant Fully automatic movement to the target area, for example autonomous parking or fully automatic parking, is performed.

[0023] In one embodiment, the control unit outputs Vacant Based on the segment that corresponds to the marked target area, Vacant Information to verify the location of the marked target area is available outside the vehicle; Communication Equipment The information is transmitted (wirelessly) to the Communication Equipment is, for example, a control center of a car park, or Telematics Providers This allows the cloud to be the backbone of Vacant Parking spaces can be marked on a map as well as offered to other traffic participants via V2X.

[0024] In one embodiment, the means for moving is a vehicle; Vacant The marked target area is a parking space or (in this case) an area for inductive charging of an electric vehicle, so that the electric vehicle can be positioned precisely at the inductive charging pad.

[0025] A further object of the present invention is to provide an individual Vacant The marked target area is segmented from the images of the vehicle's camera system. Station The present invention relates to an apparatus or system for detecting and identifying a target object.

[0026] The apparatus includes: - an input unit configured to receive at least one image of the camera system; - a data processing unit adapted to carry out the method according to the invention; and -Individual Vacant an output unit configured to output a first segment corresponding to the marked target area and at least one further class of second segments of the captured periphery of the mobile means to the control unit.

[0027] The apparatus or data processing unit particularly preferably includes a microcontroller or processor, a central processing unit (CPU), a graphic processing unit (GPU), a digital signal processor (DSP), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), and the like, and associated software for performing the method steps.

[0028] The invention further relates to said camera system, said control unit, as well as to a means of transport incorporating the device according to the invention.

[0029] The invention further provides that when a data processing unit is programmed with the aid of said programming, said data processing unit is programmed with individual Vacant The marked target area is segmented from the image of the vehicle's camera system. Station The present invention also relates to a computer program element with instructions for carrying out the method for identifying.

[0030] The present invention further encompasses a computer readable storage medium having corresponding program elements stored therein.

[0031] Thus, the invention can be implemented in digital electronic circuitry, computer hardware, firmware, or software.

[0032] In other words, the present invention provides a method for the instantiation of a neural network to achieve a more efficient automation of the automated parking process. Vacant It allows direct and highly accurate detection and vectorization of parking spaces. Both labeling and detection are performed directly on the fisheye (195°) image without additional (top-view) rectification. Vacant The accuracy of the shape of the marked target area is improved.

[0033] One aspect is the fisheye domain. semantic Segment Station by Vacant It concerns the direct detection of target regions. For this, the existing list of detectable classes (curb, road, person, car, lane line) can be extended with the class "parking space", label data can be generated and the neural network can be trained on the corresponding training data. The areas between parking lines are then detected and directly output by the algorithm. An independent detection of the bounding parking lines can be performed in the context of the class "road markings". In the last step, the points are fused into clusters in order to provide a geometric description in polygonal form to the following environment model and enable trajectory planning.

[0034] Additionally, height information determined indirectly from the classification of static and dynamic objects can also be output to support trajectory planning / parking assist functions.

[0035] A significant advantage of the present invention is that direct parking space detection simplifies known methods, where a parking space must first be defined using existing parking markers that must be manually adapted to the variety of line markings around the world (white lines, lines with red blocks, etc.), which is tedious and prone to errors.

[0036] That is, in direct parking space detection, the neural network used detects parking spaces directly as they are, and can be trained to learn and adapt to all national variations within a manageable amount of effort (recording, labelling, tracking).

[0037] This approach also directly handles situations that make parking impossible (people or objects in the parking space). Vacant Only the area Vacant It is used as a parking space.

[0038] Applications of the invention are generally marking target areas that must be reached precisely. Examples include automated inductive charging areas, helicopter landing spaces, automated parcel delivery areas, etc. Landing site etc.

[0039] The embodiments and figures are explained in detail below. [Brief description of the drawings]

[0040] [Figure 1] 1 shows a first schematic representation of an apparatus according to the invention in an embodiment; [Diagram 2] 2 shows a second schematic representation of an on-board device according to the present invention in an embodiment; [Diagram 3] 1 shows a system with a neural network for detecting individual vacant marked target areas. [Figure 4] 1 illustrates a neighborhood scenario in which individual vacant marked parking spaces are detected. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0041] As is clear from FIG. 1, the image data of the multiple cameras in the surround view system are used to generate individual Vacant The device 1 according to the invention for detecting a marked target area can comprise several units or circuit components. In the embodiment shown in Fig. 1, the device 1 comprises several (single) cameras 2-i each producing a camera image or video data. In the embodiment shown in Fig. 1, the device 1 comprises four cameras 2-i for producing camera images. The number of cameras 2-i can be different for different applications. The device 1 according to the invention comprises at least two cameras for producing camera images.

[0042] The device 1 includes a data processing unit 3 which processes the images produced by the cameras 2-i. For example, the camera images can be combined into a single overview image. As shown in FIG. 1, the data processing unit 3 comprises an artificial neural network 4 which, by machine learning methods, processes the input image data (Ini) of the cameras (2-i) into semantic Segment Station And at that time, Vacant It is trained to instantiate a marked target area, e.g., a parking space with boundary indication. semantic Segment Station The result of the above is output from the image processing unit 3 to the control unit 5. The control unit 5 Vacant Support movement to a target area, or control autonomous movement of a moving means to a target area.

[0043] Fig. 2 shows a further schematic representation of one embodiment of the device 1 according to the invention. The device 1 shown in Fig. 2 is employed in a surround view system of a vehicle 10, in particular a passenger car or a motor vehicle. Here, the respective cameras 2-1, 2-2, 2-3, 2-4 are arranged on different sides of the vehicle 10 and have corresponding viewing areas (dashed lines), i.e. front=V, rear=H, left=L and right=R of the vehicle 10. For example, the first on-board camera 2-1 is arranged in front of the vehicle 10, the second on-board camera 2-2 is arranged in the rear of the vehicle 10, the third on-board camera 2-3 is arranged on the left side of the vehicle 10 and the fourth on-board camera 2-4 is arranged on the right side of the vehicle 10. These on-board cameras 2-i can be so-called fish-eye cameras with a viewing angle of at least 195° in one possible embodiment. These on-board cameras 2-i can communicate camera images, or camera image frames, or video data, in one possible embodiment, via an Ethernet connection to a data processing unit 3. The data processing unit 3 can create a synthesized surround view image from the camera images of the on-board cameras 2-i, which can be displayed to the driver and / or passengers via the control unit 5 and the display. Vacant Each parking space can be highlighted. The display is one way of supporting the driver to maneuver into the marked parking space. In the case of (semi-)automated vehicles, the vehicle can (semi-)autonomously: Vacant A control unit 5 is responsible (in part) for controlling the vehicle (acceleration, braking, driving direction, forward / reverse switching, steering) so that it can navigate to a parking space. This embodiment supports "Environment Perception" which allows extraction of instantiated parking spaces based on the surround view.

[0044] Figure 3 shows individual Vacant Marked target area semantic Segment StationThe figure shows a system with an artificial neural network 4 for performing segmentation and classification. At the top, a very simplified image (R-4) from a fisheye lens camera is shown, in which for example a parking lot with marking lines, a vehicle, a road surface and, in the background, street lamps, trees, bushes and buildings can be recognized. During training, the artificial neural network 4 performs segmentation on this image, as partially shown in the lower part of figure 3. Station Here, we learn to assign semantic Segment Station A class is assigned to each pixel by the above formula. In particular, the object class, such as the vehicle 30 parked on the left side, is important. The white line 32 indicating the parking space is also semantic Segment Station On the other hand, the surface of the parking space does not necessarily have to be a lane surface. However, the artificial neural network 4 can be Vacant The marked parking space 35 (instance) is a segment Station Here, multiple individual Vacant A parking space 35 is located to the right of the rightmost parked vehicle 30, and there is only one parking space on the left of the vehicle 30. Vacant Parking space 35 is present. After training, the artificial neural network attributes segments with corresponding recognized parking space instances to previously unseen fisheye camera images.

[0045] 4 shows, from a bird's-eye view, a schematic representation of a car park 44, which may be, for example, a supermarket or one floor of an indoor car park. The car park 44 has marking lines 52 which define (individual) parking spaces 46. The vehicle 20 arrives at the car park 44 and VacantA parking space 46 is required. Other vehicles 50 are already parked in the respective marked parking spaces. Ideally, the other vehicles 50 are within the lines 52 of the parking spaces in which they are parked. During the movement, the ego-vehicle 20 takes images (R-4) using at least one (fisheye lens) (2-1; 2-2; 2-3; 2-4) thereby capturing the parking space 44, the marking lines 52 as well as the other vehicles 50. The images (R-4) from the on-board cameras (2-1; 2-2; 2-3; 2-4) are processed by a data processing unit 3. A trained artificial neural network 4 of the data processing unit 3 derives from the images semantic Identify the segments assigned to a class. semantic Examples of segment classes include: For example, objects such as buildings, signs, plants, vehicles 50, motorcycles, towed vehicles, containers, people, animals, etc., as well as road markings such as lane lines, parking lot lines 52, road arrows, road traffic symbols, stop lines, etc.

[0046] Based on the training described above with reference to FIG. 3, the artificial neural network detects individual Vacant Parking space 46 semantic Segment Station This allows the vehicle 20 to automatically recognize the following: Vacant The car can be parked in parking space 46 . This application relates to the invention described in the claims, but also includes the following as other aspects. 1. From the image of the camera system (2-i) for capturing the surroundings of the mobile means (10), Vacant Segmentation of the marked target area (46) Station and a machine learning system for performing classification of images. semantic Segment Station A method for training in the above category, comprising: StationThe learning data including (S-4) is provided, and the parameters of the machine learning system are set by using the learning data (R-4, S-4) to obtain a learning target segment when the machine learning system receives a learning input image (R-4). Station (S-4) is adjusted to generate output data similar to the target segment for learning, Station The method, wherein (S-4) comprises: -Individual Vacant a first segment (35) corresponding to a marked target area (46); and at least one second segment (30; 32) of at least one further class of the captured periphery of the mobile means (10). 2. Learning Objective Segment Station The method of claim 1, wherein (S-4) additionally indicates the type of the first segment (35). 3. 3. The method according to claim 1 or 2, characterized in that the second segment (30; 32) contains an object (50; 52) and at least one class information of said object. 4. Any one of the above methods, characterized in that the camera system (2-i) includes a camera (2-1; 2-2; 2-3; 2-4) equipped with a fisheye lens, and the learning input image (R-4) is an image of the fisheye lens camera. 5. A segment that includes the following steps: Station and extracting individual images (R-4) from the camera system (2-i) of the mobile means (10). Vacant Method for identifying a marked target area (46): - capturing an image (R-4) of the surroundings of at least one mobile means (10) by a camera system (2-i); -Individual Vacant The marked target region (46) is segmented from at least one image (R-4) by any one of the trained machine learning systems described above. Station(S-4) Step; -Individual Vacant outputting to the control unit (5) the first segment (35) corresponding to the marked target area (46) and at least one further class of second segments (30; 32) of the captured periphery of the mobile means (10). 6. Any one of the above methods, characterized in that the camera system (2-i) includes at least one camera (2-1; 2-2; 2-3; 2-4) equipped with a fisheye lens. 7. Segment Station 7. The method according to claim 6, wherein (S-4) is carried out on an image (R-4) from a fisheye lens camera. 8. 8. The method according to claim 6 or 7, characterized in that the camera system (2-i) comprises a plurality of cameras (2-1; 2-2; 2-3; 2-4) arranged and configured to perform a 360° capture of the surroundings of the mobile means (10). 9. The control unit (5) determines the position of the moving means (10) based on the output segments (30; 32; 35). Vacant 9. The method according to any one of claims 5 to 8, wherein support for movement to the target area (46) is provided by a visual display. 10. The control unit (5) determines the position of the moving means (10) based on the output segments (30; 32; 35). Vacant 10. The method according to any one of claims 5 to 9, characterized in that the support for movement to the target area (46) is performed acoustically or tactilely, or the movement of the means of movement is performed in part by control measures. 11. The control unit (5) determines the position of the moving means (10) based on the output segments (30; 32; 35). Vacant 9. The method according to any one of claims 5 to 8, characterized in that a fully automatic movement to the target area (46) is performed. 12. The control unit (5) outputs Vacant Based on the segment (35) corresponding to the marked target area, Vacant 12. The method according to any one of claims 5 to 11, characterized in that information identifying the location of the marked target area is transmitted to an infrastructure facility outside the mobile vehicle. 13. The transportation means is a vehicle, Vacant 13. The method of any one of claims 5 to 12, wherein the marked target area is a parking space or an area for inductive charging of electric vehicles. 14. Segments that include: Station and extracting individual images (R-4) from the camera system (2-i) of the mobile means (10). Vacant Apparatus and / or system for identifying a marked target area (46): - an input unit configured to receive at least one image (R-4); - a data processing unit (3) configured to implement any one of the methods of 5 to 13 above; and -Individual Vacant an output unit configured to output to the control unit (5) the first segment (35) corresponding to the marked target area (46) and at least one further class of second segments (30; 32) of the captured periphery of the mobile means (10). 15. A camera system (2-i), a control unit (5), and a moving means (10) including the above-mentioned 14 devices.

Claims

1. A method for training a machine learning system to segment and identify individual vacant marked target areas (46) from images of a camera system (2-i) of a mobile vehicle (10) capturing a periphery within a semantic segmentation of the images, comprising: A method for producing a machine learning system comprising: a learning input image (R-4) and corresponding learning target segmentation data (S-4); and a parameter of the machine learning system is adjusted using the learning data (R-4, S-4) so ​​that, when the machine learning system receives the learning input image (R-4), the machine learning system produces output data corresponding to the learning target segmentation data (S-4); and the learning target segmentation data (S-4) comprises: - the location, extent or size of the first segments (35) corresponding to the individual free marked target areas (46), and - Class information of a second segment (30; 32) corresponding to at least one object other than the target area (46) around the captured vehicle (10).

2. The method according to claim 1, characterized in that the learning target segmentation data (S-4) also indicates, when the target area corresponds to a parking space, a parking space for disabled people, a parking space for women only, a parking space for electric vehicles or a parking space for children as the type of the first segment (35).

3. 3. The method according to claim 1 or 2, characterized in that the class information of the second segment (30; 32) is a structure, a sign, a plant, a vehicle, a motorcycle, a towed vehicle, a container, a person, an animal, a road marking, a lane boundary, a parking lot boundary, a road arrow, a road traffic symbol or a stop line.

4. The method according to claim 1 or 2, characterized in that the camera system (2-i) includes a camera (2-1; 2-2; 2-3; 2-4) equipped with a fisheye lens, and the learning input image (R-4) is an image of the fisheye lens camera.

5. A method for segmenting and identifying individual vacant marked target areas (46) from images (R-4) of a camera system (2-i) of a mobile means (10), comprising the steps of: - capturing at least one image (R-4) of the surroundings of the mobile means (10) by means of the camera system (2-i); - segmenting each free marked target area (46) from at least one image (R-4) by a trained machine learning system according to claim 1; - outputting to the control unit (5) first segments (35) corresponding to the respective free marked target areas (46) and at least one further class of second segments (30; 32) around the captured mobile means (10).

6. Method according to claim 5, characterized in that said camera system (2-i) comprises at least one camera (2-1; 2-2; 2-3; 2-4) equipped with a fish-eye lens.

7. 7. The method according to claim 6, characterized in that the segmentation is performed on the image (R-4) of the fisheye lens camera.

8. 8. The method according to claim 6 or 7, characterized in that the camera system (2-i) comprises a plurality of cameras (2-1; 2-2; 2-3; 2-4) arranged and configured to perform a 360° capture of the surroundings of the mobile means (10).

9. 7. The method according to claim 5 or 6, characterized in that the control unit (5) implements the support of the movement of the moving means (10) towards the free target area (46) by visual indications.

10. 7. The method according to claim 5 or 6, characterized in that the control unit (5) implements the support for the movement of the means of movement (10) towards the open target area (46) acoustically or haptically.

11. A method as described in claim 5 or 6, characterized in that the control unit (5) performs the movement of the moving means (10) to the vacant target area (46) by steering support or braking support during the parking process.

12. 7. The method according to claim 5 or 6, characterized in that the control unit (5) performs the movement of the moving means (10) to the free target area (46) fully automatically.

13. 7. The method according to claim 5 or 6, characterized in that by the control unit (5), information for identifying the location of the free marked target areas is transmitted to a communication facility outside the mobile means.

14. 7. The method according to claim 5 or 6, characterized in that the mobile means is a vehicle and the vacant marked target area is a vacant parking space or an area for inductive charging of electric vehicles.

15. 1. An apparatus for identifying and segmenting individual vacant marked target areas (46) from images (R-4) of a camera system (2-i) of a mobile means (10), comprising: an input unit for receiving at least one image (R-4); - a data processing unit (3) for implementing the method according to claim 5 or 6; and an output unit for outputting to the control unit (5) the position, extent or size of the first segments (35) corresponding to the respective free marked target areas (46) and class information of the second segments (30; 32) corresponding to at least one object other than the target area (46) in the periphery of the captured mobile means (10).

16. A camera system (2-i), a control unit (5) and a means of movement (10) comprising the device according to claim 15.

Images