Method and system for calibrating a camera

Abstract

A method and system for calibrating a camera (24) disposed on an ownship (70) is disclosed. The method includes receiving image data (ID) of an image (40) captured by the camera (24), the image (40) showing at least one imaged solid object (46, 48) and at least one imaged solid object interface (50, 52, 54, 56) between the imaged solid object (46, 48) and the imaged sky (46) or between the imaged solid object (44, 48) and the imaged water surface (42) of a body of water in which the ownship (70) is navigating, extracting the imaged solid object interface (50, 52, 54, 56) from the image (40), determining an attitude of the ownship (70) at the time the image (40) was captured, and receiving reference data (RD), the reference data (RD) including a reference to the imaged solid object interface (50, 52, 54, 56) in the image (40). and representing at least one real solid object interface (64, 82, 84, 86) in the real world (80) corresponding to the imaged solid object (50, 52, 54, 56), wherein the real solid object interface (64, 82, 84, 86) is between the corresponding real solid object (66, 94) and the real sky (96) or between the corresponding real solid object (66, 94) and the corresponding real water surface (92) of the body of water, respectively, and comprising: extracting the real solid object interface (64, 82, 84, 86) from reference data (RD); determining a difference between the extracted imaged solid object interface (50, 52, 54, 56) and the extracted real solid object interface (64, 82, 84, 86) taking into account the determined attitude of the own ship (70); and determining a calibration value (CV) of the camera (24) according to the difference so that the difference is reduced.

Description

Technical Field

[0001] The present invention relates to the field of camera calibration. In particular, the present invention relates to a method and system for calibrating a camera disposed on a host ship, which can be used for target detection based on, for example, monocular vision.

Background Art

[0002] Camera calibration is a well-established process used to estimate the intrinsic and extrinsic parameters of a camera, and optionally lens distortion. The former parameters define the internal characteristics of the camera, such as the optical center, focal length, skew, and / or distortion, and the latter parameters define the position and orientation of the camera in the real world. For many applications, such as for navigation purposes, knowing these parameters is essential as they can contribute to defining the mapping between the image coordinate system of the camera, which can be used for navigation, and the real world. For example, when a camera is disposed on a host ship navigating over water, this mapping may be used to estimate the distance and bearing of another ship on the same body of water.

[0003] Manual camera calibration is a time-consuming and labor-intensive process. Furthermore, maintaining calibration between multiple cameras is a difficult task as even a slight change in the position of a camera requires repeating the calibration process. Although not as accurate as manual calibration, automatic calibration is a desirable alternative as it does not suffer from the same problems.

[0004] Automatic Identification System (AIS) is currently used by a significant number of ships. This system transmits other ship-related information, such as geographical position, surface route, surface speed, etc., in addition to a unique identifier.

[0005] The March 2013 IEEE Conference on Aeronautics and Space (pp. 1-8), "Harbour surveillance with cameras calibrated with AIS data" by Palmieri FA, Castaldo F., and Marino G., describes how to use corresponding geolocation information for camera calibration. However, current conventional techniques have two problems: they do not take into account the range of other vessels, and they do not take into account that the position of GPS receivers on other vessels is not provided or is inaccurate. Therefore, a more accurate method for calibrating cameras is needed.

[0006] Calibrated cameras are also required for other applications, such as using monocular vision to determine the position of an object on the water's surface of another vessel. A known procedure is to calculate a ray or line in the real world corresponding to a given point in the image captured by the camera. To calculate this line, camera calibration, including intrinsic parameters, extrinsic parameters, and lens distortion, must be known. Intrinsic parameters, such as focal length and principal point, as well as lens distortion characteristics, can be obtained before the camera is installed on the vessel using a known camera calibration procedure with a checkerboard pattern. More difficult is obtaining the camera's extrinsic parameters that describe the camera's position and orientation on the vessel. Orientation is difficult because an accuracy of approximately one-tenth of a degree is required. Position is also difficult due to the practical difficulty of accurate measurement on the vessel's structure. Manual calibration is time-consuming, its accuracy depends on the experience of the person calibrating the camera and can be fluid, and inaccurate calibration leads to a decrease in the overall system performance; therefore, a corresponding automated calibration is desirable.

[0007] The most modern method for obtaining at least a subset of external parameters, namely pitch and roll relative to a known camera height, is to automatically detect the ocean horizon, essentially the water-sky boundary on the image captured by the calibrated camera. However, this method has several drawbacks. Firstly, it cannot calibrate the camera's position, only its roll and pitch angles. Secondly, for this method to work, a clear view of the horizon (water-sky boundary) is required. If the horizon is, for example, the land-sky boundary, the camera cannot be calibrated because the altitude of the land is unknown. [Overview of the project]

[0008] The object of the present invention is to provide a method for calibrating cameras installed on a ship that is highly accurate and / or cost-effective, and can be performed in particular automatically.

[0009] Another object of the present invention is to provide a system for calibrating cameras positioned on a ship in a highly accurate and / or cost-effective manner, particularly one that operates automatically.

[0010] These objectives are achieved by the subject matter of the independent claims. Further exemplary embodiments are evident from the dependent claims and the following description.

[0011] The object of the present invention is achieved by a method for calibrating a camera positioned on a ship. The method comprises: receiving image data of an image captured by a camera, wherein the image shows at least one imaged solid object and at least one imaged solid object interface between the imaged solid object and the imaged sky, or between the imaged solid object and the imaged water surface of the water body in which the ship is navigating; extracting the imaged solid object interface from the image; determining the attitude of the ship at the time the image was captured; receiving reference data, wherein the reference data represents at least one real solid object interface in the real world corresponding to the imaged solid object interface, where the actual solid object interfaces are, respectively, between the corresponding real solid object and the actual sky, or between the corresponding real solid object and the actual water surface of the corresponding water body; extracting the actual solid object interface from the reference data; determining the difference between the extracted imaged solid object interface and the extracted real solid object interface, taking into account the determined attitude of the ship; and determining a calibration value for the camera in accordance with the difference such that the difference is reduced (e.g., minimized).

[0012] Comparing a real solid object, particularly the real interface between a real solid object and a real water surface and / or between a real solid object and the real sky, to an imaged representation of the solid object—i.e., an imaged solid object interface—for camera calibration allows for automatic, cost-effective, and accurate camera calibration, even when the horizon, i.e., the boundary between water and sky, is not visible. “Imaged” in this specification may mean that the corresponding object is shown in the image, while “real” in this specification may mean that the corresponding object is in the real world.

[0013] Image data may be received by an entity performing the above method. The entity may be, for example, a general-purpose computer on board the vessel, or a system for calibrating a camera positioned on board the vessel. The body of water may be a lake, ocean, sea, or river. The calibration values ​​may be external calibration values, internal calibration values, and / or distortion parameters, such as camera lens distortion. The calibration values ​​may include the position of the camera on board the vessel and / or the orientation of the camera relative to the vessel's ship coordinate system. The camera position may correspond to the position of the camera's center. The camera center may be given by the camera's pinhole. The calibration values ​​may be represented by a vector θ. The camera orientation may include yaw, pitch, and roll. The attitude of the vessel may be determined by the vessel's self-positioning system. Determining the attitude of the vessel may include receiving attitude data representing the attitude of the vessel from the self-positioning system. The attitude may include the vessel's position, yaw, pitch, roll, and / or draft. The attitude may be determined by GPS, compass, and / or IMU data. The captured solid object interfaces may be extracted by image processing techniques and / or neural networks, for example, as described in the unpublished patent application EP22180131.9. The actual solid object interfaces may be extracted by image projection techniques and / or by coordinate transformation of the corresponding reference data.

[0014] This difference may be any mathematical concept suitable for describing or defining the deviation between the captured solid object interface and the actual solid object interface. In particular, the difference between the captured solid object interface seen in the image and the projection of the actual solid object interface on the image can be calculated. For example, the difference may be given by one or more distances between one or more points along the captured solid object interface and corresponding points along the actual solid object interface projected onto the image.

[0015] In general, the interface between a solid object and the water surface, or between a solid object and the sky, can be either a physical or non-physical interface. Physical interfaces exist in the real world, while non-physical interfaces exist only in images and do not exist in the real world.

[0016] The solid object may be land or another vessel. If the solid object is land, the solid object interface may be the physical land-water interface between the water surface and the land, for example, the coastline of the land, or the physical land-air interface between the sky and the land, i.e., the skyline of the land. Alternatively, if the solid object is land, the solid object interface may be the non-physical land-water interface between the water surface and the land, for example, the interface between the water surface and the elevation of the land above the coastline. If the solid object is another vessel, the solid object interface may be the physical ship-water interface between the water surface and the other vessel, for example, the waterline of the other vessel, or the physical ship-air interface between the sky and the other vessel, i.e., the skyline of the other vessel. Alternatively, if the other vessel is the solid object, the solid object interface may be the non-physical ship-water interface between the water surface and the part of the other vessel above the waterline of the other vessel. This may be valid for imaged solid object interfaces as well as for real solid object interfaces.

[0017] Real-world solid object interfaces are also 2D lines, just like the imaged solid object interfaces. In particular, the interface between an imaged solid object and an imaged water surface, or between an imaged solid object and an imaged sky, is always a 2D line in the camera's 2D image, while the interface between a real solid object and a real water surface is always a 3D line, and the interface between a real solid object and the real sky is always a 3D surface in the real world. However, for the purposes of this invention and in this description, only the portions of 3D lines or 3D surfaces in the real world that correspond to the 2D imaged solid object interfaces are used. Therefore, with respect to this invention and in this description, real-world solid object interfaces are also 2D lines, such as coastlines, waterlines, or skylines, just like the corresponding imaged solid object interfaces.

[0018] The above method may be performed several times. For example, the above method may be executed as a loop. The calibration loop may include an optimization algorithm, such as a nonlinear optimization algorithm. The above method, i.e., the loop, may be executed until a predetermined condition is met, for example, until the difference falls below a predetermined threshold of difference. When the method is executed for the first time, a predetermined calibration value may be received from the system's memory for calibrating the camera and adjusted to determine the camera's calibration value. In this case, the predetermined calibration value may be stored in memory beforehand, for example, by the system manufacturer, as a fixed starting value. Alternatively, a "best guess" calibration value, or a calibration value determined from the last calibration, may be used as the starting point when the above method is executed for the first time. When the method is executed a second time or later during a single calibration, the predetermined calibration value may be an adjusted calibration value from the previous loop. Thus, the above method can provide a calibration loop for adjusting the calibration value.

[0019] According to one embodiment, extracting a captured solid object interface from an image comprises extracting the path of the captured solid object interface in the image, and extracting a real solid object interface from reference data comprises extracting the path of the real solid object interface in the real world, and the difference between the captured solid object interface and the real solid object interface corresponds to the difference between the extracted path of the captured solid object interface and the extracted path of the real solid object interface. The extracted path of the captured solid object interface may include the morphology and position of the captured solid object interface in the image, in particular the coordinates of the corresponding pixels in the image. The extracted path of the real solid object interface may include the morphology and position of the real solid object interface in the real world, in particular the coordinates of the corresponding pixels in the real world.

[0020] According to one embodiment, determining the difference between the imaged solid object interface path and the real solid object interface path comprises transforming the imaged solid object interface path and / or the real solid object interface path into the same coordinate system and determining the misfit function between the imaged solid object interface and the real solid object interface in the coordinate system as the difference, and determining the camera calibration value such that the difference is reduced comprises determining the calibration value such that the misfit function is reduced, for example, minimized. Alternatively, a cost function may be used instead of the misfit function. The “same” coordinate system may be the real-world coordinate system given by the reference data, the image coordinate system given by the camera and the image, or another coordinate system used only to determine the difference.

[0021] According to one embodiment, a predetermined calibration value is changed so as to minimize the misfit function. When using a cost function, the difference can be reduced by reducing the cost of the cost function, for example, by minimizing it.

[0022] According to one embodiment, when extracting the interface of a real solid object from reference data, only the portion of the interface of the real solid object visible to the camera is extracted, and determining the difference between the imaged interface and the interface of the real solid object comprises determining the difference between the imaged interface and only the determined visible portion of the interface of the real solid object. A real solid object may have an interface with respect to a water surface or the sky that is not visible to the camera. Therefore, the image may not include the imaged interface of the real solid object that corresponds to that interface of the real solid object. Determining these portions of the interface of the real solid object visible to the camera, and determining the difference between the imaged interface and the determined visible portion of the interface of the real solid object, may only contribute to solving this problem.

[0023] According to one embodiment, the method comprises determining the positional relationship between the camera and the interface of the real solid object from reference data, and determining the portion of the interface of the real solid object visible from the camera according to the determined positional relationship. The positional relationship may be determined by converting the positions of the camera and the interface of the real solid object into the same coordinate system, for example, a 2D coordinate system. For example, the reference data can comprise 2D data, such as 2D map data, or 3D data, such as 3D map data, which are converted into 2D data, such as 2D map data, and the position of the camera can be converted into the corresponding 2D map. Then, the positional relationship can be extractable and / or visible from a 2D map comprising the representation of the interface of the real solid object and the position of the camera. The positional relationship can be defined by one or more distances between the position of the camera and one or more points along the representation of the interface of the real solid object in the corresponding 2D map.

[0024] According to one embodiment, the reference data comprises 2D map data representing a 2D map around the vessel itself. The 2D map data can comprise the 2D coordinates of one or more land polygons corresponding to one or more shorelines in the water area. The 2D map data may be obtained from a digital electronic map, such as OpenStreetMap.

[0025] According to one embodiment, the portion of the interface of the real solid object visible from the camera is determined from the 2D map data according to the determined positional relationship between the camera and the interface of the real solid object by ray casting. The ray casting can be 2D uniform ray casting. The ray casting can contribute to determining the portion of the interface of the real solid object that is easily and / or accurately visible from the camera.

[0026] According to one embodiment, the reference data includes AIS data. The AIS data may include vessel-related information about one or more other vessels close to the own vessel. In particular, the AIS data may be related to at least one other vessel in the water area, where the other vessel may be used as a real solid object. The vessel-related information may include a unique identifier, geographical location, heading, and ground speed of the other vessel. Further, the AIS data may include the range of other vessels related to the geographical location of its GPS receiver. Thereby, when using the corresponding real solid object interface, i.e., the vessel-water interface between the other vessel and the water surface, to determine the calibration value, it is possible to take into account the range of other vessels in relation to the geolocation of its GPS receiver.

[0027] According to one embodiment, the AIS data includes the geolocation of the GPS receiver of the other vessel and the range of the other vessel related to the geolocation of the GPS receiver of the other vessel. From the range and geolocation of the GPS receiver of the other vessel, the real vessel-water interface between the real water surface and the other vessel is estimated. Depending on the estimated real vessel-water interface and the position of the camera, the part of the real vessel-water interface visible from the camera is determined. The visible part of the real vessel-water interface is used as the real solid object interface for comparison with the corresponding imaged solid object interface. This contributes to a very accurate determination of the calibration value. The real vessel-water interface may be determined by constructing a bounding box, for example a perfectly fitting bounding box, to which the other vessel conforms according to the range of the other vessel, and by constructing an ellipse, for example a perfectly fitting ellipse, that conforms to the bounding box, where the ellipse is transformed into the real-world coordinate system and used as the real vessel-water interface. Similarly, other geometric shapes that more accurately model the contour of the vessel may be used in addition to the ellipse.

[0028] According to one embodiment, the reference data comprises elevation map data of real land surrounding a body of water. This makes it possible to determine real solid object interfaces, particularly real land-air interfaces. The elevation map data can represent a 3D map and / or topography of the land and is sometimes called a digital elevation model (DEM), as is known in the art.

[0029] According to one embodiment, the portion of the real solid object interface visible from the camera is determined from elevation map data by generating a textureless rendering image and by extracting portions from the textureless rendering image. The textureless rendering image may be generated by a renderer as a composite view of land surrounding a water body and / or taking into account predetermined calibration values ​​and the attitude of the vessel. The textureless rendering image may comprise the sky, land, and water surface in different colors. The visible portion may then be extracted by image processing as is known in the art.

[0030] According to one embodiment, the calibration value is adjusted until a predetermined condition is met. In other words, the steps of the method may be repeated until a predetermined condition is met. For each loop of the method, a new image may be captured by the camera and received by a system for calibrating the camera, a new imaged solid object interface may be extracted from the image, and a corresponding new real solid object interface may be extracted from reference data. The predetermined condition may be that the difference is less than a predetermined threshold of difference.

[0031] According to one embodiment, the calibration value is determined by adjusting a predetermined calibration value of the camera so that the difference is reduced, for example, minimized. The predetermined calibration value may be stored in the memory of the system for calibrating the camera. The predetermined calibration value may be known by the system manufacturer or may be derived from the camera specifications of the camera and stored in memory. This may be advantageous when the method is being performed for the first time. Alternatively or additionally, the predetermined calibration value may be the final calibration value determined by the camera's last calibration. This may be advantageous when the method has already been performed one or more times.

[0032] The objective is achieved by a system for calibrating cameras positioned on the vessel. The system comprises a memory configured to store reference data and / or calibration values, such as predetermined calibration values, and a processing unit configured to perform the above method. Furthermore, the system may include a camera, or be part of a camera. Furthermore, the system may include means for receiving reference data, for example, via radio and / or satellite communication.

[0033] It should be understood that the characteristics of the methods for calibrating cameras placed on board a ship, as described above and below, may also be characteristics of the systems for calibrating cameras placed on board a ship, as described above and below.

[0034] These and other aspects of the present invention will become apparent from the examples described below and will be clarified by reference to those examples. [Brief explanation of the drawing]

[0035] The subject matter of the present invention will be described in detail below with reference to exemplary embodiments shown in the accompanying figures. [Figure 1] Figure 1 shows a block diagram illustrating an exemplary embodiment of a system for calibrating cameras installed on a ship. [Figure 2]Figure 2 shows a flowchart of an exemplary embodiment of a method for calibrating a camera. [Figure 3] Figure 3 shows an example image of another vessel in a body of water. [Figure 4] Figure 4 shows an enormous picture of the real world. [Figure 5] Figure 5 shows an example of a visualization of the real ship-water interface estimated from AIS data. [Figure 6] Figure 6 illustrates the principle for determining the estimated portion of the ship-water interface visible from the camera.

[0036] The reference numerals used in the drawings and their meanings are listed in a summary format in the reference numeral list. As a general rule, the same reference numeral is assigned to identical parts within a drawing. [Modes for carrying out the invention]

[0037] Figure 1 shows a block diagram illustrating an exemplary embodiment of a system for calibrating a camera 24 located on a vessel 70 (see Figure 6). The system may comprise a positioning device 22 for determining the position data PD of the vessel 70, a camera 24 for generating an image data ID of an image 40 (see Figure 3), an acquisition component 26 for acquiring reference data RD, an extraction component 28 for extracting calibration features CF from the image data ID, and a calibration loop 30 for determining a calibration value CV for calibrating the camera 24 according to the position data PD, calibration features CF, and reference data RD. In addition, the system may comprise a memory (not shown) configured to store the reference data RD and / or calibration value CV, and a processing unit (not shown) configured to perform a method for calibrating the camera 24 located on the vessel 70, as described, for example, with respect to Figure 2.

[0038] Camera 24 is configured to capture one or more images 40 around the vessel 70 and generate corresponding image data IDs. Positioning device 22 is configured to determine the attitude of the vessel 70 at the time the images 40 are captured and generate corresponding attitude data PD. Extraction component 28 is configured to extract calibration features CF from the image data IDs, i.e., at least one imaged solid object interface from the images 40. Acquisition component 26 may be configured to acquire reference data RD from the system's memory or from an external source, for example, via satellite, radio, and / or the internet. Acquisition component 26 may include another extraction component for extracting at least one real solid object interface from the reference data RD.

[0039] Reference data RD represents at least one real solid object interface in the real world 80 (see Figure 4) corresponding to the imaged solid object interface in image 40, where the real solid object interface lies between the corresponding real solid object and the real sky 94, or between the corresponding real solid object and the corresponding real water surface 92 of a body of water. The processing unit is configured to receive the reference data RD, determine the difference between the imaged solid object interface and the real solid object interface, taking into account the determined attitude of the vessel 70, and determine the calibration CF value of the camera 24 in accordance with the difference so that the difference is reduced, for example, minimized.

[0040] The functions and / or operation of the system will be described in more detail below with respect to how they are performed by the system, and the steps of the method shown in Figure 2 will be illustrated with reference to Figures 3 through 6.

[0041] Figure 2 shows a flowchart of an exemplary embodiment of a method for calibrating a camera 24 located on the vessel 70. The vessel 70, on which the camera 24 is located, can navigate a body of water. The body of water may be a lake, ocean, sea, or river.

[0042] In step S2, the image data ID of the image 40 (see Figure 3) captured by the camera 24 is received. The camera 24 may generate an image data ID representing the image 40. The image data ID may be sent to and / or received by a processing unit that performs a method for calibrating the camera 24. The image data ID may be stored in the memory of the system for calibrating the camera 24, where the memory is connected to the processing unit. When the method is performed, the processing unit may receive the image data ID directly from the camera 24 or from the memory. The camera 24 can be calibrated with respect to its position and / or orientation relative to the ship 70 by the system and method described above.

[0043] Figure 3 shows an example of an image 40 showing another captured vessel 48 that can be seen from the vessel 70 and is navigating in the water. The water may be represented in the image 40 by a captured water surface 42. The image 40 may show a captured water surface 42, captured land 44, for example, two captured land areas, captured sky 46, a captured other vessel 48, and a captured boundary box 58 surrounding the captured other vessel 48.

[0044] Image 40 may show at least one imaged solid object and at least one imaged solid object interface between the imaged solid object and the imaged sky 46, or between the imaged solid object and the imaged water surface 42 of the water in which the vessel 70 is navigating. For example, the imaged solid object may be an imaged land 44 or an imaged other vessel 48. In the case of an imaged land 44 as an imaged solid object, the solid object interface may be an imaged land-water interface 50 between the imaged water surface 42 and the imaged land 44, i.e., the imaged coastline of the imaged land 44, or an imaged land-air interface 52 between the imaged sky 46 and the imaged land 44, i.e., the imaged skyline of the imaged land 44. In the case of another vessel 48 that is imaged as a solid object, the imaged solid object interface may be the imaged vessel-water interface 54 between the imaged water surface 42 and the imaged other vessel 48, i.e., the imaged waterline of the imaged other vessel 48, or the imaged vessel-sky interface 56 between the imaged sky 46 and the imaged other vessel 48, i.e., the imaged skyline of the imaged other vessel 48.

[0045] In step S4, the captured solid object interfaces, i.e., the captured land-water interface 50, the captured land-air interface 52, the captured ship-water interface 54, or the captured ship-air interface 56, are extracted from the image 40, particularly from the image data ID, for example, by the extraction component 28. Extracting the captured solid object interfaces from the image 40 may also involve extracting the paths of the captured solid object interfaces within the image 40. The paths of the captured solid object interfaces may include the morphology and position of the captured solid object interfaces within the image 40, where the image data ID may comprise the pixel coordinates of the corresponding pixels in the image 40. To extract the captured solid object interfaces, particularly the corresponding pixel coordinates, a captured bounding box 58 may be determined, and the paths of the captured solid object interfaces may be explored only within the bounding box 58.

[0046] The extraction component 28 may comprise a first neural network configured to receive an image data ID and extract an imaged solid object interface from the image data ID. For example, if an imaged ship-water interface 54 is to be extracted and used as an imaged solid object interface, the imaged solid object interface may be extracted from the image data ID by determining at least one ship-water value for each pixel of the image 40 from the image data ID, where the ship-water value may represent the probability of a corresponding pixel indicating an imaged solid object, in this case another imaged ship 48, an imaged water surface 42, or an imaged sky 46. The imaged ship-water interface 54 may then be determined from the ship-water values ​​by one or more post-processing steps, for example, by edge detection and / or thresholding of the ship-water values.

[0047] The ship-water value may be determined by semantic segmentation of image 40. Semantic segmentation may be performed by a first neural network trained, for example, by supervised learning using correspondingly labeled images to distinguish solid objects, water surfaces, and the sky in an image. For example, the first neural network may be pre-trained by supervised learning using a certain amount of images 40 showing captured solid objects, e.g., other ships 48 in the water, and by corresponding captured images of the water surface 42, captured sky 46, and images 40 labeled with respect to the other ships 48. Furthermore, images 40 that do not show any solid objects may also be involved in training the first neural network.

[0048] A method for extracting solid object interfaces, particularly ship-water interfaces, from images that may be used in this situation is known from an unpublished patent application EP22180131.9, which describes it in detail. Although this patent application focuses solely on the extraction of ship-water interfaces, the corresponding method can be readily extended to the extraction of solid object interfaces in general by those skilled in the art by training a first neural network to be able to determine, for example, other solid object interfaces, namely, an imaged land-water interface 50, an imaged land-air interface 52, or an imaged ship-air interface 56.

[0049] In step S6, the attitude of the vessel 70 at the time the image 40 was captured is determined. Determining the attitude of the vessel 70 may include receiving attitude data PD representing the attitude of the vessel 70 from the positioning device 22. The attitude may include the position, yaw, pitch, roll, and / or draft of the vessel 70.

[0050] In step S8, reference data RD is received. Reference data RD represents at least one real solid object interface in the real world 80 that corresponds to the captured solid object interface in image 40. Reference data RD may include a path of a real solid object interface in the real world 80. The path of a real solid object interface may include the morphology and location of the real solid object interface in the real world 80. For example, the real-world coordinates of the path of a real solid object interface may be encoded within the reference data RD.

[0051] In step S9, the actual solid object interfaces, particularly the paths of the actual solid object interfaces, are extracted from the reference data RD. The extracted actual solid object interfaces, particularly the extracted paths of the actual solid object interfaces, essentially correspond to the calibration features CF extracted from the image data, i.e., the corresponding imaged solid object interfaces, except for the differences described below. The method for extracting the actual solid object interfaces may depend on the type of reference data RD that can be used. For example, if the reference data RD comprises a 2D map of the vessel 70, the actual solid object interfaces may correspond to the coastline represented in the 2D map, and the actual solid object interfaces can be extracted from the corresponding 2D map by simply extracting the corresponding coordinates from the 2D map. Alternatively, if the reference data RD comprises AIS data, such as described with respect to Figure 5, the actual solid object interfaces may correspond to the waterline of the corresponding other vessel 48, and the waterline is determined from the AIS data as described below. Alternatively, if the reference data RD includes elevation map data, for example, as described with respect to Figure 4, then the actual solid object interface may correspond to the corresponding land skyline, or the physical or non-physical water-land interface, as described below.

[0052] Figure 4 shows an exemplary picture of the real world 80, given, for example, by an altitude map, or, for example, by reference data RD. The portion of the real world 80 shown in the picture of Figure 4 does not correspond to the portion of the real world 80 shown in Image 40, and is shown to illustrate at least some of the real solid object interfaces available for use in the present invention. Generally, a real solid object interface lies between a corresponding real solid object and the real sky 96, or between a corresponding real solid object and the corresponding real water surface 92 of a body of water. In this example, the real solid object could be real land 94, or, for example, another real ship (not shown in Figure 4) corresponding to another imaged ship 48 (shown in Figure 3). In the case of a real solid object, such as a real landmass 94, the real solid object interface may be a first real land-water interface 82 (continuous line) between the real water surface 92 and the real landmass 94, for example, the real coastline of the real landmass 94, also called the physical land-water interface, or a second real land-water interface 84 (dashed line) between the real water surface 92 and the real landmass 94, for example, the line between the altitude of the real landmass 94 and the real water surface 92, which is also called a non-physical land-water interface because it is not the real world but merely the land-water interface as seen from the ship 70. Alternatively, the real solid object interface may be a real land-air interface 86 (dotted line) between the real sky 96 and the real landmass 94, i.e., the real skyline of the actual landmass 94. If the real object is another real ship, the interface of the real object may be the real ship-water interface (not shown) between the real water surface 92 and the other real ship, for example, the real waterline of the other real ship, or the real ship-sky interface (not shown) between the real sky 96 and the other real ship, i.e., the real skyline of the other real ship.

[0053] In the optional step S10, the portion of the real solid object interface visible from the camera 24 may be determined. A real solid object, such as real land 94, may have an interface with the real water surface 92 or the real sky 96 that is not visible from the camera 24. Therefore, the image 40 may not include and / or show the captured solid object interface corresponding to that real solid object interface. For example, the first real land-water interface 82 includes a portion of the coastline of the real land 94 visible from the camera 24, but does not include a portion of the coastline of the real land 94 that is covered by the altitude of the real land 94. Therefore, the first real land-water interface 82 is obscured by the second real land-water interface 84. The second real land-water interface 84 does not include the coastline of the real land 94, but only includes the 2D line between the real water surface 92 and the altitude of the real land 94 as seen from the ship 70. Furthermore, only a portion of the real skyline of the real landmass 94, for example, the real land-air interface 86 as shown in Figure 4, may be visible from the camera 24. The portion of the real solid object interface visible to the ship 70, particularly from the camera 24, may be determined by raycasting, as described below with respect to Figure 6.

[0054] In particular, in an optional step S12 which may be performed after an optional step S10 has been performed, the positional relationship between the camera 24 and the interface of a real solid object may be determined from reference data RD. In this case, the portion of the interface of the real solid object visible to the camera 24 may be determined according to the determined positional relationship. The positional relationship may be determined by transforming the position of the camera 24 and the interface of the real solid object into the same coordinate system, for example, a 2D coordinate system, by any suitable coordinate transformation method known in the art. For example, the reference data RD may consist of 2D data, for example, 2D map data, or 3D data, for example, 3D map data, which can be transformed into 2D data, for example, 2D map data. The position of the camera 24 can then be mapped to the corresponding 2D map. Subsequently, the positional relationship may be extractable and / or visible from a 2D map comprising a representation of the interface of the real solid object and the position of the camera 24. The positional relationship may be defined by one or more distances between the position of the camera 24 and one or more points along the representation of the interface of the real solid object in the corresponding 2D map. Therefore, the reference data RD may include 2D map data representing a 2D map of the area around the vessel 70, particularly of camera 24. The 2D map data may include 2D coordinates of one or more land polygons corresponding to one or more coastlines in the water. The 2D map data may be obtained from OpenStreetMap.

[0055] 2D map data can also be used to extract real solid object interfaces, particularly visible real solid object interfaces, from the corresponding 2D map. The extracted real solid object interfaces may be projected onto image 40.

[0056] Alternatively, the reference data RD may include AIS data. Generally, the AIS data may include ship-related information about one or more other vessels close to the vessel 70. In particular, the AIS data may include the geographical location of the GPS receiver 62 of a real other vessel 66 (see Figure 5) corresponding to the imaged other vessel 48. The ship-related information may include a unique identifier, a land route, the land speed of the real other vessel 66, and the geographical location of the GPS receiver 62. Furthermore, the AIS data may include the range of the real other vessel 66 in relation to the geographical location of its GPS receiver 62.

[0057] Figure 5 shows an example of visualization of the real ship-to-water interface estimated from AIS data, i.e., the estimated real ship-to-water interface 64. The extent of the corresponding real ship 66 may be given by the width to starboard WS, the width to port WP, the length to bow LB, and / or the length to stern LS, each related to the position of the GPS receiver 62. The estimated real ship-to-water interface 64 may be determined, depending on the extent of the real ship 66, by constructing, for example, a perfectly fitting AIS bounding box 60 to which the real ship 66 fits, and by constructing, for example, a perfectly fitting ellipse to which the bounding box fits, where the ellipse may correspond to the estimated real ship-to-water interface 64, may be transformed into a real-world coordinate system, and may be used as the real ship-to-water interface. Alternatively, the real ship-to-water interface may be estimated using a form other than an ellipse. Furthermore, the shape extracted from the AIS data may be supplemented with corresponding information from radar and / or lidar measurements.

[0058] Therefore, the actual ship-water interface between the actual water surface 92 and the actual other ship 66 can be estimated from the extent and geolocation of the actual other ship 66 corresponding to the imaged other ship 48. The portion of the actual ship-water interface visible from camera 24 can be determined depending on the estimated actual ship-water interface 64 and the position of camera 24. The actual ship-water interface 64, in particular the visible portion of the actual ship-water interface 64, may be projected onto image 40 and used to determine the difference between the imaged solid object interface and the actual solid object interface.

[0059] Alternatively, the reference data RD may include elevation map data of the real land 94 surrounding the body of water. The elevation map data can represent a 3D map and / or topography of the real land 94 and may be called a digital elevation model (DEM) as is known in the art. In this case, the portion of the real solid object interface visible to the camera 24 may be determined from the elevation map data by generating a textureless rendering image and extracting the portion of the real solid object interface visible to the camera 24 from the textureless rendering image. The textureless rendering image may be generated, for example, as a composite view of the real land 94 surrounding the body of water and / or by a renderer, taking into account predetermined calibration values ​​of the camera 24 and the attitude of the vessel 70. The textureless rendering image may include representations of the real sky 96, real land 94, and real water surface 92 having different properties, for example, different colors. The visible portion of the real solid object interface visible to the camera 24 may then be extracted by image processing as is known in the art. The extracted real solid object interfaces, particularly the visible portions of the real solid object interfaces, may be projected onto image 40 and used to determine the difference between the imaged three-dimensional object interface and the real solid object interface.

[0060] In step S12, the difference between the imaged solid object interface and the actual solid object interface is determined considering the determined orientation of the ship 70. If the optional step S10 is performed, only the difference between the imaged solid object interface and the determined visible portion of the actual solid object interface may be determined. This difference may be any mathematical concept suitable for describing or defining the deviation of the imaged solid object interface from the corresponding actual solid object interface. In particular, the difference between the imaged solid object interface seen in image 40 and the projection of the actual solid object interface on image 40 may be calculated. For example, the difference may be given by one or more distances between one or more points along the imaged solid object interface, e.g., along the path of the imaged solid object interface, and corresponding points along the actual solid object interface projected onto image 40, e.g., along the path of the actual solid object interface projected onto image 40. In general, the difference between the imaged solid object interface and the actual solid object interface may correspond to the difference between the path of the imaged solid object interface and the path of the actual solid object interface transformed into the same coordinate system, e.g., the image coordinate system. Determining the difference between the imaged solid object interface path and the actual solid object interface path may involve transforming the imaged solid object interface path and / or the actual solid object interface path into the same coordinate system, and determining the misfit function between the imaged solid object interface and the actual solid object interface within the coordinate system as the difference.

[0061] For example, the calibration features CF extracted from image 40, i.e., the coordinates of one or more of the solid object interfaces described above, can be given by the following:

number

[0062] Another set of points, expressed in world coordinates, can be obtained from the discretization of the visible portion of the interface of real solid objects given by the reference data RD, as follows:

number

[0063] Next, the distance from the nearest calibration feature CF, i.e., the nearest imaged solid object interface, to the corresponding extracted reference data RD, i.e., the corresponding real solid object interface, can be given by the following:

number

number

[0064] The vector θ is defined by the following equation:

number

[0065] In this context, for example, a method for projecting real-world 3D points onto a 2D image 40 using calibration values ​​from the aforementioned reference, namely "Multiple View Geometry in Computer Vision," is already known in the art. It should be noted that the projection of real-world points onto an image from an ideal camera is described on page 154 of Chapter 6, and a method for correcting lens distortion is described on page 191 of Chapter 7.

[0066] In step S14, the calibration value CV of camera 24 may be determined such that the difference is reduced, for example, minimized, depending on the difference described above. For example, the misfit and / or cost function may be defined as follows:

number

number

[0067] From this optimization, a set of improved calibration values ​​θ* may be determined, where θ* is an improved calibration value CV that can be further improved in subsequent iterations.

[0068] Alternatively, the Fréchet distance can be used to determine the difference between the imaged solid object interface and the actual solid object interface.

[0069] The calibration value CV may be determined by adjusting a predetermined calibration value of the camera 24 so that the difference is reduced, for example, minimized. The predetermined calibration value may be stored in the system's memory for calibrating the camera 24. The predetermined calibration value may be stored in memory by the system manufacturer. Alternatively or additionally, the predetermined calibration value may be the final calibration value CV determined by the last calibration of the camera 24, or it may be a "best guess" calibration value.

[0070] The calibration value CV may be adjusted until a predetermined condition is met. In other words, the steps of the above method may be repeated until a predetermined condition is met. For each loop of the method, a new image 40 may be captured by the camera 24 and received by a system for calibrating the camera 24, a newly captured solid object interface may be extracted from the image 40, and a corresponding new real solid object interface may be extracted from the reference data RD. The predetermined condition may be that the difference is less than a predetermined threshold of difference.

[0071] Figure 6 illustrates the principle for determining the estimated portion of the ship-water interface visible from camera 24, i.e., the visible ship-water interface 74, which is shown by a solid line in Figure 6. Camera 24 is positioned on the ship 70. Camera 24 has a field of view 72 in which the estimated real ship-water interface 64 is located, where the field of view 72 may be larger than depicted in Figure 6, and in particular larger than the estimated real ship-water interface 64. As can be seen from Figure 6, the visible ship-water interface 74 is visible from camera 24, while the rest of the estimated real ship-water interface 64 (dashed line) is not visible from camera 24. Therefore, the visible ship-water interface 74 may be used only to determine the calibration value CV. A numerical method can be used to distinguish between the visible ship-water interface 74 and the rest of the ship-water interface. For example, a point on the entire ship-water interface and a line segment connecting that point to the camera center can be defined. Next, points on the ship-water interface that have relevant line segments that do not intersect with other parts of the ship-water interface belong to the visible ship-water interface 74. The visible ship-water interface 74 can be determined by raycasting.

[0072] The aforementioned neural networks and components, such as the first neural network, acquisition component 26, extraction component 28, and / or calibration loop 30, can each be implemented by software, hardware, or a combination of software and hardware.

[0073] Although the present invention has been illustrated and described in detail in the drawings and the foregoing description, such illustrations and descriptions should be considered illustrative or exemplary and not limiting, and the present invention is not limited to the disclosed embodiments. Other modifications to the disclosed embodiments can be understood and practiced by those skilled in the art practicing the claimed invention, by reference to the drawings, disclosure and the appended claims. In the claims, the word “equipped with” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude plurals. A single processing unit or other units may perform the functions of several items described in the claims. The mere fact that certain means are described in different dependent claims does not mean that combinations of these means cannot be used advantageously. None of the reference numerals in the claims should be construed as limiting the scope.

[0074] List of reference symbols 20 Calibration System 22 Positioning devices 24 cameras 26 Acquired Components 28 Extraction Components 30 Calibration Loops 40 images 42 Imaged water surface 44. Land areas that were photographed 46 The captured sky 48 Other ships that were photographed 50 Imaged land-water interface 52 Imaged land-air interface 54 Imaged ship-water interface 56 Imaged ship-air interface 58 Imaged boundary box 60 AIS Boundary Box 62 GPS receivers 64 Estimated Real-World Ship-Water Interface 66 Other real ships 70. Own ship 72 Field of view 74 Visible ship-water interface 80 Real World (Picture) 82 The first real land-water interface 84 The second reality: the land-water interface 86 Real-world land-air interface 88 Horizontal line 90 High-level contour 92 The actual water surface 94 Real land 96 The Real Sky S2-S14 Steps 1-14 PD location data ID image data RD Reference Data CF Calibration Features CV calibration value WS (Wide to starboard) Width to WP port LB (Length to Bow) LS (Length to the stern) The invention described in the original claims of this application is listed below. [1] A method for calibrating a camera (24) positioned on the ship (70), The camera (24) receives image data (ID) of the image (40) captured by the camera (24), and the image (40) shows at least one imaged solid object (44, 48) and at least one imaged solid object interface (50, 52, 54, 56) between the imaged solid object (44, 48) and the imaged sky (46), or between the imaged solid object (44, 48) and the imaged water surface (42) of the water area in which the ship (70) is navigating. Extracting the captured solid object interfaces (50, 52, 54, 56) from the image data (ID), To determine the attitude of the ship (70) at the time the aforementioned image (40) was captured, Receiving reference data (RD), the reference data (RD) represents at least one real solid object interface (64, 82, 84, 86) in the real world (80) corresponding to the captured solid object interface (50, 52, 54, 56) in the image (40), where the real solid object interface (64, 82, 84, 86) is between the corresponding real solid object (66, 94) and the real sky (96), or between the corresponding real solid object (66, 94) and the corresponding real water surface (92) of the body of water. Extracting the actual solid object interfaces (64, 82, 84, 86) from the aforementioned reference data (RD), The difference between the extracted imaged solid object interfaces (50, 52, 54, 56) and the extracted real solid object interfaces (64, 82, 84, 86) is determined considering the determined attitude of the vessel (70), A method comprising determining the calibration value (CV) of the camera (24) according to the difference such that the difference decreases. [2] Extracting the captured solid object interfaces (50, 52, 54, 56) from the image (40) comprises extracting the paths of the captured solid object interfaces (50, 52, 54, 56) in the image (40), Extracting the actual solid object interfaces (64, 82, 84, 86) from the reference data (RD) comprises extracting the paths of the actual solid object interfaces (64, 82, 84, 86) from the reference data (RD), The difference between the imaged solid object interface (50, 52, 54, 56) and the actual solid object interface (64, 82, 84, 86) corresponds to the difference between the path of the imaged solid object interface (50, 52, 54, 56) and the path of the actual solid object interface (64, 82, 84, 86) according to [1]. [3] Determining the difference between the paths of the imaged solid object interfaces (50, 52, 54, 56) and the paths of the actual solid object interfaces (64, 82, 84, 86) comprises moving the paths of the imaged solid object interfaces (50, 52, 54, 56) and / or the paths of the actual solid object interfaces (64, 82, 84, 86) to the same coordinate system, and determining the misfit function between the imaged solid object interfaces (50, 52, 54, 56) and the actual solid object interfaces (64, 82, 84, 86) in the coordinate system as the difference in the coordinate system. The method according to [2], wherein determining the calibration value (CV) of the camera (24) such that the difference is reduced is the method according to [2], wherein determining the calibration value (CV) such that the misfit function is reduced. [4] The method according to [3], wherein a predetermined calibration value (CV) is varied such that the misfit function is minimized. [5] When extracting the actual solid object interfaces (64, 82, 84, 86) from the reference data (RD), only the portion of the actual solid object interfaces (64, 82, 84, 86) visible from the camera (24) is extracted. The method according to any one of [1] to [4], wherein determining the difference between the imaged solid object interface (50, 52, 54, 56) and the actual solid object interface (64, 82, 84, 86) is determined to determine the difference between the imaged solid object interface (50, 52, 54, 56) and only the extracted visible portion of the actual solid object interface (64, 82, 84, 86). [6] Determining the positional relationship between the camera (24) and the actual solid object interface (64, 82, 84, 86) from the reference data (RD), The method according to [5], further comprising determining the portion of the real solid object interface (64, 82, 84, 86) visible from the camera (24) according to the determined positional relationship. [7] The method according to any one of [1] to [6], wherein the reference data (RD) comprises 2D map data representing a 2D map of the area around the vessel (70). [8] The portion of the real solid object interface (64, 82, 84, 86) visible from the camera (24) is determined from the 2D map data by raycasting in accordance with the determined positional relationship between the camera (24) and the real solid object interface (64, 82, 84, 86), as described in [7], citing [6]. [9] The method according to any one of [1] to [6], wherein the reference data (RD) comprises AIS data relating to at least one other vessel (48) in the waters.

[10] The AIS data comprises the geolocation of the GPS receiver (62) of the other vessel (48) and the range (WS, WP, LB, LS) of the other vessel (48) with respect to the geolocation of the GPS receiver (62) of the other vessel (48), The actual ship-water interface (64) between the actual water surface (92) and the other vessel (48) is estimated from the range (WS, WP, LB, LS) and the geolocation of the GPS receiver (62) of the other vessel (48). The portion (74) of the actual ship-water interface (64) visible from the camera (24) is determined according to the estimated actual ship-water interface (64) and the position of the camera (24). The method according to [9], citing [6], wherein the visible portion (74) of the actual ship-water interface (64) is used as the actual solid object interface for comparison with the corresponding imaged solid object interface (54).

[11] The method according to any one of [1] to [6], wherein the reference data (RD) comprises elevation map data of the actual land (94) adjacent to the water body.

[12] The portion of the real solid object interface (82, 84, 86) visible from the camera (24) is determined from elevation map data by generating a textureless rendering image and extracting the portion from the textureless rendering image, as described in

[11] , citing [6].

[13] The method according to any one of [1] to

[12] , wherein the calibration value (CV) is adjusted until predetermined conditions are met.

[14] The method according to any one of [1] to

[13] , wherein the calibration value (CV) is determined by adjusting a predetermined calibration value (CV) of the camera (24) such that the difference is reduced.

[15] A system (20) for calibrating a camera (24) positioned on a ship (70), the system (20) comprising a memory configured to store reference data (RD) and / or calibration values ​​(CV), and a processing unit configured to perform the method described in any one of [1] to

[14] .

Claims

1. A method for calibrating a camera (24) positioned on the ship (70), The camera (24) receives image data (ID) of the image (40) captured by the camera (24), and the image (40) shows at least one imaged solid object (44, 48) and at least one imaged solid object interface (50, 52, 54, 56) between the imaged solid object (44, 48) and the imaged sky (46), or between the imaged solid object (44, 48) and the imaged water surface (42) of the water area in which the ship (70) is navigating. Extracting the captured solid object interfaces (50, 52, 54, 56) from the image data (ID), To determine the attitude of the ship (70) at the time the aforementioned image (40) was captured, Receiving reference data (RD), the reference data (RD) represents at least one real solid object interface (64, 82, 84, 86) in the real world (80) corresponding to the captured solid object interface (50, 52, 54, 56) in the image (40), where the real solid object interface (64, 82, 84, 86) is between the corresponding real solid object (66, 94) and the real sky (96), or between the corresponding real solid object (66, 94) and the corresponding real water surface (92) of the body of water. Extracting the actual solid object interfaces (64, 82, 84, 86) from the aforementioned reference data (RD), The difference between the extracted imaged solid object interfaces (50, 52, 54, 56) and the extracted real solid object interfaces (64, 82, 84, 86) is determined considering the determined attitude of the vessel (70), A method comprising determining the calibration value (CV) of the camera (24) according to the difference such that the difference decreases.

2. Extracting the captured solid object interfaces (50, 52, 54, 56) from the image (40) comprises extracting the paths of the captured solid object interfaces (50, 52, 54, 56) within the image (40), Extracting the actual solid object interfaces (64, 82, 84, 86) from the reference data (RD) comprises extracting the paths of the actual solid object interfaces (64, 82, 84, 86) from the reference data (RD), The method according to claim 1, wherein the difference between the imaged solid object interfaces (50, 52, 54, 56) and the actual solid object interfaces (64, 82, 84, 86) corresponds to the difference between the paths of the imaged solid object interfaces (50, 52, 54, 56) and the paths of the actual solid object interfaces (64, 82, 84, 86).

3. Determining the difference between the paths of the imaged solid object interfaces (50, 52, 54, 56) and the paths of the actual solid object interfaces (64, 82, 84, 86) comprises moving the paths of the imaged solid object interfaces (50, 52, 54, 56) and / or the paths of the actual solid object interfaces (64, 82, 84, 86) to the same coordinate system, and determining the misfit function between the imaged solid object interfaces (50, 52, 54, 56) and the actual solid object interfaces (64, 82, 84, 86) in the coordinate system as the difference within the coordinate system. The method according to claim 2, wherein determining the calibration value (CV) of the camera (24) such that the difference is reduced is further comprising determining the calibration value (CV) such that the misfit function is reduced.

4. The method according to claim 3, wherein a predetermined calibration value (CV) is changed such that the misfit function is minimized.

5. When extracting the actual solid object interfaces (64, 82, 84, 86) from the reference data (RD), only the portions of the actual solid object interfaces (64, 82, 84, 86) visible from the camera (24) are extracted. The method according to any one of claims 1 to 4, wherein determining the difference between the imaged solid object interfaces (50, 52, 54, 56) and the actual solid object interfaces (64, 82, 84, 86) is further comprising determining the difference between the imaged solid object interfaces (50, 52, 54, 56) and only the extracted visible portions of the actual solid object interfaces (64, 82, 84, 86).

6. The positional relationship between the camera (24) and the actual solid object interface (64, 82, 84, 86) is determined from the aforementioned reference data (RD), The method according to claim 5, further comprising determining the portion of the actual solid object interface (64, 82, 84, 86) visible from the camera (24) according to the determined positional relationship.

7. The method according to claim 6, wherein the reference data (RD) comprises 2D map data representing a 2D map of the area around the vessel (70).

8. The method according to claim 7, wherein the portion of the real solid object interface (64, 82, 84, 86) visible from the camera (24) is determined from the 2D map data by raycasting according to the determined positional relationship between the camera (24) and the real solid object interface (64, 82, 84, 86).

9. The method according to claim 6, wherein the reference data (RD) comprises AIS data relating to at least one other vessel (48) in the waters.

10. The AIS data comprises the geolocation of the GPS receiver (62) of the other vessel (48) and the range (WS, WP, LB, LS) of the other vessel (48) with respect to the geolocation of the GPS receiver (62) of the other vessel (48). The actual ship-to-water interface (64) between the actual water surface (92) and the other vessel (48) is estimated from the range (WS, WP, LB, LS) and the geolocation of the GPS receiver (62) of the other vessel (48). The portion (74) of the actual ship-water interface (64) visible from the camera (24) is determined according to the estimated actual ship-water interface (64) and the position of the camera (24). The method according to claim 9, wherein the visible portion (74) of the actual ship-water interface (64) is used as the actual solid object interface for comparison with the corresponding imaged solid object interface (54).

11. The method according to claim 6, wherein the reference data (RD) comprises elevation map data of the actual land area (94) adjacent to the water area.

12. The method according to claim 11, wherein the portion of the real solid object interface (82, 84, 86) visible from the camera (24) is determined from elevation map data by generating a textureless rendering image and by extracting the portion from the textureless rendering image.

13. The method according to any one of claims 1 to 4, wherein the calibration value (CV) is adjusted until predetermined conditions are met.

14. The method according to any one of claims 1 to 4, wherein the calibration value (CV) is determined by adjusting a predetermined calibration value (CV) of the camera (24) so ​​that the difference is reduced.

15. A system (20) for calibrating a camera (24) positioned on a ship (70), the system (20) comprising a memory configured to store reference data (RD) and / or calibration values ​​(CV), and a processing unit configured to perform the method according to any one of claims 1 to 4.

Images