An Automatic Calibration Method for Automotive Panoramic Cameras with Variable Environmental Reference Points

Abstract

The present invention belongs to the technical field of panoramic cameras, and specifically relates to an automatic calibration method for an automotive panoramic camera with variable environmental reference points. The method includes configuring the initial environmental reference point positions according to the calibration pattern of the calibration site; parking the vehicle in the calibration site and placing a checkerboard calibration cloth; obtaining the current environmental reference point positions of the calibration site; determining whether the environmental reference points are lost based on the current environmental reference point positions and the initial environmental reference point positions, and modifying the initial environmental reference point positions; calculating the extrinsic parameters of the camera according to the modified initial environmental reference point positions; and checking the extrinsic parameters of the camera. This realizes avoiding the decrease in calibration accuracy caused by the lack of environmental reference points in the calibration site, improves the success rate of calibration, optimizes the traditional checkerboard calibration method, and has better applicability to the dirty calibration site of the production line.

Description

Technical Field

[0001] The present invention relates to the technical field of panoramic cameras, and particularly to an automatic calibration method for an automotive panoramic camera with variable environmental reference points. Background Art

[0002] The 360-degree surround view system is increasingly applied to vehicles, bringing a more comfortable driving experience to drivers. The 360-degree surround view system stitches the images of four cameras into a panoramic view around the vehicle, and the stitching effect of the panoramic view depends on camera calibration. The black-and-white checkerboard pattern is a currently widely used calibration pattern. However, after long-term use of the calibration site on the production line, the entry and exit of vehicles will cause the calibration site to become dirty, affecting the calibration result.

[0003] Therefore, based on the above technical problems, it is necessary to design a new automatic calibration method for an automotive panoramic camera with variable environmental reference points. Summary of the Invention

[0004] The purpose of the present invention is to provide an automatic calibration method for an automotive panoramic camera with variable environmental reference points.

[0005] To solve the above technical problems, the present invention provides an automatic calibration method for an automotive panoramic camera with variable environmental reference points, including:

[0006] Configuring the initial environmental reference point positions according to the calibration pattern of the calibration site;

[0007] Parking the vehicle in the calibration site and placing a checkerboard calibration cloth;

[0008] Obtaining the current environmental reference point positions of the calibration site;

[0009] Judging whether the environmental reference points are lost according to the current environmental reference point positions and the initial environmental reference point positions, and modifying the initial environmental reference point positions;

[0010] Calculating the external parameters of the camera according to the modified initial environmental reference point positions; and

[0011] Checking the external parameters of the camera.

[0012] Further, the configuring the initial environmental reference point positions according to the calibration pattern of the calibration site includes:

[0013] Dispersedly selecting five environmental reference point positions for each of the four images as the initial environmental reference point positions according to the calibration pattern of the calibration site, and writing them into a configuration file;

[0014] The calibration pattern is a checkerboard pattern;

[0015] Five environmental reference points are suitable for selecting the environmental reference points on both sides and in the middle of the calibration pattern.

[0016] Further, parking the vehicle in the calibration site and placing the checkerboard calibration cloth includes:

[0017] The vehicle is suitable for parking in the middle position of the calibration site. After the vehicle is parked in the calibration site, a black-and-white checkerboard calibration cloth is placed around the vehicle, so that there are complete patterns in the calibration pattern in front of, behind, left and right of the vehicle.

[0018] Further, obtaining the current environmental reference point position of the calibration site includes:

[0019] Through the OpenCV vision operator, according to the image of the black-and-white checkerboard calibration cloth, obtain the contour of the black squares, and record the rows and columns where the black squares are located in the order from top to bottom and from left to right;

[0020] Determine the initial environmental reference point position in the configuration file in the black-and-white checkerboard calibration cloth.

[0021] Further, judging whether the environmental reference point is lost according to the current environmental reference point position and the initial environmental reference point position, and modifying the initial environmental reference point position includes:

[0022] Judge whether there is an environmental reference point lost in the four-way images. If the middle environmental reference point is lost, it does not affect the calibration. If the environmental reference points on both sides are lost, select a suitable current environmental reference point and synchronously modify the position of the initial environmental reference point in the configuration file;

[0023] If the loss of the environmental reference point is caused by the inability to recognize the upper black square, but the lower environmental reference point exists, and the number of rows and columns of the black squares in the current view has not changed, look for a black square in the column above the non-outermost black square where the current environmental reference point is located, and take the leftmost black square in this row as the alternative black square. The four vertices of the black square are used as alternative environmental reference points, and the optimal reference point is selected according to the principle that the row number of the reference point is the smallest and the column number is the smallest;

[0024] If the loss of the environmental reference point is caused by the inability to recognize the lower black square, but the upper environmental reference point exists, and the number of rows and columns of the black squares in the current view has not changed, look for a black square in the column below the non-outermost black square where the current environmental reference point is located, and take the leftmost black square in this row as the alternative black square. The four vertices of the black square are used as alternative environmental reference points, and the priority is selected according to the principle that the row number is the largest and the column number is the smallest.

[0025] Further, perform a verification on the optimal reference point P with the two environmental reference points Q and R on the other side respectively, calculate the distances between P and these two points respectively. For the front and rear views, the distance is required to be at least 5 times the side length of a black square. For the left and right views, the distance is required to be at least 9 times the side length of a black square. The side length of a black square in the image is c, and generate a judgment model:

[0026]

[0027] where n is the distance;

[0028] If the optimal reference point meets the judgment model, then use this optimal reference point as the new initial environmental reference point and write the position of the new initial environmental reference point into the configuration file.

[0029] Further, if the upper right black square part is missing, when looking for alternative black squares, select the rightmost black square in this row as the alternative black square, and select the alternative environmental reference point according to the principle of the smallest row number and the largest column number;

[0030] If the lower right black square part is missing, select the rightmost black square in this row as the alternative black square, and select the alternative environmental reference point according to the principle of the largest row number and the largest column number;

[0031] If the loss of black squares causes the loss of all two environmental reference points on the left or right, the calibration fails.

[0032] Further, the calculation of the external camera parameters based on the modified position of the initial environmental reference point includes:

[0033]

[0034] where P is the scale factor, is a point in the pixel coordinate system, is the internal camera parameter, is the external camera parameter, is a point in the world coordinate system.

[0035] Further, the inspection of the external camera parameters includes:

[0036] Use the root mean square error RMSE to judge whether the external camera parameters meet the requirements;

[0037]

[0038] where Y i represents the predicted value; f(x i ) represents the true value;

[0039] If the RMSE is greater than the given threshold, it means that the calculated external parameters are incorrect.

[0040] On the other hand, the present invention also provides a calibration system for an automotive panoramic camera automatic calibration method using the above variable environmental reference points, including:

[0041] An initial module that configures the position of the initial environmental reference point according to the calibration pattern of the calibration site;

[0042] A comparison module that parks the vehicle in the calibration site and places a checkerboard calibration cloth;

[0043] An acquisition module that acquires the current environmental reference point position of the calibration site;

[0044] A judgment module that judges whether the environmental reference point is lost according to the current environmental reference point position and the initial environmental reference point position, and modifies the initial environmental reference point position;

[0045] A parameter module that calculates the external parameters of the camera according to the modified initial environmental reference point position; and

[0046] An inspection module that inspects the external parameters of the camera.

[0047] The beneficial effects of the present invention are as follows: The present invention configures the position of the initial environmental reference point according to the calibration pattern of the calibration site; parks the vehicle in the calibration site and places a checkerboard calibration cloth; acquires the current environmental reference point position of the calibration site; judges whether the environmental reference point is lost according to the current environmental reference point position and the initial environmental reference point position, and modifies the initial environmental reference point position; calculates the external parameters of the camera according to the modified initial environmental reference point position; and inspects the external parameters of the camera. This avoids the decrease in calibration accuracy caused by the lack of environmental reference points in the calibration site, improves the success rate of calibration, optimizes the traditional checkerboard calibration method, and has better applicability to the dirty calibration site of the production line.

[0048] Other features and advantages of the present invention will be described in the following specification, and some of them will become obvious from the specification or be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification and the drawings.

[0049] To make the above objectives, features, and advantages of the present invention more obvious and understandable, the following specific preferred embodiments are given, and in conjunction with the accompanying drawings, the detailed description is as follows. Description of the Drawings

[0050] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0051] Figure 1 is a flowchart of an automatic calibration method for a panoramic camera of a vehicle with a variable environmental reference point according to the present invention;

[0052] Figure 2 is a schematic diagram of the calibration site according to the present invention;

[0053] Figure 3 is a schematic diagram of the inability to recognize the middle black square according to the present invention;

[0054] Figure 4 is a schematic diagram of the inability to recognize the upper black square according to the present invention;

[0055] Figure 5 is a schematic diagram of the inability to recognize the lower black square according to the present invention;

[0056] Figure 6 is a flowchart of the calibration process when the calibration site is dirty according to the present invention. Specific Embodiments

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions of the present invention in conjunction with the drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0058] Example 1

[0059] As Figures 1 to 6As shown, this embodiment 1 provides an automatic calibration method for a car panoramic camera with variable environmental reference points, including: configuring the initial environmental reference point position according to the calibration pattern of the calibration site; parking the vehicle in the calibration site and placing a checkerboard calibration cloth; obtaining the current environmental reference point position of the calibration site; judging whether the environmental reference point is lost according to the current environmental reference point position and the initial environmental reference point position, and modifying the initial environmental reference point position; calculating the camera extrinsic parameters according to the modified initial environmental reference point position; and checking the camera extrinsic parameters; avoiding the loss of environmental reference points in the calibration site that causes a decrease in calibration accuracy, improving the success rate of calibration, optimizing the traditional checkerboard calibration method, and having better applicability to dirty production line calibration sites.

[0060] In this embodiment, two environmental reference points are selected from the left and right parts of each calibration pattern, and one environmental reference point is selected from the middle part for recording; the vehicle is parked in the middle of the black and white grid calibration pattern (calibration pattern); the original image of the black and white grid pattern (calibration pattern) is obtained through four cameras; the environmental reference points in the image are identified through visual operators. If the ground is dirty due to long-term use of the venue, causing some environmental reference points to be lost, the algorithm will use the existing environmental reference points to select alternative environmental reference points to continue calibration, and calculate the external parameters in the world coordinate system of the four cameras based on the above environmental reference points.

[0061] In this embodiment, the configuration of the initial environment reference point position according to the calibration pattern of the calibration site includes: selecting five environment reference point positions of each of the four images as the initial environment reference point positions according to the calibration pattern of the calibration site, and writing them into the configuration file; the calibration pattern is a checkerboard pattern; the five environment reference points are suitable for selecting environment reference points on both sides and in the middle of the calibration pattern; the four vertices of the black squares of the checkerboard pattern in the calibration pattern can be used as environment reference points, the selected environment reference points are used as initial environment reference points, and the row numbers and column numbers of these initial environment reference points are written into the configuration file as unique identifiers. The calibration pattern of the calibration site is a checkerboard pattern, and there are multiple environment reference points in the checkerboard pattern. However, in order to ensure accurate splicing, the environment reference points on both sides and in the middle are given priority. Since the checkerboard pattern is black and white, the sizes of the black and white blocks are the same, and each view can see the number of black and white blocks in fixed rows and columns, the environment reference point can be represented by the position of the row and column where it is located, and the selected environment reference point is written into the configuration file.

[0062] In this embodiment, parking the vehicle in the calibration site and placing the checkerboard calibration cloth includes: the vehicle is suitable to be parked in the middle position of the calibration site to ensure that each camera can see all the patterns in the view. After the vehicle is parked in the calibration site, a black-and-white checkerboard calibration cloth (checkerboard pattern marker) is placed around the vehicle, so that there are complete patterns in the calibration patterns on the front, back, left, and right of the vehicle; the distance between the checkerboard patterns in the black-and-white checkerboard calibration cloth and the placement of the vehicle between the patterns should be appropriate, that is, to ensure that there are complete patterns in the original images on the front, back, left, and right sides.

[0063] In this embodiment, obtaining the current environmental reference point position of the calibration site includes: through the OpenCV vision operator, according to the image of the black-and-white checkerboard calibration cloth, obtaining the contours of the black squares, and recording the rows and columns where the black squares are located in the order from top to bottom and from left to right; determining the initial environmental reference point position in the configuration file in the black-and-white checkerboard calibration cloth.

[0064] In this embodiment, judging whether the environmental reference point is lost according to the current environmental reference point position and the initial environmental reference point position, and modifying the initial environmental reference point position includes: normally, there are 5 environmental reference points in each view, but dirt on the calibration cloth or the site will cause the environmental reference points to be undetectable; using the vision operator to process the picture can obtain the black squares in the pattern. Ideally, the black squares are clear and the environmental reference points can be easily identified. However, due to the long-term use of the calibration site, it is inevitable that there will be dirt, and the wheel marks are difficult to clean, which will cause the environmental reference points to be unrecognizable. In this case, the traditional calibration algorithm will fail due to the loss of reference points.

[0065] In this embodiment, record the rows and columns where the black squares are located in the order from top to bottom and from left to right. If the black squares in the middle part cannot be recognized and cause the loss of the environmental reference point, as Figure 3 shown, secondary processing will be carried out, and subsequent calibration will be performed through the remaining four environmental reference points; if some of the black squares on the left and right sides cannot be recognized, resulting in the loss of any one of the four environmental reference points, the algorithm will reselect a new reference point according to the environmental reference point closest to the lost environmental reference point.

[0066] Judge whether there is a loss of environmental reference points in the four-way images. If the middle environmental reference point is lost, it does not affect the calibration. If the environmental reference points on both sides are lost, select a suitable current environmental reference point and synchronously modify the position of the initial environmental reference point in the configuration file.

[0067] If the upper black square cannot be recognized, resulting in the loss of the environmental reference point, but the lower environmental reference point exists and the number of rows and columns of the black squares in the current view has not changed, search for a black square in the column above the non-outermost black square where the current environmental reference point is located. Select the leftmost black square in this row as the alternative black square, and use the four vertices of the black square as alternative environmental reference points. Select the optimal reference point according to the principle of the smallest row number and the smallest column number of the reference point; as follows Figure 4 As shown, the alternative reference points are A, B, C, and D. According to the selection principle, A is the optimal reference point;

[0068] If the lower black square cannot be recognized, resulting in the loss of the environmental reference point, but the upper environmental reference point exists and the number of rows and columns of the black squares in the current view has not changed, search for a black square in the column below the non-outermost black square where the current environmental reference point is located. Select the leftmost black square in this row as the alternative black square, and use the four vertices of the black square as alternative environmental reference points. The priority is selected according to the principle of the largest row number and the smallest column number; as follows Figure 5 As shown, the alternative reference points are A, B, C, and D. According to the selection principle, C is the optimal reference point.

[0069] In this embodiment, perform a verification on the optimal reference point P with the other two environmental reference points Q and R on the other side respectively, calculate the distances between P and these two points respectively. If it is a front and rear view, the distance is required to be at least the side length of 5 black squares (n = 5). If it is a left and right view, the distance is required to be at least the side length of 9 black squares (n = 9). The side length of a black square in the image is c, and a judgment model is generated:

[0070]

[0071] Among them, n is the distance; if the optimal reference point meets the judgment model, then use this optimal reference point as the new initial environmental reference point, and write the position of the new initial environmental reference point into the configuration file.

[0072] In this embodiment, if the upper right black square is partially lost, when searching for the alternative black square, select the rightmost black square in this row as the alternative black square, and select the alternative environmental reference points according to the principle of the smallest row number and the largest column number; if the lower right black square is partially lost, select the rightmost black square in this row as the alternative black square, and select the alternative environmental reference points according to the principle of the largest row number and the largest column number; if the loss of the black square causes the loss of both the left and right environmental reference points, the calibration fails; by complementing the lost environmental reference points, it is convenient to accurately obtain the external parameters of the camera, so that the camera can accurately obtain the image.

[0073] In this embodiment, calculating the extrinsic parameters of the camera based on the modified initial environmental reference point position includes: calculating the extrinsic parameters of the four cameras through OpenCV operators according to the found environmental reference points and the positions of the environmental reference points in the configuration file recorded by the vehicle;

[0074]

[0075] where P is the scale factor, is a point in the pixel coordinate system, is the intrinsic parameter of the camera,

[0076] is the extrinsic parameter of the camera, is a point in the world coordinate system; by obtaining the extrinsic parameters of the camera,

[0077] it is convenient to better display the images of the camera.

[0078] In this embodiment, checking the extrinsic parameters of the camera includes:

[0079] using the root mean square error RMSE to judge whether the extrinsic parameters of the camera meet the requirements;

[0080]

[0081] where Y i represents the predicted value; f(x i ) represents the true value; if the RMSE is greater than the given threshold, it indicates that the calculated extrinsic parameters are incorrect; by checking the extrinsic parameters of the camera, it can be judged whether the extrinsic parameters of the camera are accurate, which is convenient for determining whether each camera can accurately obtain images for stitching.

[0082] Embodiment 2

[0083] Based on Embodiment 1, this Embodiment 2 also provides a calibration system for an automotive panoramic camera using the variable environmental reference point in Embodiment 1, including: an initial module for configuring the initial environmental reference point position according to the calibration pattern of the calibration site; a comparison module for parking the vehicle in the calibration site and placing a checkerboard calibration cloth; an acquisition module for acquiring the current environmental reference point position of the calibration site; a judgment module for judging whether the environmental reference point is lost according to the current environmental reference point position and the initial environmental reference point position, and modifying the initial environmental reference point position; a parameter module for calculating the extrinsic parameters of the camera according to the modified initial environmental reference point position; and an inspection module for inspecting the extrinsic parameters of the camera.

[0084] In summary, the present invention configures the initial environmental reference point positions according to the calibration patterns of the calibration site; parks the vehicle in the calibration site and places the checkerboard calibration cloth; obtains the current environmental reference point positions of the calibration site; determines whether the environmental reference points are lost based on the current environmental reference point positions and the initial environmental reference point positions, and modifies the initial environmental reference point positions; calculates the external parameters of the camera according to the modified initial environmental reference point positions; and checks the external parameters of the camera, thus avoiding the reduction of calibration accuracy caused by the lack of environmental reference points in the calibration site, improving the success rate of calibration, optimizing the traditional checkerboard calibration method, and having better applicability to the dirty production line calibration site.

[0085] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0086] In addition, the functional modules in each embodiment of the present invention may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.

[0087] When the above-mentioned functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0088] Taking the above-mentioned ideal embodiments of the present invention as an inspiration, through the above description, relevant staff can, without departing from the technical idea of this invention, make various changes and modifications. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. An automatic calibration method for a panoramic camera of a vehicle with a variable environmental reference point, characterized in that, Including: Configuring the initial environmental reference point positions according to the calibration pattern of the calibration site; Parking the vehicle in the calibration site and placing a checkerboard calibration cloth; Obtaining the current environmental reference point positions of the calibration site; Judging whether the environmental reference points are lost according to the current environmental reference point positions and the initial environmental reference point positions, and modifying the initial environmental reference point positions; Calculating the external parameters of the camera according to the modified initial environmental reference point positions; And Checking the external parameters of the camera; The judging whether the environmental reference points are lost according to the current environmental reference point positions and the initial environmental reference point positions, and modifying the initial environmental reference point positions includes: Judging whether there are lost environmental reference points in the four-way images. If the middle environmental reference point is lost, it does not affect the calibration. If the environmental reference points on both sides are lost, select appropriate current environmental reference points and synchronously modify the positions of the initial environmental reference points in the configuration file; If the loss of the environmental reference points is caused by the inability to recognize the upper black square, but the environmental reference points below exist and the number of rows and columns of the black squares in the current view has not changed, search for black squares in the column above the non-outermost black square where the current environmental reference point is located, take the leftmost black square in this row as the alternative black square, and the four vertices of the black square as the alternative environmental reference points, and select the optimal reference point according to the principle of the smallest number of rows and the smallest number of columns of the reference points; If the loss of the environmental reference points is caused by the inability to recognize the lower black square, but the environmental reference points above exist and the number of rows and columns of the black squares in the current view has not changed, search for black squares in the column below the non-outermost black square where the current environmental reference point is located, take the leftmost black square in this row as the alternative black square, and the four vertices of the black square as the alternative environmental reference points, and select the priority according to the principle of the largest number of rows and the smallest number of columns; Perform a check on the optimal reference point P with the other two environmental reference points Q and R on the other side respectively, calculate the distances between P and these two points respectively. If it is a front and rear view, the distance is required to be at least the side length of 5 black squares. If it is a left and right view, the distance is required to be at least the side length of 9 black squares. The side length of a black square in the image is c, and a judgment model is generated: Where n is the distance; If the optimal reference point meets the judgment model, take the optimal reference point as the new initial environmental reference point and write the position of the new initial environmental reference point into the configuration file.

2. The automatic calibration method for an automotive panoramic camera with variable environmental reference points according to claim 1, wherein The configuring the initial environmental reference point positions according to the calibration pattern of the calibration site includes: Dispersedly selecting five environmental reference point positions in each of the four-way images according to the calibration pattern of the calibration site as the initial environmental reference point positions and writing them into the configuration file; The calibration pattern is a checkerboard pattern; The five environmental reference points are suitable for selecting the environmental reference points on both sides and in the middle of the calibration pattern.

3. The automatic calibration method for an automotive panoramic camera with variable environmental reference points according to claim 2, wherein The parking the vehicle in the calibration site and placing a checkerboard calibration cloth includes: The vehicle is suitable for being parked at the middle position of the calibration site. After the vehicle is parked in the calibration site, a black-and-white checkerboard calibration cloth is placed around the vehicle, so that there is a complete pattern in the calibration pattern in front of, behind, left and right of the vehicle.

4. The automatic calibration method of an automotive panoramic camera with variable environmental reference points according to claim 3, characterized in that The obtaining of the current environmental reference point position of the calibration site includes: Through the OpenCV vision operator, according to the image of the black-and-white checkerboard calibration cloth, obtain the contour of the black square, and record the rows and columns where the black squares are located in the order from top to bottom and from left to right; Determine the initial environmental reference point position in the configuration file in the black-and-white checkerboard calibration cloth.

5. The automatic calibration method of an automotive panoramic camera with variable environmental reference points according to claim 4, characterized in that If a part of the upper right black square is missing, when looking for an alternative black square, select the rightmost black square in this row as the alternative black square, and the alternative environmental reference point is selected according to the principle of the smallest row number and the largest column number; If a part of the lower right black square is missing, the alternative black square is selected as the rightmost black square in this row, and the alternative environmental reference point is selected according to the principle of the largest row number and the largest column number; If the loss of the black square causes the loss of all two environmental reference points on the left or right, the calibration fails.

6. The automatic calibration method of an automotive panoramic camera with variable environmental reference points according to claim 5, characterized in that The calculating of the external parameters of the camera according to the modified initial environmental reference point position includes: where P is the scale factor, is a point in the pixel coordinate system, is the camera internal parameter, is the camera external parameter, is a point in the world coordinate system.

7. The automatic calibration method of an automotive panoramic camera with variable environmental reference points according to claim 6, characterized in that The checking of the external parameters of the camera includes: Use the root mean square error RMSE to judge whether the external parameters of the camera meet the requirements; Among them, Y i represents the predicted value; f(x i ) represents the true value; If the RMSE is greater than the given threshold, it means that the calculated external parameters are incorrect.

8. A calibration system for an automatic calibration method of an automotive panoramic camera using a variable environmental reference point as described in claim 1, characterized in that, Including: An initial module, which configures the initial environmental reference point position according to the calibration pattern of the calibration site; A comparison module, which parks the vehicle in the calibration site and places the checkerboard calibration cloth; An obtaining module, which obtains the current environmental reference point position of the calibration site; A judgment module, which judges whether the environmental reference point is lost according to the current environmental reference point position and the initial environmental reference point position, and modifies the initial environmental reference point position; A parameter module, which calculates the external parameters of the camera according to the modified initial environmental reference point position; And An inspection module, which inspects the external parameters of the camera.

Images