A single-station-based rotation axis calibration method, device, system and storage medium

Abstract

The application relates to the technical field of calibration, and discloses a single-station-based rotating shaft calibration method, which comprises the following steps: obtaining a local initial point cloud by rotating scanning a house to be measured by a three-dimensional scanner, and calculating the transformation relationship between the local initial point cloud and a rotating shaft coordinate system to obtain the relative initial external parameter of the three-dimensional scanner and the rotating shaft; extracting a plane feature of the local initial point cloud in the rotating shaft coordinate system to obtain a plane point cloud and an index associated with the plane point cloud, and processing the plane point cloud based on the relative initial external parameter and the index associated with the plane point cloud to obtain a global local point cloud in a global coordinate system; constructing a target optimization function by using the index associated with the plane point cloud and the global local point cloud in the global coordinate system, and iteratively optimizing the relative initial external parameter by using the target optimization function to obtain a relative external parameter. The method provided by the application not only ensures the accuracy of the relative external parameter, but also reduces the time cost and improves the house measurement efficiency.

Description

Technical Field

[0001] This invention relates to the field of calibration technology, and in particular to a single-site-based shaft calibration method, apparatus, system, and storage medium. Background Technology

[0002] A 3D scanner is a scanning device that obtains point cloud data from a scene. Point cloud data can be used to represent the 3D shape of a scene and plays an important role in fields such as building measurement and model building. Existing methods for building measurement using 3D scanners mostly rely on single-site acquisition technology using high-precision automated turntables with feedback measurement, such as TLS (Terrestrial Laser Scanning) from Leica and Faro. This method stitches together the scanned point clouds to obtain a global point cloud by detecting the rotation angle of the automated turntable with each rotation. In other words, each time point cloud stitching is performed, the automated turntable with feedback measurement needs to provide feedback on its current rotation angle. This means that for every point cloud data obtained by the 3D scanner, the rotation angle of the automated turntable needs to be detected and used accordingly, resulting in low efficiency in building measurement. Summary of the Invention

[0003] This invention provides a single-site shaft calibration method, device, system, and storage medium, the main purpose of which is to solve the problem of low efficiency in building measurement.

[0004] To achieve the above objectives, this application provides a single-site axis calibration method applied to an axis calibration system. The axis calibration system includes an automatic turntable, a support, and a processor, used to acquire the relative extrinsic parameters of a 3D scanner mounted on the support and with the support as the axis of rotation. The method includes: controlling the automatic turntable to rotate the 3D scanner attached to the automatic turntable support, so as to obtain local initial point clouds corresponding to different rotation angles by scanning the test building with the 3D scanner at different rotation angles; calculating the transformation relationship between multiple frames of local initial point clouds in the 3D scanner coordinate system and the axis of rotation coordinate system to obtain the relative initial extrinsic parameters of the 3D scanner and the axis of rotation and the local initial point cloud in the axis of rotation coordinate system; extracting planar features from the local initial point cloud in the axis of rotation coordinate system to obtain planar point clouds and indices associated with the planar point clouds, and processing the planar point clouds based on the relative initial extrinsic parameters and the indices associated with the planar point clouds to obtain global and local point clouds in the global coordinate system; constructing a target optimization function using the indices associated with the planar point clouds and the global and local point clouds in the global coordinate system, and iteratively optimizing the relative initial extrinsic parameters through the target optimization function to obtain the relative extrinsic parameters.

[0005] To address the aforementioned issues, this application also provides a single-site axis calibration device, comprising: a local point cloud acquisition module for controlling an automatic turntable to rotate a 3D scanner attached to an automatic turntable support, thereby obtaining local initial point clouds corresponding to different rotation angles by scanning the building under test with the 3D scanner at different rotation angles; an initial extrinsic parameter calculation module for calculating the transformation relationship between multiple frames of local initial point clouds in the 3D scanner coordinate system and the axis coordinate system, obtaining the relative initial extrinsic parameters of the 3D scanner and the axis, and the local initial point cloud in the axis coordinate system; a local point cloud transformation module for extracting planar features from the local initial point cloud in the axis coordinate system to obtain a planar point cloud and an index associated with the planar point cloud, and processing the planar point cloud based on the relative initial extrinsic parameters and the index associated with the planar point cloud to obtain a global local point cloud in the global coordinate system; and an initial extrinsic parameter optimization module for constructing a target optimization function using the index associated with the planar point cloud and the global local point cloud in the global coordinate system, and iteratively optimizing the relative initial extrinsic parameters using the target optimization function to obtain the relative extrinsic parameters.

[0006] To address the aforementioned issues, this application also provides a single-site-based axis calibration system, comprising an automatic turntable, a support, and a processor, for acquiring the relative extrinsic parameters of the 3D scanner mounted on the support and using the support as the axis of rotation. The automatic turntable includes a turntable motor base and a turntable tray, the turntable motor base driving the turntable tray to rotate. The support is mounted at a position corresponding to the rotation center of the turntable tray, rotating with the turntable tray and forming the axis of rotation for the axis calibration system. The processor processes multiple frames of initial local point clouds obtained by the 3D scanner synchronously scanning the local scene of the building under test as the support rotates, according to the aforementioned single-site-based axis calibration method, to obtain the relative extrinsic parameters of the 3D scanner and the axis of rotation.

[0007] To address the aforementioned issues, this application also provides a single-site-based 3D scanning system, comprising: an automatic turntable, a support, a 3D scanner, and a processor, wherein: the automatic turntable includes a turntable motor base and a turntable tray, the turntable motor base being used to drive the turntable tray to rotate; the support is installed at a position corresponding to the rotation center of the turntable tray to follow the rotation of the turntable tray and form the rotation axis of the 3D scanning system; the 3D scanner is attached to the support and synchronously performs local scene scanning of the house under test with the rotation of the support to obtain multiple frames of local initial point clouds; the processor is used to stitch the multiple frames of local initial point clouds to obtain the global point cloud of the house under test based on the relative extrinsic parameters of the 3D scanner and the rotation axis obtained by the above-mentioned single-site-based rotation axis calibration method.

[0008] To address the aforementioned issues, this application also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor to implement the aforementioned single-site-based shaft calibration method.

[0009] This application continuously optimizes the relative extrinsic parameters by constructing an objective optimization function. This ensures that planar point clouds extracted from local point clouds acquired from different angles can be stitched together using these relative extrinsic parameters, resulting in a high-precision, complete point cloud of the building under test. Because the true value of the relative extrinsic parameters is unique after the scanning system is completed, the relative extrinsic parameters obtained in this application have generalization capabilities and can be used for point cloud stitching from other angles. This solves the problem of low applicability of relative extrinsic parameters and avoids the need to obtain different relative extrinsic parameters for different rotation angles when acquiring point clouds at a single site, or the need to detect the rotation angle of the automatic turntable for each point cloud data acquired by the 3D scanner. The single-site rotation axis calibration method provided in this application not only ensures the accuracy of the relative extrinsic parameters but also eliminates the need to detect the rotation angle of the automatic turntable, reducing time costs and improving the efficiency of building measurement. Attached Figure Description

[0010] Figure 1 A system architecture diagram of a single-site-based shaft calibration system provided in an embodiment of the present invention;

[0011] Figure 2 This is a flowchart illustrating a single-site-based shaft calibration method according to an embodiment of the present invention.

[0012] Figure 3 This is a functional block diagram of a single-site shaft calibration device provided in an embodiment of the present invention.

[0013] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0015] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0016] Figure 1 This is a schematic diagram of the architecture of a single-site axis calibration system provided in an embodiment of the present invention. The system includes an automatic turntable 101, a support 102, a 3D scanner 103, and a processor 104, used to acquire the relative extrinsic parameters of the axis and the 3D scanner 103 mounted on the support 102. The automatic turntable 101 includes a turntable motor base 1011 and a turntable tray 1012, the turntable motor base 1011 being used to drive the turntable tray 1012 to rotate. The support 102 is mounted at a position corresponding to the rotation center of the turntable tray 1012, so as to follow the rotation of the turntable tray 1012 and form the axis of the axis calibration system. The processor 104 is used to process multiple frames of local initial point cloud obtained by the 3D scanner 103 synchronously scanning the local scene of the house under test as the support 102 rotates, according to the single-site axis calibration method provided in one or more embodiments of this application, to obtain the relative extrinsic parameters of the 3D scanner 103 and the axis.

[0017] In one embodiment, the 3D scanner 103 can be a laser scanner, which scans the building under test to obtain local point clouds from a specific perspective. The set of local point clouds obtained from each scanning point is the local point cloud set. In another embodiment, the 3D scanner 103 can also be a depth camera, which scans the building under test to obtain local depth images from a specific perspective. The local depth images obtained from each scanning point are converted into local point clouds, and the set of local point clouds obtained from each scanning point is the local point cloud set. It should be noted that when the 3D scanner 103 is a depth camera, the depth camera can be a structured light camera, a time-of-flight camera, a binocular camera, or a multi-view camera, etc. In addition, it should be noted that this application allows for one or more depth cameras or laser scanners installed on the same side of the support, and no limitation is made here.

[0018] In some embodiments, the processor 104 is composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 104 is the control unit of the spindle calibration system, connecting various components of the entire spindle calibration system via various interfaces and lines. It performs various functions of the electronic device and processes data by running or executing programs or modules stored in memory (e.g., executing a single-site spindle calibration method program) and calling data stored in memory.

[0019] The figure only shows the components of the system. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the system and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0020] For example, although not shown, the system may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to at least one processor via a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power sources, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The system may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be elaborated here. It should be understood that the embodiments are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0021] Figure 2 This is a flowchart illustrating a single-site-based shaft calibration method according to an embodiment of the present invention. The method is applied to... Figure 1 The axis calibration system shown is used to acquire the relative extrinsic parameters of a 3D scanner mounted on a support, with the support as the axis of rotation. Figure 2 The methods shown include:

[0022] S21. Control the automatic turntable to drive the 3D scanner attached to the automatic turntable support to rotate, so as to obtain the local initial point cloud corresponding to different rotation angles by scanning the house under test by the 3D scanner at different rotation angles.

[0023] Specifically, the automatic turntable is controlled to rotate around a support frame (i.e., the center of rotation) by a preset angle, driving the 3D scanner on the automatic turntable to rotate as well. When the automatic turntable reaches different rotation angles (e.g., 10°, 20°, ..., 350°), the automatic turntable stops, and simultaneously the 3D scanner with a certain field of view scans a local scene of the building under test, obtaining a local initial point cloud frame. Furthermore, when the automatic turntable completes one full rotation (360°), it is considered to have completed one global scan of the building under test, obtaining multiple local initial point clouds corresponding to different rotation angles.

[0024] In one embodiment, when the 3D scanner is a depth camera, controlling the depth camera on the automatic turntable to scan the house under test at different rotation angles to obtain the local initial point cloud corresponding to each rotation angle further includes: controlling the 3D scanner on the automatic turntable to perform a global scan of the house under test at different rotation angles to obtain multiple frames of local depth images, and converting the multiple frames of local depth images into point clouds based on the depth camera intrinsic parameters to obtain the local initial point clouds corresponding to each rotation angle.

[0025] Preferably, the intrinsic parameters of the depth camera are represented in matrix form as follows:

[0026]

[0027] Where f represents the focal length of the depth camera, (C x C y ) represents the optical center of the depth camera.

[0028] Converting multiple frames of local depth images into point clouds based on depth camera intrinsic parameters to obtain initial local point clouds corresponding to different rotation angles includes: converting local depth images into point clouds by combining depth camera intrinsic parameters and depth point cloud conversion formulas to obtain initial local point clouds; wherein, the depth point cloud conversion formulas are as follows:

[0029]

[0030] Where (X, Y, Z) represent the coordinates of the initial local point cloud in the depth camera coordinate system, (x1, y1) represent the pixel coordinates of the local depth image, and z represents the depth value corresponding to the pixel in the local depth image. In this embodiment of the invention, by using a depth transformation algorithm to convert the target image into a camera point cloud data set, subsequent point cloud data stitching can be facilitated, thereby achieving the acquisition of the global point cloud of the building under test.

[0031] S22. Calculate the transformation relationship between the local initial point cloud of multiple frames in the coordinate system of the 3D scanner and the rotation axis coordinate system to obtain the relative initial extrinsic parameters of the 3D scanner and the rotation axis and the local initial point cloud in the rotation axis coordinate system.

[0032] In one embodiment, multiple frames of local initial point clouds are transformed from the 3D scanner coordinate system to the rotation axis coordinate system, thereby calculating the transformation relationship between the multiple frames of local initial point clouds in the 3D scanner coordinate system and the rotation axis coordinate system, and obtaining the relative initial extrinsic parameters between the 3D scanner and the rotation axis. Specifically, research shows that the relative extrinsic parameters between the 3D scanner and the rotation axis describe the transformation relationship between the 3D scanner coordinate system and the rotation axis coordinate system, which is independent of the rotation angle in step S21. Based on this, the relative extrinsic parameters between the 3D scanner and the rotation axis can be represented by four degrees of freedom, namely r x r y x2, y2, are represented by rotation matrix R and translation matrix T, which together represent the initial relative extrinsic parameters between the 3D scanner and the axis of rotation.

[0033] p′=R*p+t

[0034] R = exp((0, r y ,0) ∧ )exp((r x ,0,0) ∧ )

[0035] T = (x2, y2, 0)

[0036] Where p represents the point cloud coordinates in the 3D scanner coordinate system, p′ represents the point cloud coordinates in the rotation axis coordinate system, and exp(·) represents the Rodriguez rotation formula, (r x ,r y ,r z ) ∧ Defined as r y r represents the rotation along the y-axis. x x1 represents the rotation along the x-axis, x2 represents the translation along the x-axis, and y2 represents the translation along the y-axis. It should be noted that the relative extrinsic parameters in this application represent the transformation relationship between the 3D scanner coordinate system and the rotation axis coordinate system. Determining the relative extrinsic parameters requires relative parameters with 6 degrees of freedom. Since these are independent of the rotation angle in step S21, the two degrees of freedom parameters related to this rotation angle can have any value. For ease of calculation, this application sets the rotation-related degrees of freedom parameters to zero; however, this is not a restriction.

[0037] S23. Perform planar feature extraction on the local initial point cloud in the rotation axis coordinate system to obtain the planar point cloud and the index associated with the planar point cloud. Then, process the planar point cloud based on the relative initial extrinsic parameters and the index associated with the planar point cloud to obtain the global local point cloud in the global coordinate system.

[0038] In one embodiment, extracting planar features from the local initial point cloud in the rotation coordinate system to obtain a planar point cloud and associated indices includes: extracting planar features from each frame of the local initial point cloud in the rotation coordinate system using methods such as Random Sample Consensus (RANSAC), region growing, or deep learning to obtain the planar point cloud of the building under test, such as wall point clouds, ground point clouds, and ceiling point clouds; and associating each frame of the local initial point cloud obtained based on different rotation angles with the angle index and planar index of each point in the planar point cloud obtained based on each frame of the local initial point cloud. It should be noted that during the extraction of the planar point cloud, filtering can also be performed on the point cloud to remove noise, thereby obtaining a more accurate planar point cloud.

[0039] In one embodiment, associating the local initial point clouds of each frame acquired based on different rotation angles with the planar point clouds obtained based on the local initial point clouds of each frame specifically includes: extracting planar point clouds based on the local initial point clouds of each frame acquired by the 3D scanner at different rotation angles as the automatic turntable rotates; associating the local initial point clouds of each frame at different rotation angles with the planar point clouds extracted based on the local initial point clouds of each frame; and assigning each point in the planar point cloud a corresponding angle index and a planar index; wherein, the angle index represents the rotation angle corresponding to the current point, such as the current point being acquired based on a rotation angle of 45°; and the planar index represents the plane corresponding to the current point, such as the current point being a wall or the ground.

[0040] In one embodiment, processing the planar point cloud based on relative initial extrinsic parameters and the index associated with the plane to obtain a global-local point cloud in the global coordinate system includes: converting the planar point cloud in the rotation-axis coordinate system into a planar point cloud in the global coordinate system using a global rotation-axis transformation formula, based on the relative initial extrinsic parameters and the angle index associated with the plane. Specifically, the global rotation-axis transformation formula is as follows:

[0041] p″=exp((θ*ω) ∧ )p′

[0042] Where p″ is the point cloud data in the global point cloud coordinate system, exp(·) refers to the Rodriguez rotation formula, θ is the known rotation angle of the automatic turntable around the rotation axis, and ω refers to the rotation axis direction in the rotation axis coordinate system. θ*ω can be expressed as (ω x ,ω y ,ω z If (θ*ω) is true, then (θ*ω) is true. ∧ for It should be noted that the point cloud data in the rotation axis coordinate system is converted to the global coordinate system so that the point cloud can be stitched together based on the same coordinate system to obtain the global point cloud of the house to be measured. This allows for the application of loop constraints on the relative extrinsic parameters between the 3D scanner and the rotation axis based on the point cloud stitching.

[0043] S24. Construct a target optimization function using the index associated with the planar point cloud and the global and local point clouds in the global coordinate system. Iterate and optimize the relative initial extrinsic parameters through the target optimization function to obtain the relative extrinsic parameters.

[0044] In one embodiment, a target optimization function is constructed using the planar index associated with the planar point cloud and the global and local point clouds in the global coordinate system. Points obtained from different rotation angles are linked together through the planar index associated with the planar point cloud to perform planar constraints. The optimal relative extrinsic parameters are then obtained through iterative optimization based on the target optimization function, using the relative initial extrinsic parameters. Preferably, the specific form of the target optimization function is:

[0045]

[0046] Where arg min represents the minimum function of the independent variable, represents the value of the independent variable when the function reaches its minimum, R1 represents the rotation matrix relative to the extrinsic parameters, t1 represents the translation matrix relative to the extrinsic parameters, i represents the plane index corresponding to the current point, j represents the j-th point in the global local point cloud, and w i n represents the weight of the plane corresponding to the current point. i p″ represents the normal vector of the plane corresponding to the current point. j c represents the coordinates of the j-th point in the global and local point clouds. i This indicates the center point of the plane corresponding to the current point.

[0047] In one embodiment, iteratively optimizing the relative extrinsic parameters using a target optimization function to obtain the relative extrinsic parameters includes: comparing the loss value obtained based on the target optimization function with a preset threshold; if the loss value is greater than or equal to the preset threshold, then using the optimized relative extrinsic parameters as the initial value, repeatedly obtaining the local point cloud in the rotation coordinate system based on the relative extrinsic parameters and steps S23-S24, continuously iterating and optimizing the relative extrinsic parameters until the loss value obtained based on the target optimization function is less than the preset threshold. When the loss value obtained based on the target optimization function is less than the preset threshold, the relative extrinsic parameters obtained at this time are considered the true values ​​of the extrinsic parameters, i.e., the optimal relative extrinsic parameters.

[0048] This embodiment continuously optimizes the relative extrinsic parameters by constructing an objective optimization function. This results in a high degree of flatness when the planar point clouds extracted from local point clouds acquired from different angles are stitched together using these relative extrinsic parameters, ultimately leading to a high-precision complete point cloud of the building under test. It should be noted that once the system is finalized, its extrinsic parameter true values ​​are unique. The relative extrinsic parameters obtained in this application have generalization capabilities and can be used for point cloud stitching from other angles, solving the problem of low applicability of relative extrinsic parameters. This avoids the need to obtain different relative extrinsic parameters for different rotation angles when acquiring point clouds at a single site, or the need to detect the rotation angle of the automatic turntable for each point cloud data scanned by the 3D scanner. The single-site rotation axis calibration method provided in this application not only ensures the accuracy of the relative extrinsic parameters but also eliminates the need to detect the rotation angle of the automatic turntable, reducing time costs and improving the efficiency of building measurement.

[0049] Figure 3 This is a functional block diagram of the single-site-based axis calibration device provided in this application. The device is applied to an axis calibration system and includes a local point cloud acquisition module 301, an initial extrinsic parameter calculation module 302, a local point cloud conversion module 303, and an initial extrinsic parameter optimization module 304. The local point cloud acquisition module 301 controls an automatic turntable to rotate a 3D scanner attached to the automatic turntable support, thereby obtaining local initial point clouds corresponding to different rotation angles through a global scan of the building under test by the 3D scanner at different rotation angles. The initial extrinsic parameter calculation module 302 calculates the local initial point clouds of multiple frames relative to the axis in the 3D scanner coordinate system. The transformation relationship between coordinate systems is used to obtain the relative initial extrinsic parameters between the 3D scanner and the rotation axis, as well as the local initial point cloud in the rotation axis coordinate system. The local point cloud transformation module 303 is used to extract planar features from the local initial point cloud in the rotation axis coordinate system to obtain a planar point cloud and an index associated with the planar point cloud. Based on the relative initial extrinsic parameters and the index associated with the planar point cloud, the planar point cloud is processed to obtain a global and local point cloud in the global coordinate system. The initial extrinsic parameter optimization module 304 is used to construct a target optimization function using the index associated with the planar point cloud and the global and local point cloud in the global coordinate system. The relative extrinsic parameters are obtained by iteratively optimizing the relative initial extrinsic parameters through the target optimization function.

[0050] It should be noted that the module of this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by a processor in a system and can perform a fixed function. Furthermore, the modules in the single-site shaft calibration device 300 in this embodiment of the application employ the same methods as described above. Figure 2 The method is based on the same techniques as the single-site shaft calibration method and can produce the same technical effect, so it will not be elaborated here.

[0051] Furthermore, if the integrated modules / units of the system are implemented as software functional units and sold or used as independent products, they can be stored in a storage medium. The storage medium can be volatile or non-volatile. For example, computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, and read-only memory (ROM).

[0052] The present invention also provides a storage medium, which stores a computer program. When the computer program is executed by the system's processor, it can implement one or more embodiments of the single-site-based shaft calibration method provided in this application.

[0053] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0054] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0055] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0056] This application also provides a single-site-based 3D scanning system, which includes an automatic turntable, a support, a 3D scanner, and a processor. The automatic turntable includes a turntable motor base and a turntable tray, the turntable motor base driving the turntable tray to rotate. The support is installed at a position corresponding to the rotation center of the turntable tray to follow the turntable tray's rotation and form the rotation axis of the 3D scanning system. The 3D scanner is attached to the support and synchronously scans the local scene of the house under test with the rotation of the support to obtain multiple frames of local initial point clouds. The processor is used to stitch together the multiple frames of local initial point clouds to obtain the global point cloud of the house under test based on the relative extrinsic parameters of the 3D scanner and the rotation axis obtained by the single-site-based rotation axis calibration method provided in one or more embodiments of this application.

[0057] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0058] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0059] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A single-site-based shaft calibration method, applied to a shaft calibration system, the shaft calibration system comprising an automatic turntable, a support, and a processor, characterized in that, The method for obtaining relative extrinsic parameters of a 3D scanner mounted on the bracket and with the bracket as a rotation axis includes: The automatic turntable is controlled to drive the 3D scanner attached to the automatic turntable support to rotate, so as to obtain the local initial point cloud corresponding to different rotation angles by scanning the house under test by the 3D scanner at different rotation angles. Calculate the transformation relationship between the local initial point cloud of multiple frames in the coordinate system of the 3D scanner and the rotation axis coordinate system to obtain the relative initial extrinsic parameters of the 3D scanner and the rotation axis and the local initial point cloud in the rotation axis coordinate system; Planar features are extracted from the local initial point cloud in the rotation axis coordinate system to obtain a planar point cloud and an index associated with the planar point cloud. Based on the relative initial extrinsic parameters and the index associated with the planar point cloud, the planar point cloud is processed to obtain a global local point cloud in the global coordinate system. A target optimization function is constructed using the index associated with the planar point cloud and the global and local point clouds in the global coordinate system. The relative extrinsic parameters are obtained by iteratively optimizing the relative initial extrinsic parameters using the target optimization function.

2. The single-site-based shaft calibration method as described in claim 1, characterized in that, When the 3D scanner is a depth camera, the local initial point cloud corresponding to different rotation angles is obtained by scanning the building under test with the 3D scanner at different rotation angles, including: The depth camera on the automatic turntable is controlled to scan the building under test at different rotation angles to obtain multiple frames of local depth images. These local depth images are then converted into point clouds using the depth camera's intrinsic parameters and a depth point cloud conversion formula, resulting in an initial local point cloud. The depth point cloud conversion formula is as follows: in,( , , () represents the coordinates of the local initial point cloud in the depth camera coordinate system. This represents the pixel coordinates of the local depth image. This represents the depth value corresponding to a pixel in the local depth image. This indicates the focal length of the depth camera. This represents the optical center of the depth camera.

3. The single-site-based shaft calibration method as described in claim 1, characterized in that, The relative extrinsic parameters between the 3D scanner and the rotating axis can be represented by four degrees of freedom, namely: , , , The relative initial extrinsic parameters between the 3D scanner and the rotation axis are represented by the rotation matrix R and the translation matrix T, respectively: in, This represents the point cloud coordinates in the 3D scanner coordinate system. This represents the point cloud coordinates in the rotation coordinate system. This represents the Rodriguez rotation formula. Defined as , Axis rotation amount, Axis rotation amount, Axis translation amount, Axis translation amount.

4. The single-site-based shaft calibration method as described in claim 3, characterized in that, The step of extracting planar features from the local initial point cloud in the rotation axis coordinate system to obtain a planar point cloud and an index associated with the planar point cloud includes: The planar features of the local initial point cloud in each frame under the rotation coordinate system are extracted using random consistency, region growing or deep learning methods to obtain the planar point cloud of the house under test; By associating the local initial point clouds of each frame obtained based on the different rotation angles with the local initial point clouds of each frame, the angle index and the plane index of each point in the planar point cloud are obtained; wherein, the angle index represents the rotation angle corresponding to the current point, and the plane index represents the plane corresponding to the current point.

5. The single-site-based shaft calibration method as described in claim 4, characterized in that, The step of processing the planar point cloud based on the relative initial extrinsic parameters and the index associated with the planar point cloud to obtain a global-local point cloud in the global coordinate system includes: converting the planar point cloud in the rotation axis coordinate system into a planar point cloud in the global coordinate system according to the relative initial extrinsic parameters and the angle index associated with the plane using a global rotation axis transformation formula; wherein, the global rotation axis transformation formula includes: in, It is point cloud data in the global point cloud coordinate system. This refers to the Rodriguez rotation formula. It is the known rotation angle of the automatic turntable around its axis. This refers to the direction of rotation in the rotation coordinate system. It can be represented as ,but for .

6. The single-site-based shaft calibration method as described in claim 5, characterized in that, The objective optimization function includes: in, Let represent the minimum function of the independent variable, and let represent the value of the independent variable when the function reaches its minimum value. Represents the rotation matrix of the relative extrinsic parameter. This represents the translation matrix of the relative extrinsic parameter. This represents the plane index corresponding to the current point. This indicates that the current point is the [number]th [point] in the global / local point cloud. point, This indicates the weight of the plane corresponding to the current point. This represents the normal vector of the plane corresponding to the current point. Represents the first point in the global and local point clouds. The coordinates of the point This indicates the center point of the plane corresponding to the current point.

7. A shaft calibration device based on a single station, characterized in that, The device includes: The local point cloud acquisition module is used to control the automatic turntable to drive the 3D scanner attached to the automatic turntable to rotate, so as to obtain the local initial point cloud corresponding to different rotation angles by scanning the house under test by the 3D scanner at different rotation angles. The initial extrinsic parameter calculation module is used to calculate the transformation relationship between the local initial point cloud in multiple frames in the 3D scanner coordinate system and the rotation axis coordinate system, so as to obtain the relative initial extrinsic parameters of the 3D scanner and the rotation axis and the local initial point cloud in the rotation axis coordinate system. The local point cloud conversion module is used to extract planar features from the local initial point cloud in the rotation axis coordinate system to obtain a planar point cloud and an index associated with the planar point cloud, and to process the planar point cloud based on the relative initial extrinsic parameters and the index associated with the planar point cloud to obtain a global local point cloud in the global coordinate system. The initial extrinsic parameter optimization module is used to construct a target optimization function using the index associated with the planar point cloud and the global and local point clouds in the global coordinate system, and to obtain the relative extrinsic parameters by iteratively optimizing the relative initial extrinsic parameters through the target optimization function.

8. A single-site-based shaft calibration system, comprising an automatic turntable, a support, and a processor, characterized in that, Used to obtain the relative extrinsic parameters of the rotating shaft and the 3D scanner mounted on the bracket, wherein: The automatic turntable includes a turntable motor base and a turntable tray, wherein the turntable motor base is used to drive the turntable tray to rotate; The bracket is installed at the corresponding position of the rotation center of the turntable tray so as to rotate with the turntable tray and form the rotation axis of the rotation axis calibration system; The processor is used to process multiple frames of local initial point cloud obtained by the 3D scanner synchronously scanning the local scene of the house under test as the support rotates with the single-site rotation axis calibration method according to any one of claims 1 to 6 to obtain the relative external parameters of the 3D scanner and the rotation axis.

9. A single-site-based 3D scanning system, characterized in that, include: The components include an automated turntable, a support frame, a 3D scanner, and a processor, among which: The automatic turntable includes a turntable motor base and a turntable tray, wherein the turntable motor base is used to drive the turntable tray to rotate; The bracket is installed at the corresponding position of the rotation center of the turntable tray so as to rotate with the turntable tray and form the rotation axis of the three-dimensional scanning system; The 3D scanner is attached to a support and rotates synchronously with the support to perform local scene scanning of the house under test, obtaining multiple frames of local initial point cloud. The processor is used to stitch together the multi-frame local initial point cloud with the relative extrinsic parameters of the 3D scanner and the axis of rotation obtained by the single-site axis calibration method according to any one of claims 1 to 6 to obtain the global point cloud of the house under test.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the single-site-based shaft calibration method as described in any one of claims 1 to 6.

Images