Image calibration method and device, electronic equipment and storage medium

Abstract

The invention provides an image calibration method, and the method comprises the steps: sequentially adjusting the shooting of a calibration object at each attitude angle of a target camera according to a preset attitude angle sequence; for each attitude angle, video data of the target camera adjusted from the initial attitude angle to an expected attitude angle is obtained, each attitude angle corresponds to one video data, and the expected attitude angle is determined according to an expected image position of the calibration object in the video data; performing analysis processing on the video data of each attitude angle to obtain a calibration external parameter matrix of the target camera; obtaining a to-be-calibrated image of the target camera; and calibrating the target object in the to-be-calibrated image by calibrating the external parameter matrix. The method solves the problems that according to an existing calibration method, an operator needs to capture images among multiple devices at the same time, operation is tedious, the number of required devices is large, the operator cannot interact in the whole calibration process, and the operator cannot know whether the collected images are available or not in the collection process, so that the calibration result is inaccurate.

Description

Technical Field

[0001] This invention relates to the field of image management technology, and in particular to an image calibration method, apparatus, electronic device, and storage medium. Background Technology

[0002] Current calibration methods for cameras with limited fields of view primarily utilize multiple calibration boards and joint calibration across multiple calibration devices to form a shared field of view. These methods require operators to simultaneously capture images from multiple devices, making the process cumbersome and requiring numerous devices. Furthermore, the operator cannot interact with the data during the calibration process, and cannot determine the usability of the captured images. Therefore, a highly accurate image calibration method is urgently needed to address the problems of existing methods, such as the need for operators to simultaneously capture images from multiple devices, the lack of operator interaction during the calibration process, and the inability of the operator to determine the usability of the captured images, leading to inaccurate calibration results. Summary of the Invention

[0003] This invention provides an image calibration method to address the problems of existing calibration methods, such as the need for operators to simultaneously capture images across multiple devices, cumbersome operation, high equipment requirements, lack of operator interaction during the calibration process, and inaccurate calibration results due to the operator's inability to determine the usability of captured images. This invention adjusts the target camera's attitude angles sequentially according to a preset order, capturing images of the calibration object. For each attitude angle, video data of the target camera adjusting from its initial attitude angle to the desired attitude angle is acquired. The video data for each attitude angle is analyzed to obtain the target camera's calibration extrinsic parameter matrix. The image to be calibrated is then obtained from the target camera. The target object in the image is calibrated using the calibration extrinsic parameter matrix. This method solves the problems of existing calibration methods, which require operators to simultaneously capture images across multiple devices, are cumbersome, lack operator interaction during the calibration process, and have no way of knowing whether captured images are usable, leading to inaccurate calibration results.

[0004] In a first aspect, embodiments of the present invention provide an image calibration method, the method comprising the following steps:

[0005] According to the preset attitude angle sequence, adjust the attitude angles of the target camera to capture the target object in sequence;

[0006] For each attitude angle, video data of the target camera adjusting from the initial attitude angle to the desired attitude angle is acquired. Each attitude angle corresponds to one set of video data. The desired attitude angle is determined based on the desired image position of the calibration object in the video data.

[0007] The video data of each of the aforementioned attitude angles are analyzed and processed to obtain the calibration extrinsic parameter matrix of the target camera;

[0008] Acquire the image to be calibrated from the target camera;

[0009] The target object in the image to be calibrated is calibrated using the calibration extrinsic parameter matrix.

[0010] Optionally, acquiring video data of the target camera adjusting from the initial attitude angle to the desired attitude angle includes:

[0011] The target camera is adjusted from the first initial attitude angle to the first desired attitude angle to obtain the first video data;

[0012] Fix the first attitude angle, and adjust the target camera from the second initial attitude angle to the second desired attitude angle to obtain the second video data;

[0013] The first attitude angle and the second attitude angle are fixed, and the target camera is adjusted from the third initial attitude angle to the third desired attitude angle to obtain the third video data.

[0014] Optionally, the step of parsing the video data for each of the attitude angles to obtain the calibration extrinsic parameter matrix of the target camera includes:

[0015] The first video data is parsed and processed to obtain the first extrinsic parameter matrix corresponding to the first attitude angle;

[0016] The second video data is parsed and processed to obtain the second extrinsic parameter matrix corresponding to the second attitude angle;

[0017] The third video data is parsed to obtain the third extrinsic parameter matrix corresponding to the third attitude angle;

[0018] The calibration extrinsic matrix of the target camera is obtained based on the first extrinsic matrix, the second extrinsic matrix, and the third extrinsic matrix.

[0019] Optionally, the step of parsing the first video data to obtain the first extrinsic parameter matrix corresponding to the first attitude angle includes:

[0020] The first video data is parsed and processed to obtain multiple first attitude angle image data;

[0021] Based on multiple first attitude angle image data, the projection relationship between the calibration board coordinate system and the camera's phase plane coordinate system is determined;

[0022] Based on the aforementioned deployment relationship, the first attitude angle extrinsic parameter matrix of the current image frame is calculated;

[0023] Based on the first attitude angle extrinsic parameter matrix of the current image frame, the first extrinsic parameter matrix corresponding to the first attitude angle is obtained.

[0024] Optionally, the step of parsing the second video data to obtain the second extrinsic parameter matrix corresponding to the second attitude angle includes:

[0025] The second video data is parsed and processed to obtain multiple second attitude angle image data;

[0026] Based on the second attitude angle image data, a second attitude angle list is determined;

[0027] Based on the second attitude angle list and the equation of the fitted circle, the second extrinsic parameter matrix corresponding to the second attitude angle is obtained.

[0028] Optionally, determining the second extrinsic parameter matrix corresponding to the second attitude angle based on the second attitude angle list and the equation of the fitted circle includes:

[0029] Based on the second attitude angle list and the equation of the fitted circle, the change in the radius of the circle of the second initial attitude angle with respect to the second desired attitude angle is calculated;

[0030] Based on the change in radius of the circle corresponding to the second initial attitude angle with respect to the second desired attitude angle, the second extrinsic parameter matrix corresponding to the second attitude angle is obtained.

[0031] Optionally, the step of parsing the third video data to obtain the third extrinsic parameter matrix corresponding to the third pose angle includes:

[0032] The third video data is parsed and processed to obtain multiple third attitude angle image data;

[0033] Based on the third attitude angle image data, a list of third attitude angles is determined;

[0034] Based on the list of third attitude angles and the equation of the fitted circle, the third extrinsic parameter matrix corresponding to the third attitude angle is obtained.

[0035] Optionally, determining the third extrinsic parameter matrix corresponding to the third attitude angle based on the third attitude angle list and the equation of the fitted circle includes:

[0036] Based on the third attitude angle list and the equation of the fitted circle, the change of the third initial attitude angle with respect to the third desired attitude angle is calculated;

[0037] Based on the change of the third initial attitude angle with respect to the third desired attitude angle, the third extrinsic parameter matrix corresponding to the third attitude angle is obtained.

[0038] In a second aspect, embodiments of the present invention provide an image calibration device, the image calibration device comprising:

[0039] The shooting module is used to sequentially adjust the various attitude angles of the target camera to shoot at the calibration object according to a preset attitude angle sequence;

[0040] The first acquisition module is used to acquire video data of the target camera adjusting from the initial attitude angle to the desired attitude angle for each attitude angle. Each attitude angle corresponds to one set of video data, and the desired attitude angle is determined based on the desired image position of the calibration object in the video data.

[0041] The parsing and processing module is used to parse and process the video data of each of the attitude angles to obtain the calibration extrinsic parameter matrix of the target camera;

[0042] The second acquisition module is used to acquire the image to be calibrated from the target camera;

[0043] The calibration module is used to calibrate the target object in the image to be calibrated using the calibration extrinsic parameter matrix.

[0044] Thirdly, embodiments of the present invention provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the image calibration method provided in embodiments of the present invention.

[0045] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the image calibration method provided in the embodiments of the present invention.

[0046] In this embodiment of the invention, the target camera's attitude angles are sequentially adjusted to capture images of the calibration object according to a preset attitude angle order. For each attitude angle, video data of the target camera adjusting from the initial attitude angle to the desired attitude angle is acquired, with each attitude angle corresponding to one video data point. The desired attitude angle is determined based on the desired image position of the calibration object in the video data. The video data for each attitude angle is parsed to obtain the calibration extrinsic parameter matrix of the target camera. The image to be calibrated from the target camera is acquired. The target object in the image to be calibrated is calibrated using the calibration extrinsic parameter matrix. This invention solves the problems of existing calibration methods, which require operators to simultaneously capture images from multiple devices, resulting in cumbersome operations, numerous devices, and a lack of operator interaction during the calibration process. Furthermore, the operator cannot determine the usability of the captured images during the acquisition process, leading to inaccurate calibration results. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of an image calibration method provided in an embodiment of the present invention;

[0049] Figure 2 This is a diagram of another image calibration method provided in an embodiment of the present invention;

[0050] Figure 3 This is a diagram of acquiring video data provided in an embodiment of the present invention.

[0051] Figure 4 This is a perspective field diagram provided in an embodiment of the present invention;

[0052] Figure 5 This is a schematic diagram of the structure of an image calibration device provided in an embodiment of the present invention;

[0053] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] like Figure 1 As shown, Figure 1 This is a flowchart of an image calibration method provided by an embodiment of the present invention. The image calibration method includes the following steps:

[0056] 101. Following the preset attitude angle sequence, adjust the various attitude angles of the target camera sequentially to capture images of the calibration object.

[0057] In this embodiment of the invention, the image calibration method described above can be applied to an image calibration platform. This platform can be built on a server-based or distributed architecture and includes a data interface (for sensors or users to upload data), a knowledge database, and a knowledge database construction program. The data interface can be used to sequentially adjust the various attitude angles of the target camera to capture images of the calibration object according to a preset attitude angle sequence. The knowledge database construction program can be used to construct the knowledge database, which is specifically designed to provide additional relational information for the identified data entities, thereby enhancing the data recognition system's understanding of the content.

[0058] The preset attitude angle sequence can be understood as the order in which the system adjusts the camera's attitude angles. These attitude angles include roll, pitch, and yaw, which determine the orientation of the calibration object in three-dimensional space. The preset attitude angle sequence can be roll, pitch, yaw, or yaw, pitch, roll, etc.

[0059] The aforementioned target camera can be understood as a camera that photographs the target object.

[0060] The aforementioned calibration material can be a standard measuring instrument or item used in the calibration process.

[0061] The video data described above records a continuous sequence of images of the target camera at various attitude angles relative to the calibration object.

[0062] Understandably, the target camera can be adjusted sequentially according to a preset attitude angle order to take pictures of the calibration object. Each attitude angle adjustment is followed by a picture until all attitude angle adjustments are completed.

[0063] In one possible embodiment, when the preset attitude angle sequence is roll angle, pitch angle, and yaw angle, the target camera's roll angle is first adjusted to capture the image of the calibration object, then the target camera's pitch angle is adjusted to capture the image of the calibration object, and finally the target camera's yaw angle is adjusted to capture the image of the calibration object; when the preset attitude angle sequence is yaw angle, pitch angle, and roll angle, the target camera's yaw angle is first adjusted to capture the image of the calibration object, then the target camera's pitch angle is adjusted to capture the image of the calibration object, and finally the target camera's roll angle is adjusted to capture the image of the calibration object, and so on.

[0064] 102. For each attitude angle, acquire video data of the target camera as it adjusts from the initial attitude angle to the desired attitude angle.

[0065] In this embodiment of the invention, each attitude angle corresponds to one video data point. The aforementioned desired attitude angle can be understood as the ideal attitude angle position to be achieved when performing the task, and the desired attitude angle is determined based on the desired image position of the calibration object in the video data.

[0066] The aforementioned initial attitude angle can be understood as the predetermined or default attitude angle when starting to execute a task or operation.

[0067] The aforementioned attitude angles include roll, pitch, and yaw.

[0068] Specifically, for roll, pitch, and yaw, it is necessary to acquire video data of the target camera adjusting from the initial roll angle to the desired roll angle, video data of the target camera adjusting from the initial pitch angle to the desired pitch angle, and video data of the target camera adjusting from the initial yaw angle to the desired yaw angle.

[0069] 103. Analyze and process the video data at each attitude angle to obtain the calibration extrinsic parameter matrix of the target camera.

[0070] In this embodiment of the invention, the above-mentioned parsing process can be understood as the process of extracting the calibration extrinsic parameter matrix of the target camera from video data.

[0071] Furthermore, the extrinsic parameter matrices for each pose angle can be extracted from the video data, and the calibration extrinsic parameter matrix of the target camera can be calculated based on these matrices. Functions in the OpenCV library can be used to extract feature points from the video data for each pose angle, and these feature points can then be used to calculate the calibration extrinsic parameter matrix of the target camera. The OpenCV library is an open-source computer vision library that provides image processing and computer vision algorithms for tasks such as feature detection and image processing.

[0072] The aforementioned calibration extrinsic parameter matrix can be understood as a matrix used to describe the pose relationship of the target camera relative to the world coordinate system. It can be understood that the calibration extrinsic parameter matrix contains the position and orientation information of the target camera in the world coordinate system.

[0073] 104. Obtain the image to be calibrated from the target camera.

[0074] In this embodiment of the invention, the image to be calibrated can be understood as an image used for camera calibration. The image to be calibrated can be multiple scene images taken by the target camera at different angles and distances, or it can be an image directly selected from existing scene images.

[0075] 105. The target object in the image to be calibrated is calibrated by calibrating the extrinsic parameter matrix.

[0076] In this embodiment of the invention, the position of the target object in the image and its position in actual space can be mapped by using the calibration external acquisition matrix of the target camera and the camera extrinsic parameter matrix, thereby achieving accurate positioning of the target object.

[0077] The aforementioned target objects can be understood as specific objects that need to be identified and located in the image, such as roadblocks or road surface defects.

[0078] It should be noted that this invention allows for a visual observation of changes in the camera's field of view using visualization tools. Based on these changes, the final target calibration position can be determined, and the target object in the image to be calibrated can be calibrated according to this position. Throughout the entire process, the visible range and field of view of the target camera can be observed via a mobile device, allowing for a direct perception of the changes in perspective caused by changes in camera posture. This solves the problems of existing calibration methods, which require operators to simultaneously capture images across multiple devices, resulting in cumbersome operations, numerous devices, and a lack of operator interaction during the calibration process. Furthermore, operators cannot determine the usability of the captured images during the acquisition process, leading to inaccurate calibration results.

[0079] In this embodiment of the invention, the target camera's attitude angles are sequentially adjusted to capture images of the calibration object according to a preset attitude angle order. For each attitude angle, video data of the target camera adjusting from the initial attitude angle to the desired attitude angle is acquired, with each attitude angle corresponding to one video data point. The desired attitude angle is determined based on the desired image position of the calibration object in the video data. The video data for each attitude angle is parsed to obtain the calibration extrinsic parameter matrix of the target camera. The image to be calibrated from the target camera is acquired. The target object in the image to be calibrated is calibrated using the calibration extrinsic parameter matrix. This invention solves the problems of existing calibration methods, which require operators to simultaneously capture images from multiple devices, resulting in cumbersome operations, numerous devices, and a lack of operator interaction during the calibration process. Furthermore, the operator cannot determine the usability of the captured images during the acquisition process, leading to inaccurate calibration results.

[0080] It is understood that in the specific implementation of this application, data such as attitude angle data, image data, video data, knowledge data, and user data are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required. Furthermore, the collection, use, and processing of related data, as well as the training, deployment, and invocation of algorithm models, must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0081] like Figure 2 As shown, Figure 2This is an illustration of another image calibration method provided by an embodiment of the present invention. Specifically, during the operation of the present invention, the operator can intuitively see the changes in the camera's field of view using visualization tools, and adjust the final target position according to the changes in the camera's field of view. The entire process is perceptible and interactive. The present invention allows the operator to intuitively see the changes in the camera's field of view using visualization tools, and determine the final target calibration position based on the changes in the camera's field of view. The target object in the image to be calibrated is then calibrated based on the target calibration position. Throughout the entire process of the present invention, the visible range and field of view of the target camera can be observed through a mobile phone, allowing the operator to intuitively perceive the changes in the field of view caused by changes in the camera's posture. This solves the problems of existing calibration methods, which require the operator to simultaneously capture images from multiple devices, resulting in cumbersome operation, numerous devices required, and the inability for the operator to interact during the entire calibration process. Furthermore, the operator cannot know whether the captured images are usable during the acquisition process, leading to inaccurate calibration results.

[0082] Optionally, in the step of acquiring video data of the target camera adjusted from the initial attitude angle to the desired attitude angle, the target camera can be adjusted from the first initial attitude angle to the first desired attitude angle to obtain the first video data; the first attitude angle is fixed, and the target camera is adjusted from the second initial attitude angle to the second desired attitude angle to obtain the second video data; the first attitude angle and the second attitude angle are fixed, and the target camera is adjusted from the third initial attitude angle to the third desired attitude angle to obtain the third video data.

[0083] In this embodiment of the invention, the target camera can be understood as a camera that captures images of a calibration object.

[0084] The aforementioned attitude angles include roll, pitch, and yaw. Attitude angles determine the orientation of the calibration object in three-dimensional space.

[0085] The initial attitude angle can be understood as the predetermined or default attitude angle at the start of a task or operation. The desired attitude angle is determined based on the desired image position of the calibration object in the video data.

[0086] The aforementioned first initial attitude angle, second initial attitude angle, and third initial attitude angle can be determined according to a preset attitude angle order. For example, when the preset attitude angle order is roll angle, pitch angle, and yaw angle, the first initial attitude angle is the initial roll angle, the second initial attitude angle is the initial pitch angle, and the third initial attitude angle is the initial yaw angle, etc. When the first initial attitude angle is the initial roll angle, the first desired attitude angle is the desired roll angle; when the second initial attitude angle is the initial pitch angle, the second desired pitch angle is the desired pitch angle; and when the third initial attitude angle is the initial yaw angle, the third desired attitude angle is the desired yaw angle.

[0087] It should be noted that the system can acquire video data of the target camera adjusting from its initial attitude angle to the desired attitude angle according to a preset attitude angle order. The preset attitude angle order is the order in which the camera attitude angles are adjusted, which can be roll angle, pitch angle, yaw angle, or yaw angle, pitch angle, roll angle, etc.

[0088] In one possible embodiment, when the preset attitude angle sequence is roll angle, pitch angle, and yaw angle, the first initial attitude angle can be the initial roll angle, the first desired attitude angle can be the desired roll angle, and the first video data can be roll angle video data. The target camera can be adjusted from the initial roll angle to the desired roll angle position to obtain roll angle video data. The second initial attitude angle can be the initial pitch angle, the second desired attitude angle can be the desired pitch angle, and the second video data can be pitch angle video data. By fixing the roll angle, the target camera can be adjusted from the initial pitch angle to the desired pitch angle position to obtain pitch angle video data. The third initial attitude angle can be the initial yaw angle, the third desired attitude angle can be the desired yaw angle, and by fixing the roll angle and pitch angle, the target camera can be adjusted from the initial yaw angle to the desired yaw angle position to obtain yaw angle video data.

[0089] like Figure 3 As shown, Figure 3 This is a diagram of video data acquisition provided by an embodiment of the present invention. Specifically, the attitude angles include roll, pitch, and yaw, which define the orientation of the calibration object in three-dimensional space. When the preset attitude angle sequence is yaw, pitch, and roll, the first initial attitude angle can be the initial yaw angle, the first desired attitude angle can be the desired yaw angle, and the first video data can be yaw angle video data. The target camera can be adjusted from the initial yaw angle to the desired yaw angle position to obtain yaw angle video data. The second initial attitude angle can be the initial pitch angle, the second desired attitude angle can be the desired pitch angle, and the second video data can be pitch angle video data. By fixing the yaw angle, the target camera can be adjusted from the initial pitch angle to the desired pitch angle position to obtain pitch angle video data. The third initial attitude angle can be the initial roll angle, the third desired attitude angle can be the desired roll angle, and by fixing the yaw angle and pitch angle, the target camera can be adjusted from the initial roll angle to the desired roll angle position to obtain roll angle video data.

[0090] Optionally, in the step of parsing and processing the video data at each attitude angle to obtain the calibration extrinsic parameter matrix of the target camera, the first video data can be parsed and processed to obtain the first extrinsic parameter matrix corresponding to the first attitude angle; the second video data can be parsed and processed to obtain the second extrinsic parameter matrix corresponding to the second attitude angle; the third video data can be parsed and processed to obtain the third extrinsic parameter matrix corresponding to the third attitude angle; and the calibration extrinsic parameter matrix of the target camera can be obtained based on the first extrinsic parameter matrix, the second extrinsic parameter matrix, and the third extrinsic parameter matrix.

[0091] In this embodiment of the invention, the above-described analytical processing can be understood as the process of analyzing and processing video data to extract the extrinsic parameter matrix corresponding to the attitude angle in the video data.

[0092] The video data described above records a continuous sequence of images of the target camera at various attitude angles relative to the calibration object.

[0093] The aforementioned first video data can be video data obtained when the target camera adjusts from a first initial attitude angle to a first desired attitude angle. For example, when the first initial attitude angle is the initial roll angle and the first desired attitude angle is the desired roll angle, the first video data is roll angle video data; when the first initial attitude angle is the initial pitch angle and the first desired attitude angle is the desired pitch angle, the first video data is pitch angle video data, etc.

[0094] The aforementioned second video data can be video data obtained by fixing the first attitude angle and adjusting the target camera from the second initial attitude angle to the second desired attitude angle. For example, when the first attitude angle is the roll angle, the second initial attitude angle is the initial pitch angle, and the second desired attitude angle is the desired pitch angle, the second video data is pitch angle video data; when the first attitude angle is the yaw angle, the second initial attitude angle is the initial roll angle, and the second desired attitude angle is the desired roll angle, the second video data is roll angle video, etc.

[0095] The third video data can be video data obtained by fixing the first and second attitude angles and adjusting the target camera from the third initial attitude angle to the third desired attitude angle. For example, when the first attitude angle is the roll angle, the second attitude angle is the pitch angle, the third initial attitude angle is the initial yaw angle, and the third desired attitude angle is the desired yaw angle, the third video data is yaw angle video data; when the first attitude angle is the yaw angle, the second attitude angle is the roll angle, the third initial attitude angle is the initial pitch angle, and the third desired attitude angle is the desired pitch angle, etc.

[0096] The aforementioned extrinsic parameter matrix can be understood as the parameters of position and orientation in the world coordinate system.

[0097] The first extrinsic parameter matrix can be the extrinsic parameter matrix corresponding to the first attitude angle, the second extrinsic parameter matrix can be the extrinsic parameter matrix corresponding to the second attitude angle, and the third extrinsic parameter matrix can be the extrinsic parameter matrix corresponding to the third attitude angle.

[0098] In one possible embodiment, when the first video data is roll angle video data, the second video data is yaw angle video data, and the third video data is pitch angle video data, the roll angle video data can be parsed to obtain the extrinsic parameter matrix corresponding to the roll angle; the yaw angle video data can be parsed to obtain the extrinsic parameter matrix corresponding to the yaw angle; the pitch angle video data can be parsed to obtain the extrinsic parameter matrix corresponding to the pitch angle; and using the extrinsic parameter matrices corresponding to the roll angle, yaw angle, and pitch angle, the calibration object extrinsic parameter matrix of the target camera can be obtained.

[0099] In another possible embodiment, the first extrinsic parameter matrix is ​​T 第一期望姿态角 The second extrinsic parameter matrix is ​​T. 第二期望姿态角 The third extrinsic parameter matrix is ​​T. 第三期望姿态角 The calibration extrinsic parameter matrix of the target camera can be calculated using the following formula:

[0100] T 标定外参矩阵 =T 第一期望姿态角 *T 第二期望姿态角 *T 第三期望姿态角 .

[0101] Optionally, in the step of parsing the first video data to obtain the first extrinsic parameter matrix corresponding to the first attitude angle, the first video data can be parsed to obtain multiple first attitude angle image data; based on the multiple first attitude angle image data, the projection relationship between the calibration board coordinate system and the camera's phase plane coordinate system can be determined; based on the projection relationship, the first attitude angle extrinsic parameter matrix of the current image frame can be calculated; and based on the first attitude angle extrinsic parameter matrix of the current image frame, the first extrinsic parameter matrix corresponding to the first attitude angle can be obtained.

[0102] In this embodiment of the invention, the above-mentioned parsing process can be understood as the process of analyzing and processing the first video data to obtain multiple first attitude angle image data.

[0103] The above-mentioned deployment relationship can be understood as a mapping relationship between the calibration plate coordinate system and the camera phase plane coordinate system.

[0104] The aforementioned extrinsic parameter matrix can be understood as the parameters of position and orientation in the world coordinate system.

[0105] In one possible embodiment, for example, if the calibration board coordinate system is (x, y) and the camera phase plane coordinate system is (x', y'), then the projection relationship can be expressed as: x = k1 * x' + k2, y = k3 * y' + k4, where k1, k2, k3, and k4 are parameters to be determined. The parameters of the projection relationship can be obtained through the system of equations, and the extrinsic parameter matrix of the first attitude angle of the current image frame can be calculated based on the projection relationship. For example, if the first attitude angle is θ, then the extrinsic parameter matrix can be expressed as: [cosθ sinθ 0; -sinθ cosθ 0; 0 0 1]. By mapping the extrinsic parameter matrix to the first attitude angle, the first extrinsic parameter matrix corresponding to the first attitude angle is obtained.

[0106] It should be noted that the first video data can be parsed to obtain multiple first attitude angle image data. The calibration board can be seen in these first attitude angle image data. Based on the dimensions of the calibration board and the dimensions of the calibration corner points, the projection relationship between the calibration board coordinate system and the camera's phase plane coordinate system can be determined. Then, using this projection relationship and the Epnp method, the extrinsic parameter matrix of the current image frame can be calculated. Based on this extrinsic parameter matrix, the first extrinsic parameter matrix corresponding to the first attitude angle can be obtained. The Epnp method is an efficient algorithm for solving the Perspective-n-Point (PnP) problem. The core of the Epnp method is to estimate the camera pose in O(n) time complexity through linearization.

[0107] Optionally, in the step of parsing the second video data to obtain the second extrinsic parameter matrix corresponding to the second attitude angle, the second video data can be parsed to obtain multiple second attitude angle image data; based on the second attitude angle image data, a list of second attitude angles can be determined; based on the list of second attitude angles and the equation of the fitted circle, the second extrinsic parameter matrix corresponding to the second attitude angle can be obtained.

[0108] In this embodiment of the invention, the above-mentioned parsing process can be understood as the process of analyzing and processing the second video data to obtain multiple second pose angle image data.

[0109] The above list of second attitude angles contains an array or list of multiple second attitude angles.

[0110] The above-mentioned fitted circle equation can be understood as obtaining a circle equation to describe the distribution of a set of data points by minimizing the sum of squared errors. Specifically, the center and radius parameters of the circle can be obtained by solving a system of quadratic equations, thus yielding the circle equation.

[0111] Specifically, multiple second attitude angle image data can be parsed from the second video data, and the second attitude angle of each image from the second initial attitude angle at the beginning of the video movement position to the second desired attitude angle position can be calculated using the perspective field method, forming a second attitude angle list. The equation of the circle satisfied by the extrinsic parameter matrix points can be fitted to the second attitude angle list, and the second extrinsic parameter matrix corresponding to the second attitude angle can be calculated based on the equation of the circle.

[0112] like Figure 4 As shown, Figure 4 This is a perspective field diagram provided in an embodiment of the present invention. Specifically, the perspective field consists of an upward vector u and a latitude v. The upward vector u is the opposite direction of gravity, and the latitude v is the angle between the line connecting the 3D point to the optical center and the horizontal plane.

[0113] Optionally, in the step of determining the second extrinsic parameter matrix corresponding to the second attitude angle based on the second attitude angle list and the equation of the fitted circle, the change in the radius of the circle of the second initial attitude angle with respect to the second desired attitude angle can be calculated based on the second attitude angle list and the equation of the fitted circle; and the second extrinsic parameter matrix corresponding to the second attitude angle can be obtained based on the change in the radius of the circle of the second initial attitude angle with respect to the second desired attitude angle.

[0114] In this embodiment of the invention, the above-mentioned second attitude angle list includes an array or list of multiple second attitude angles.

[0115] The above-mentioned fitted circle equation can be understood as obtaining a circle equation to describe the distribution of a set of data points by minimizing the sum of squared errors. Specifically, the center and radius parameters of the circle can be obtained by solving a system of quadratic equations, thus yielding the circle equation.

[0116] Specifically, based on the list of second attitude angles and the equation of the fitted circle, the change in the radius of the circle between the second initial attitude angle and the second desired attitude angle can be calculated. Then, using this change in the radius of the circle between the second initial attitude angle and the second desired attitude angle, the second extrinsic parameter matrix corresponding to the second attitude angle can be obtained. For example, the change in the radius of the circle between each second initial attitude angle and the second desired attitude angle can be calculated using the following formula:

[0117] R = (x - h)^2 + (y - k)^2

[0118] Where (x, y) are the coordinates of the second initial attitude angle, and (h, k) are the coordinates of the second desired attitude angle. Each radius change can be divided by the maximum radius change, and the result can be multiplied by 100 to obtain the percentage change. The second extrinsic parameter matrix corresponding to the second attitude angle can then be obtained based on the percentage change.

[0119] In one possible embodiment, an image is parsed from the second video data, and the position P from the start of movement in the video is calculated using a perspective field method. 第二初始姿态角 To position P 第二期望姿态角 The second pose angle of each image forms , where at position P 第二初始姿态角 Since the calibration plate is still visible, at position P 第二初始姿态角 At the same time, the entire extrinsic parameter matrix T is calculated. 第二姿态角 The entire extrinsic parameter matrix T is fitted from the second attitude angle list. 第二姿态角 The equation of a circle satisfied by points, refined based on the equation of the circle.

[0120] P 第二期望姿态角 The second attitude angle at point P is such that, since only the camera's second attitude angle changes throughout the entire process, the algorithm is based on P. 第二期望姿态角 The second attitude angle at point P is calculated relative to P. 第二初始姿态角 radius of the circle at the location The change in P is used to calculate the value of P. 第二期望姿态角 The extrinsic parameter matrix T at the location 第二期望姿态角 .

[0121] In another possible embodiment, when the second attitude angle is the yaw angle, the image is parsed from the second video data, and the position from the start of movement in the video is calculated using the perspective field method. arrive Location of each image Angles, form a list Among them, in The calibration plate can be seen in the location, therefore, in At the same time, the entire extrinsic parameter matrix is ​​calculated at this location. ,from Fit the entire extrinsic parameter matrix from the list The equation of a circle satisfied by points, refined based on the equation of the circle. place Angle, because only the camera is present throughout the entire process. The angle has changed, and the algorithm is based on... place Calculate the angle relative to Location The change, thus calculating Location .

[0122] Optionally, in the step of parsing the third video data to obtain the third extrinsic parameter matrix corresponding to the third attitude angle, the third video data can be parsed to obtain multiple third attitude angle image data; based on the third attitude angle image data, a list of third attitude angles can be determined; and based on the list of third attitude angles and the equation of the fitted circle, the third extrinsic parameter matrix corresponding to the third attitude angle can be obtained.

[0123] In this embodiment of the invention, the above-mentioned parsing process can be understood as the process of analyzing and processing the third video data to obtain multiple third pose angle image data.

[0124] The above list of third attitude angles contains an array or list of multiple third attitude angles.

[0125] The above-mentioned fitted circle equation can be understood as obtaining a circle equation to describe the distribution of a set of data points by minimizing the sum of squared errors. Specifically, the center and radius parameters of the circle can be obtained by solving a system of quadratic equations, thus yielding the circle equation.

[0126] Specifically, multiple third attitude angle image data can be parsed from the third video data, and the third attitude angle of each image from the third initial attitude angle at the beginning of the video movement to the third desired attitude angle position can be calculated using the perspective field method, forming a list of third attitude angles. The equation of the circle satisfied by the extrinsic parameter matrix points can be fitted to the list of third attitude angles, and the third extrinsic parameter matrix corresponding to the third attitude angle can be calculated based on the equation of the circle.

[0127] The aforementioned perspective field consists of an upward vector u and a latitude φ. The upward vector u is the opposite direction of gravity, and the latitude φ is the angle between the line connecting the 3D point to the optical center and the horizontal plane.

[0128] Optionally, in the step of determining the third extrinsic parameter matrix corresponding to the third attitude angle based on the third attitude angle list and the equation of the fitted circle, the change of the third initial attitude angle with respect to the third desired attitude angle can be calculated based on the third attitude angle list and the equation of the fitted circle; and the third extrinsic parameter matrix corresponding to the third attitude angle can be obtained based on the change of the third initial attitude angle with respect to the third desired attitude angle.

[0129] In this embodiment of the invention, the above-mentioned third attitude angle list includes an array or list of multiple third attitude angles.

[0130] The above-mentioned fitted circle equation can be understood as obtaining a circle equation to describe the distribution of a set of data points by minimizing the sum of squared errors. Specifically, the center and radius parameters of the circle can be obtained by solving a system of quadratic equations, thus yielding the circle equation.

[0131] Specifically, based on the list of third attitude angles and the equation of the fitted circle, the change in radius of the circle between the third initial attitude angle and the third desired attitude angle can be calculated. Then, using this change in radius, the third extrinsic parameter matrix corresponding to the third attitude angle can be obtained. For example, the radius change of the circle between each third initial attitude angle and the third desired attitude angle can be calculated using the following formula:

[0132] R = (x - h)^2 + (y - k)^2

[0133] Where (x, y) are the coordinates of the third initial attitude angle, and (h, k) are the coordinates of the third desired attitude angle. Each radius change can be divided by the maximum radius change, and the result can be multiplied by 100 to obtain the percentage change. The third extrinsic parameter matrix corresponding to the third attitude angle can then be obtained based on the percentage change.

[0134] In one possible embodiment, an image is parsed from third video data, and the position P from the start of movement in the video is calculated using a perspective field method. 第三初始姿态角 To position P 第三期望姿态角 The third attitude angle of each image is used to form a third attitude angle list = {third attitude angle 1, third attitude angle, third attitude angle 3, ..., third desired attitude angle.}, where, at position P 第三初始姿态角 Since the calibration plate is still visible, at position P 第三初始姿态角 At the same time, the entire extrinsic parameter matrix T is calculated. 第三姿态角 The entire extrinsic parameter matrix T is fitted from the third attitude angle list. 第三姿态角 The equation of a circle satisfied by points, refined based on the equation of the circle.

[0135] P 第三期望姿态角 The third attitude angle at point P is determined by the fact that only the camera's third attitude angle changes throughout the entire process. The algorithm then uses P... 第三期望姿态角 The third attitude angle at point P is calculated relative to P. 第三初始姿态角 The change in the radius R of the circle at the location is used to calculate P. 第三期望姿态角 The extrinsic parameter matrix T at the location 第三期望姿态角 .

[0136] In another possible embodiment, when the third attitude angle is the yaw angle (pitch), the image is parsed from the third video data, and the pitch angle of each image from the video start position P_(pitch_begin) to the P_tpitch position is calculated using the perspective field method, forming a list PITCH={pitch1,pitch2,pitch3..tpitch.}. Here, the calibration board can be seen at the P_(pitch_begin) position. Therefore, the entire extrinsic parameter matrix T_(pitch_begin) at this time is calculated at the P_(pitch_begin) position. The equation of the circle satisfied by the entire extrinsic parameter matrix T_(pitch_begin) is fitted from the PITCH list. The pitch angle at P_tpitch is refined according to the equation of the circle. Since only the camera's pitch angle changes during the whole process, the algorithm calculates the change of R relative to the P_(pitch_begin) position based on the pitch angle at P_tpitch, thereby calculating T_tpitch at the P_tpitch position.

[0137] like Figure 5 As shown, an embodiment of the present invention provides an image calibration device, which includes:

[0138] The shooting module 501 is used to sequentially adjust the various attitude angles of the target camera to shoot the target object according to the preset attitude angle sequence.

[0139] The first acquisition module 502 is used to acquire video data of the target camera adjusting from the initial attitude angle to the desired attitude angle for each attitude angle. Each attitude angle corresponds to one set of video data, and the desired attitude angle is determined based on the desired image position of the calibration object in the video data.

[0140] The parsing and processing module 503 is used to parse and process the video data of each of the attitude angles to obtain the calibration extrinsic parameter matrix of the target camera;

[0141] The second acquisition module 504 is used to acquire the image to be calibrated from the target camera;

[0142] The calibration module 505 is used to calibrate the target object in the image to be calibrated using the calibration extrinsic parameter matrix.

[0143] Optionally, the first acquisition module 502 is further configured to adjust the target camera from a first initial attitude angle to a first desired attitude angle to obtain first video data; fix the first attitude angle and adjust the target camera from a second initial attitude angle to a second desired attitude angle to obtain second video data; fix the first attitude angle and the second attitude angle and adjust the target camera from a third initial attitude angle to a third desired attitude angle to obtain third video data.

[0144] Optionally, the parsing and processing module 503 is further configured to parse and process the first video data to obtain a first extrinsic parameter matrix corresponding to a first attitude angle; parse and process the second video data to obtain a second extrinsic parameter matrix corresponding to a second attitude angle; parse and process the third video data to obtain a third extrinsic parameter matrix corresponding to a third attitude angle; and obtain the calibration extrinsic parameter matrix of the target camera based on the first extrinsic parameter matrix, the second extrinsic parameter matrix, and the third extrinsic parameter matrix.

[0145] Optionally, the parsing processing module 503 is further configured to parse and process the first video data to obtain multiple first attitude angle image data; based on the multiple first attitude angle image data, determine the projection relationship between the calibration board coordinate system and the camera's phase plane coordinate system; based on the projection relationship, calculate the first attitude angle extrinsic parameter matrix of the current image frame; and based on the first attitude angle extrinsic parameter matrix of the current image frame, obtain the first extrinsic parameter matrix corresponding to the first attitude angle.

[0146] Optionally, the parsing and processing module 503 is further configured to parse and process the second video data to obtain multiple second attitude angle image data; determine a second attitude angle list based on the second attitude angle image data; and obtain the second extrinsic parameter matrix corresponding to the second attitude angle based on the second attitude angle list and the equation of the fitted circle.

[0147] Optionally, the parsing processing module 503 is further configured to calculate the change in radius of the circle of the second initial attitude angle with respect to the second desired attitude angle based on the second attitude angle list and the equation of the fitted circle; and to obtain the second extrinsic parameter matrix corresponding to the second attitude angle based on the change in radius of the circle of the second initial attitude angle with respect to the second desired attitude angle.

[0148] Optionally, the parsing and processing module 503 is further configured to parse and process the third video data to obtain multiple third attitude angle image data; determine a third attitude angle list based on the third attitude angle image data; and obtain the third extrinsic parameter matrix corresponding to the third attitude angle based on the third attitude angle list and the equation of the fitted circle.

[0149] Optionally, the analysis processing module 503 is further configured to calculate the change of the third initial attitude angle with respect to the third desired attitude angle based on the third attitude angle list and the equation of the fitted circle; and to obtain the third extrinsic parameter matrix corresponding to the third attitude angle based on the change of the third initial attitude angle with respect to the third desired attitude angle.

[0150] like Figure 6 As shown, this embodiment of the invention also provides an electronic device, including a processor, which can execute any of the above-described image calibration methods.

[0151] Specifically, it includes a processor 601 and a memory 602, as well as a computer program stored in the memory 602 and capable of running on the processor 601 to execute the image calibration method, wherein:

[0152] The processor 601 executes the calculator program for the image calibration method stored in the memory 602, performing the following steps:

[0153] According to the preset attitude angle sequence, adjust the attitude angles of the target camera to capture the target object in sequence;

[0154] For each attitude angle, video data of the target camera adjusting from the initial attitude angle to the desired attitude angle is acquired. Each attitude angle corresponds to one set of video data. The desired attitude angle is determined based on the desired image position of the calibration object in the video data.

[0155] The video data of each of the aforementioned attitude angles are analyzed and processed to obtain the calibration extrinsic parameter matrix of the target camera;

[0156] The target object in the image to be calibrated is calibrated using the calibration extrinsic parameter matrix.

[0157] Optionally, the acquisition of video data from the target camera adjusted from the initial attitude angle to the desired attitude angle, performed by the processor 601, includes:

[0158] The target camera is adjusted from the first initial attitude angle to the first desired attitude angle to obtain the first video data;

[0159] Fix the first attitude angle, and adjust the target camera from the second initial attitude angle to the second desired attitude angle to obtain the second video data;

[0160] The first attitude angle and the second attitude angle are fixed, and the target camera is adjusted from the third initial attitude angle to the third desired attitude angle to obtain the third video data.

[0161] Optionally, the processor 601 performs the parsing process on the video data of each of the attitude angles to obtain the calibration extrinsic parameter matrix of the target camera, including:

[0162] The first video data is parsed and processed to obtain the first extrinsic parameter matrix corresponding to the first attitude angle;

[0163] The second video data is parsed and processed to obtain the second extrinsic parameter matrix corresponding to the second attitude angle;

[0164] The third video data is parsed to obtain the third extrinsic parameter matrix corresponding to the third attitude angle;

[0165] The calibration extrinsic matrix of the target camera is obtained based on the first extrinsic matrix, the second extrinsic matrix, and the third extrinsic matrix.

[0166] Optionally, the process of parsing the first video data performed by the processor 601 to obtain the first extrinsic parameter matrix corresponding to the first attitude angle includes:

[0167] The first video data is parsed and processed to obtain multiple first attitude angle image data;

[0168] Based on multiple first attitude angle image data, the projection relationship between the calibration board coordinate system and the camera's phase plane coordinate system is determined;

[0169] Based on the aforementioned deployment relationship, the first attitude angle extrinsic parameter matrix of the current image frame is calculated;

[0170] Based on the first attitude angle extrinsic parameter matrix of the current image frame, the first extrinsic parameter matrix corresponding to the first attitude angle is obtained.

[0171] Optionally, the parsing process performed by the processor 601 to obtain the second extrinsic parameter matrix corresponding to the second attitude angle includes:

[0172] The second video data is parsed and processed to obtain multiple second attitude angle image data;

[0173] Based on the second attitude angle image data, a second attitude angle list is determined;

[0174] Based on the second attitude angle list and the equation of the fitted circle, the second extrinsic parameter matrix corresponding to the second attitude angle is obtained.

[0175] Optionally, the processor 601 executes the equation based on the second attitude angle list and the fitted circle to determine the second extrinsic parameter matrix corresponding to the second attitude angle, including:

[0176] Based on the second attitude angle list and the equation of the fitted circle, the change in the radius of the circle of the second initial attitude angle with respect to the second desired attitude angle is calculated;

[0177] Based on the change in radius of the circle corresponding to the second initial attitude angle with respect to the second desired attitude angle, the second extrinsic parameter matrix corresponding to the second attitude angle is obtained.

[0178] Optionally, the process of parsing the third video data performed by the processor 601 to obtain the third extrinsic parameter matrix corresponding to the third attitude angle includes:

[0179] The third video data is parsed and processed to obtain multiple third attitude angle image data;

[0180] Based on the third attitude angle image data, a list of third attitude angles is determined;

[0181] Based on the list of third attitude angles and the equation of the fitted circle, the third extrinsic parameter matrix corresponding to the third attitude angle is obtained.

[0182] Optionally, the processor 601 executes the equation based on the third attitude angle list and the fitted circle to determine the third extrinsic parameter matrix corresponding to the third attitude angle, including:

[0183] Based on the third attitude angle list and the equation of the fitted circle, the change of the third initial attitude angle with respect to the third desired attitude angle is calculated;

[0184] Based on the change of the third initial attitude angle with respect to the third desired attitude angle, the third extrinsic parameter matrix corresponding to the third attitude angle is obtained.

[0185] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the image calibration method provided in this invention and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0186] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0187] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. An image calibration method characterized by, The method comprises the following steps: sequentially adjusting each pose angle of the target camera to the calibration object according to a preset pose angle sequence; for each pose angle, obtaining video data of the target camera from an initial pose angle to a desired pose angle, each pose angle corresponding to one of the video data, and the desired pose angle being determined according to a desired image position of the calibration object in the video data; analyzing the video data of each pose angle to obtain an external parameter matrix of the target camera; obtaining a to-be-calibrated image of the target camera; calibrating a target object in the to-be-calibrated image through the external parameter matrix.

2. The image calibration method of claim 1, wherein, The method comprises the following steps: adjusting the target camera from a first initial pose angle to a first desired pose angle to obtain first video data; fixing the first pose angle and adjusting the target camera from a second initial pose angle to a second desired pose angle to obtain second video data; fixing the first pose angle and the second pose angle and adjusting the target camera from a third initial pose angle to a third desired pose angle to obtain third video data.

3. The image calibration method of claim 2, wherein, The method comprises the following steps: analyzing the first video data to obtain a first external parameter matrix corresponding to the first pose angle; analyzing the second video data to obtain a second external parameter matrix corresponding to the second pose angle; analyzing the third video data to obtain a third external parameter matrix corresponding to the third pose angle; obtaining the external parameter matrix of the target camera based on the first external parameter matrix, the second external parameter matrix and the third external parameter matrix.

4. The image calibration method of claim 3, wherein, The method comprises the following steps: analyzing the first video data to obtain a plurality of first pose angle image data; determining a projection relationship between a calibration board coordinate system and a camera plane coordinate system based on the plurality of first pose angle image data; calculating a first pose angle external parameter matrix of a current image frame based on the projection relationship; obtaining the first external parameter matrix corresponding to the first pose angle based on the first pose angle external parameter matrix of the current image frame.

5. The image calibration method of claim 3, wherein, The method comprises the following steps: analyzing the second video data to obtain a plurality of second pose angle image data; determining a second pose angle list based on the second pose angle image data; obtaining the second external parameter matrix corresponding to the second pose angle based on the second pose angle list and an equation of a fitted circle.

6. The image calibration method of claim 5, wherein, The method comprises the following steps: calculating a change of a radius of the circle of the second initial pose angle with respect to the second desired pose angle based on the second pose angle list and the equation of the fitted circle; Based on the change of the second initial attitude angle to the radius of the circle of the second expected attitude angle, a second external parameter matrix corresponding to a second attitude angle is obtained.

7. The image calibration method of claim 3, wherein, The analyzing and processing of the third video data to obtain a third external parameter matrix corresponding to a third attitude angle comprises: The third video data is analyzed and processed to obtain a plurality of third attitude angle image data; Based on the third attitude angle image data, a third attitude angle list is determined; Based on the third attitude angle list and the equation of the fitted circle, a third external parameter matrix corresponding to a third attitude angle is obtained.

8. The image calibration method of claim 7, wherein, The third external parameter matrix corresponding to the third attitude angle is determined based on the third attitude angle list and the equation of the fitted circle, comprising: Based on the third attitude angle list and the equation of the fitted circle, the change of the third initial attitude angle to the third expected attitude angle is calculated; Based on the change of the third initial attitude angle to the third expected attitude angle, a third external parameter matrix corresponding to the third attitude angle is obtained.

9. An image calibration apparatus characterized by comprising: The image calibration device comprises: A shooting module is configured to sequentially adjust each attitude angle of a target camera to shoot a calibration object according to a preset attitude angle sequence. A first acquisition module is configured to acquire, for each attitude angle, video data of the target camera adjusted from an initial attitude angle to an expected attitude angle, each attitude angle corresponding to one of the video data, and the expected attitude angle being determined according to an expected image position of the calibration object in the video data. An analysis and processing module is configured to analyze and process the video data of each attitude angle to obtain a calibration external parameter matrix of the target camera. A second acquisition module is configured to acquire a to-be-calibrated image of the target camera. A calibration module is configured to calibrate a target object in the to-be-calibrated image by using the calibration external parameter matrix.

10. An electronic device, comprising: It comprises: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the image calibration method according to any one of claims 1 to 8.

11. A computer readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium, and when executed by the processor, the computer program implements the steps in the image calibration method according to any one of claims 1 to 8.

Images