Scanning method and device of three-dimensional scanner, computer equipment and storage medium

Abstract

The invention relates to a scanning method and device of a three-dimensional scanner, computer equipment and a storage medium, and the method comprises the steps: obtaining a three-dimensional point cloud set and a first matrix set of each frame based on the scanning of a scanning object; performing point cloud fusion on the three-dimensional point cloud set based on the first matrix set to obtain a first fusion point cloud; registering the three-dimensional point cloud in each frame of three-dimensional point cloud set with the first fusion point cloud to obtain a second matrix set; and based on the second matrix set, carrying out point cloud fusion on the three-dimensional point cloud set to obtain a target fusion point cloud. According to the method and the device, the problem that the quality of the final scanning data is reduced due to the fact that an RT matrix calculated by a single frame is not accurate enough in the prior art is solved, the accuracy of matrix calculation is improved by means of registration, and point cloud fusion is performed on the three-dimensional point cloud set by replacing the first matrix set with the second matrix set, so that the quality of the scanning data is improved.

Description

Technical Field

[0001] This application relates to the field of three-dimensional scanning technology, and in particular, to a scanning method, device, computer device, and storage medium of a three-dimensional scanner. Background Art

[0002] For a laser three-dimensional scanner, as the scanning progresses, it will continuously stitch the images scanned in each frame; however, as the scanning accuracy of the laser three-dimensional scanner decreases or the scanning distance becomes farther, the RT matrix calculated for a single frame will not be accurate enough, resulting in a decrease in the quality of the final scanned data, that is, situations such as warping and knife marks occur in the scanned data.

[0003] Regarding the problem in the related art that the RT matrix calculated for a single frame is not accurate enough, resulting in a decrease in the quality of the final scanned data, no effective solution has been proposed yet. Summary of the Invention

[0004] In this embodiment, a scanning method, device, computer device, and storage medium of a three-dimensional scanner are provided to solve the problem in the related art that the RT matrix calculated for a single frame is not accurate enough, resulting in a decrease in the quality of the final scanned data.

[0005] In a first aspect, in this embodiment, a scanning method of a three-dimensional scanner is provided, including:

[0006] Based on the scanning of a scanning object, a three-dimensional point cloud set and a first matrix set are obtained; the first matrix set is a set of transformation matrices for transforming each frame of the three-dimensional point cloud set into a preset third-party coordinate system;

[0007] Based on the first matrix set, point cloud fusion is performed on the three-dimensional point cloud set to obtain a first fused point cloud;

[0008] The three-dimensional point cloud in each frame of the three-dimensional point cloud set is registered with the first fused point cloud to obtain a second matrix set; the second matrix set is a set of transformation matrices for adjusting the positions of each frame of the three-dimensional point cloud set;

[0009] Based on the second matrix set, point cloud fusion is performed on the three-dimensional point cloud set to obtain a target fused point cloud.

[0010] In some of these embodiments, based on the first matrix set, performing point cloud fusion on the three-dimensional point cloud set to obtain a first fused point cloud includes:

[0011] Based on the first matrix set, the three-dimensional point cloud set is transformed into the third-party coordinate system; in the third-party coordinate system, point cloud fusion of the three-dimensional point cloud set is performed at a preset resolution to obtain a first fused point cloud;

[0012] Alternatively, downsample the three-dimensional point cloud set to obtain a downsampled point cloud set at a preset resolution;

[0013] Based on the first matrix set, transform the downsampled point cloud set to the third-party coordinate system for point cloud fusion to obtain a first fused point cloud.

[0014] In some embodiments, based on scanning a scanned object, obtaining a three-dimensional point cloud set and a first matrix set for each frame includes:

[0015] Based on scanning the marker points on the scanned object, obtain the global marker point data; at least one marker point is provided on the scanned object;

[0016] Based on scanning the scanned object, obtain the three-dimensional point cloud set, the first matrix set, and the marker point data for each frame; and determine the matching relationship between the marker point data and the global marker point data.

[0017] In some embodiments, registering the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set includes:

[0018] Traverse the three-dimensional point cloud set, the first matrix set, and the marker point data, and register the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set.

[0019] In some embodiments, registering the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set includes:

[0020] Taking the coincidence of the section where each frame of the three-dimensional point cloud set is located and the section where the first fused point cloud is located, and the marker points remaining fixed as constraint conditions, register the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set.

[0021] In some embodiments, registering the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set includes:

[0022] Based on the pre-constructed index of the first fused point cloud, determine the potential matching pairs between the three-dimensional point cloud in the three-dimensional point cloud set and the first fused point cloud;

[0023] Filter out target matching pairs from the potential matching pairs based on preset search conditions; the search conditions are that for search points with the same index, a target matching pair is formed between the first target point with the smallest Euclidean distance and the search point; the search point is a point in the three-dimensional point cloud; the first target point is a point in the first fused point cloud;

[0024] Taking the minimum distance between the search point and the first target point in the target matching pair as a condition, obtain the second matrix set.

[0025] In some embodiments, based on the pre-constructed index of the first fused point cloud, determining potential matching pairs between the three-dimensional point cloud in the three-dimensional point cloud set and the first fused point cloud includes:

[0026] Construct an index for each first target point in the first fused point cloud;

[0027] Determine the nearest point of the three-dimensional points in the three-dimensional point cloud in the first fused point cloud;

[0028] Based on a preset distance threshold, eliminate abnormal points in the three-dimensional points to obtain non-abnormal three-dimensional points and corresponding proportions;

[0029] Based on the non-abnormal three-dimensional points, determine potential matching pairs.

[0030] In some embodiments, the method further includes:

[0031] After obtaining the second matrix set, based on the second matrix set, convert the three-dimensional point cloud in each frame of the three-dimensional point cloud set into a three-dimensional updated point cloud;

[0032] Respectively determine the mean square error of the distances of the three-dimensional points in the three-dimensional updated point cloud and the three-dimensional point cloud on the corresponding tangent plane where the nearest points are located, to obtain a first mean square error and a second mean square error;

[0033] According to the first mean square error and the second mean square error, determine whether the current registration meets the convergence condition; the convergence condition is that the change amplitude between the first matrix set and the second matrix set is less than a preset amplitude threshold;

[0034] If the convergence condition is met, replace the three-dimensional point cloud in the three-dimensional point cloud set with the three-dimensional updated point cloud;

[0035] If the convergence condition is not met, then according to the number of the target matching pairs and the proportion, replace the three-dimensional point cloud in the three-dimensional point cloud set with the three-dimensional updated point cloud; re-register the updated three-dimensional updated point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud.

[0036] In some of these embodiments, the method further includes:

[0037] Based on the second matrix set, convert the first target point to a second target point; and calculate the average values of the distances from the first target point and the second target point to the tangent planes where the corresponding nearest points are located respectively, to obtain a first target average value and a second target average value;

[0038] Compare the first target average value and the second target average value;

[0039] If the first target average value is greater than the second target average value, the second matrix set meets the configuration requirements;

[0040] If the first target average value is less than or equal to the second target average value, re - estimate the second matrix set.

[0041] In some of these embodiments, the method further includes:

[0042] When the scanning object is a spherometer and at least one marking point is set beside the spherometer, determine the sphere diameter of the spherometer based on the target fused point cloud;

[0043] Determine the scanning accuracy based on the sphere diameter of the spherometer.

[0044] In a second aspect, a scanning device of a 3D scanner is provided in this embodiment, including: an acquisition module, a first point cloud fusion module, a registration module, and a second point cloud fusion module;

[0045] The acquisition module is configured to obtain a three - dimensional point cloud set and a first matrix set for each frame based on the scanning of the scanning object; the first matrix set is a set of transformation matrices for transforming each frame of the three - dimensional point cloud set into a preset third - party coordinate system;

[0046] The first point cloud fusion module is configured to perform point cloud fusion on the three - dimensional point cloud set based on the first matrix set to obtain a first fused point cloud;

[0047] The registration module is configured to register the three - dimensional point clouds in each frame of the three - dimensional point cloud set with the first fused point cloud to obtain a second matrix set; the second matrix set is a set of transformation matrices for adjusting the positions of each frame of the three - dimensional point cloud set;

[0048] The second point cloud fusion module is configured to perform point cloud fusion on the three - dimensional point cloud set based on the second matrix set to obtain a target fused point cloud.

[0049] In a third aspect, a computer device is provided in this embodiment, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the scanning method of the 3D scanner described in the first aspect above is implemented.

[0050] In a fourth aspect, a storage medium is provided in this embodiment, on which a computer program is stored. When the program is executed by a processor, the scanning method of the 3D scanner described in the first aspect above is implemented.

[0051] Compared with the related art, in the scanning method, device, computer device, and storage medium of the 3D scanner provided in this embodiment, by scanning the scanning object, a three-dimensional point cloud set and a first matrix set of each frame are obtained; the first matrix set is a set of transformation matrices obtained by transforming each frame of the three-dimensional point cloud set into a preset third-party coordinate system; based on the first matrix set, point cloud fusion is performed on the three-dimensional point cloud set to obtain a first fused point cloud; the three-dimensional points in each frame of the three-dimensional point cloud set are registered with the first fused point cloud to obtain a second matrix set; the second matrix set is a set of transformation matrices for adjusting the positions of each frame of the three-dimensional point cloud set; based on the second matrix set, point cloud fusion is performed on the three-dimensional point cloud set to obtain a target fused point cloud; the problem that the RT matrix calculated for a single frame in the related art is not accurate enough, resulting in a reduction in the quality of the final scanned data, is solved. Registration is used to improve the accuracy of matrix calculation, and the second matrix set is used to replace the first matrix set to perform point cloud fusion on the three-dimensional point cloud set, thereby improving the quality of the scanned data.

[0052] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects, and advantages of this application more concise and understandable. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings described herein are used to provide a further understanding of this application and form a part of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an improper limitation of this application. In the drawings:

[0054] Figure 1 is a hardware structure block diagram of a terminal device for the scanning method of the 3D scanner provided in an embodiment of this application;

[0055] Figure 2 is a flowchart of the scanning method of the 3D scanner provided in an embodiment of this application;

[0056] Figure 3 is a flowchart of step S210;

[0057] Figure 4 is a flowchart of step S230;

[0058] Figure 5 is the flowchart of step S410;

[0059] Figure 6 is the schematic diagram for determining the scanning accuracy provided by an embodiment of the present application;

[0060] Figure 7 is the structural block diagram of the scanning device of the 3D scanner provided by an embodiment of the present application.

[0061] In the figure, 210 is the acquisition module; 220 is the first point cloud fusion module; 230 is the registration module; 240 is the second point cloud fusion module. Detailed implementation manners

[0062] For a clearer understanding of the purpose, technical solution and advantages of the present application, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments.

[0063] Unless otherwise defined, the technical terms or scientific terms involved in the present application shall have the general meanings understood by those with ordinary skills in the technical field to which the present application belongs. In the present application, words such as "a", "an", "one kind", "the", "these", etc. do not indicate a limitation in quantity, and they can be singular or plural. The terms "including", "comprising", "having" and any variations thereof involved in the present application are intended to cover non-exclusive inclusion; for example, a process, method, system, product or device including a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent in these processes, methods, products or devices. The terms "connected", "coupled" and the like involved in the present application do not limit to physical or mechanical connections, but may include electrical connections, whether directly or indirectly. The "multiple" involved in the present application means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" may represent: A exists alone, A and B exist simultaneously, and B exists alone. Usually, the character " / " indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in the present application only distinguish similar objects and do not represent a specific sorting for the objects.

[0064] The method embodiment provided in this embodiment can be executed on a terminal, a computer or a similar computing device. For example, it runs on a terminal. Figure 1 is the hardware structural block diagram of the terminal of the scanning method of the 3D scanner in this embodiment. As Figure 1As shown, the terminal may include one or more ( Figure 1 only one is shown in the figure) processors 102 and a memory 104 for storing data. Among them, the processor 102 may include, but is not limited to, processing devices such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those of ordinary skill in the art can understand that Figure 1 the structure shown is only schematic and does not limit the structure of the above terminal. For example, the terminal may further include more or fewer components than Figure 1 shown in the figure, or have a different configuration from Figure 1 shown in the figure.

[0065] The memory 104 can be used to store computer programs. For example, software programs and modules of application software, such as the computer program corresponding to the scanning method of the 3D scanner in this embodiment. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely set relative to the processor 102, and these remote memories can be connected to the terminal through a network. Examples of the above network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0066] The transmission device 106 is used to receive or send data via a network. The above network includes a wireless network provided by the communication provider of the terminal. In one instance, the transmission device 106 includes a network adapter (Network Interface Controller, abbreviated as NIC), which can be connected to other network devices through a base station and thus communicate with the Internet. In one instance, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0067] In this embodiment, a scanning method of a 3D scanner is provided. Figure 2 is a flowchart of the scanning method of the 3D scanner in this embodiment. As Figure 2 shown, the process includes the following steps:

[0068] Step S210, based on the scanning of the scanning object, obtain the three-dimensional point cloud set and the first matrix set of each frame; the first matrix set is a set of transformation matrices obtained by transforming the three-dimensional point cloud set of each frame into a preset third-party coordinate system;

[0069] Step S220: Based on the first matrix set, perform point cloud fusion on the three-dimensional point cloud set to obtain a first fused point cloud;

[0070] Step S230: Register the three-dimensional point clouds in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set; the second matrix set is a set of transformation matrices for position adjustment of each frame of the three-dimensional point cloud set;

[0071] Step S240: Based on the second matrix set, perform point cloud fusion on the three-dimensional point cloud set to obtain a target fused point cloud.

[0072] It should be noted that the method embodiment provided in this embodiment can be applied to a three-dimensional scanner; for example, the three-dimensional scanner can be a handheld three-dimensional scanner, a tracking scanner, etc. The three-dimensional scanner includes: an image acquisition device and a processor. Among them, the image acquisition device is used to scan a scanning object to obtain each frame of the three-dimensional point cloud set and the first matrix set; the processor can run the above processing method. The three-dimensional scanner may further include a fill light for filling light for the scanning object.

[0073] In this embodiment, there is no need to set marker points on the scanning object; directly using the three-dimensional scanner to scan the scanning object can obtain each frame of the three-dimensional point cloud set and the first matrix set; that is, for each frame of scanning, there is a corresponding three-dimensional point cloud set and the first matrix set. Among them, the first matrix set is a set of transformation matrices for converting each frame of the three-dimensional point cloud set to a preset third-party coordinate system. The transformation matrix includes a rotation matrix R and a translation matrix T. Using a neural network model, a point cloud fusion algorithm, etc., based on the first matrix set, perform point cloud fusion on the three-dimensional point cloud set to obtain a first fused point cloud; among them, the point cloud fusion algorithm is: align each three-dimensional point cloud to the same coordinate system through the rotation matrix R and the translation matrix T, and then perform fusion in ways such as surface reconstruction and voxel meshing to obtain the first fused point cloud. At this time, due to the lack of marker points, the RT matrix calculated for a single frame is not accurate enough. If the first fused point cloud is directly output as the final fused point cloud, it will lead to a decrease in the quality of the scanned data, that is, situations such as warping and knife marks occur in the scanned data.

[0074] Among them, it should be noted that if the third-party coordinate system is the world coordinate system; then both the registration and point cloud fusion steps are performed in this world coordinate system.

[0075] Therefore, registration is used to improve the accuracy of matrix calculation, and the second matrix set is used to replace the first matrix set to perform point cloud fusion on the three-dimensional point cloud set, thereby improving the quality of the scanned data. Specifically, registration algorithms such as the ICP algorithm, the NDT algorithm, and the 3DSC algorithm can be used to register the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain the second matrix set; the second matrix set is a set of transformation matrices for position adjustment of each frame of the three-dimensional point cloud set. Since the three-dimensional point cloud set, as the global point cloud, is fixed; then registering each frame of the three-dimensional point cloud with the global point cloud will make each frame of the three-dimensional point cloud fit more closely to the global point cloud. Therefore, when all the three-dimensional point clouds converge towards the same global point cloud, the tightness between the final frames of the three-dimensional point clouds will be more accurate, so that a second matrix set with higher accuracy can be obtained. Then, based on the second matrix set, point cloud fusion is performed on the three-dimensional point cloud set, thereby improving the accuracy of the target fused point cloud and improving the quality of the scanned data.

[0076] The above steps will be described in detail below:

[0077] In some of these embodiments, the step of performing point cloud fusion on the three-dimensional point cloud set based on the first matrix set to obtain the first fused point cloud includes the following steps:

[0078] Based on the first matrix set, the three-dimensional point cloud set is transformed into a third-party coordinate system; in the third-party coordinate system, point cloud fusion is performed on the three-dimensional point cloud set at a preset resolution to obtain the first fused point cloud;

[0079] Alternatively, the three-dimensional point cloud set is downsampled to obtain a downsampled point cloud set at a preset resolution;

[0080] Based on the first matrix set, the downsampled point cloud set is transformed to the third-party coordinate system for point cloud fusion to obtain the first fused point cloud.

[0081] Specifically, in order to reduce noise, a preset resolution is used to participate in point cloud fusion to obtain the first fused point cloud that meets this resolution. Among them, the resolution can be from 2mm to dozens of mm; there is no limitation on this. Preferably, the resolution can be 5mm. Preferably, the third-party coordinate system is the world coordinate system.

[0082] The following will be described in different implementation manners:

[0083] The first implementation method is as follows: Align to the world coordinate system based on the transformation matrices in the first matrix set; then perform point cloud fusion at a preset resolution in the world coordinate system to obtain a first fused point cloud that conforms to this resolution. This method has sufficient transformation matrices for its world coordinate system, can improve the quality of point cloud fusion, and combines the resolution to improve the calculation efficiency and reduce noise.

[0084] The second implementation method is as follows: Downsample the three-dimensional point cloud set to obtain a downsampled point cloud set at a preset resolution; among them, downsampling is pooling, which can be implemented using relevant technologies of pooling. The purpose is to reduce the dimension of features and retain effective information, avoid overfitting to a certain extent, and thus improve the quality of the sampled point cloud set. On this basis, based on the first matrix set, transform the downsampled point cloud set to the world coordinate system for point cloud fusion to obtain a first fused point cloud. This method can improve the quality of the first fused point cloud by improving the quality of the sampled point cloud set, combining the resolution to improve the calculation efficiency and reduce noise.

[0085] In some of these embodiments, registering the three-dimensional point clouds in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set in step S230 includes the following steps:

[0086] Step S231, traverse the three-dimensional point cloud set and the first matrix set, and register the three-dimensional point clouds in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set.

[0087] Specifically, the registration ICP (Iterative Closest Point) algorithm can be used to traverse the three-dimensional point cloud set and the first matrix set, and perform ICP registration on the three-dimensional point clouds in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set. The process of ICP registration is as follows: Given a reference point set (the first fused point cloud) P and a data point set (the three-dimensional point cloud set) Q (given the first matrix set (rotation matrix R and offset matrix R)), ICP registration finds the corresponding nearest point in P for each point in Q to form a matching point pair. Then, take the sum of the Euclidean distances of all paired points as the objective function to be solved, use singular value decomposition to find new R and T to minimize the objective function, transform the new Q' according to R and T, and find the corresponding point pairs again, and so on; the set of new R and T here is the second matrix set. In other embodiments, other registration methods can also be used to implement this, and no limitation is imposed on this.

[0088] Through this embodiment, using ICP registration to recalculate a set of new transformation matrices as the second matrix set can optimize the transformation matrices to improve the accuracy of subsequent point cloud fusion.

[0089] In some of these embodiments, such as Figure 3 shown, obtaining the three-dimensional point cloud set and the first matrix set for each frame based on the scanning of the scanning object in step S210 includes the following steps:

[0090] Step S211, obtaining the global marker point data based on the scanning of the marker points on the scanning object; at least one marker point is provided on the scanning object;

[0091] Step S212, obtaining the three-dimensional point cloud set, the first matrix set, and the marker point data for each frame based on the scanning of the scanning object; and determining the matching relationship between the marker point data and the global marker point data.

[0092] Specifically, this embodiment can be applicable to scanning objects with or without marker points. For a scanning object with marker points, the marker points on the scanning object will be scanned separately to obtain the global marker point data of the marker points. Among them, the global marker point data is obtained by weighted averaging the marker points on each frame of the scanned image and transferring them to the world coordinate system for weighted averaging. If there is one marker point set on the scanning object, then the global marker point data is the three-dimensional point cloud of this one marker point. It can be considered that the more the number of marker points, the better the quality of the final registration.

[0093] Among them, the scanning of the scanning object will scan the scanning object and the marker points together; therefore, for each frame of scanning, the three-dimensional point cloud set, the first matrix set, and the marker point data can be obtained; and the matching relationship between the marker point data and the global marker point data can be determined.

[0094] Through this embodiment, by scanning the scanning object, the three-dimensional point cloud set, the first matrix set, the marker point data, the global marker point data, and the matching relationship between the marker point data and the global marker point data for each frame are automatically obtained, thereby simplifying the calculation process and improving the calculation efficiency.

[0095] In some of these embodiments, on the basis of having marker points, registering the three-dimensional point clouds in each frame of the three-dimensional point cloud set with the first fused point cloud in step S230 to obtain the second matrix set includes the following steps:

[0096] Traverse the three-dimensional point cloud set, the first matrix set, and the marker point data, and register the three-dimensional point clouds in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain the second matrix set.

[0097] Specifically, the registration ICP (Iterative Closest Point) algorithm can be adopted to traverse the three-dimensional point cloud set, the first matrix set, and the marker point data, and register the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain the second matrix set. This process can be considered as using the distance relationship between the marker point data and the global marker point data as the marker point constraint, and adopting the ICP algorithm to register the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud. Among them, the distance relationship can be the minimum three-dimensional distance or the three-dimensional distance is less than a preset distance threshold. The principle and steps of the adopted registration algorithm are similar to those of step S231, and will not be elaborated here again.

[0098] Through this embodiment, by using the registration under the marker point constraint to recalculate a new set of transformation matrices as the second matrix set, the transformation matrix can be optimized to improve the local calculation accuracy.

[0099] In some of these embodiments, registering the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain the second matrix set includes the following steps:

[0100] Taking the coincidence of the section where each frame of the three-dimensional point cloud set is located and the section where the first fused point cloud is located, and the marker points remaining unchanged as the constraint conditions, register the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain the second matrix set.

[0101] In this embodiment, it can be considered as performing registration with marker point constraints, which includes two aspects of constraints between the marker points and the sections. Specifically: taking the marker points as anchor points, and taking the coincidence of the section where each frame of the three-dimensional point cloud set is located and the section where the first fused point cloud is located as the constraint condition, so that the three-dimensional distance value between the single-frame marker point data and the global marker point data is minimized, and the section where each frame of the three-dimensional point cloud set is located and the section where the first fused point cloud is located coincide as much as possible, thereby ensuring that during the registration process, it converges towards the same global point cloud, and further improving the accuracy of the second matrix set.

[0102] In some of these embodiments, as Figure 4 shown, registering the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain the second matrix set includes the following steps:

[0103] Step S410, based on the pre-constructed index of the first fused point cloud, determine the potential matching pairs between the three-dimensional point cloud in the three-dimensional point cloud set and the first fused point cloud;

[0104] Step S420: Screen out target matching pairs from the potential matching pairs based on preset search conditions. The search condition is that for search points with the same index, a target matching pair is formed between the first target point with the minimum Euclidean distance and the search point. The search points are points in the three-dimensional point cloud, and the first target points are points in the first fused point cloud.

[0105] Step S430: Obtain a second matrix set on the condition that the distance between the search point and the first target point in the target matching pair is the smallest.

[0106] Specifically, the implementation method adopted in this embodiment is the implementation method other than using the registration algorithm in step S230. Although the registration algorithm can reduce the configuration error of the point cloud and fill the gaps, if the scanning device and the tracking device in the three-dimensional scanning system are far apart, the calculated first matrix set will deviate from the true value, resulting in noise in some areas of the first fused point cloud, and further resulting in incorrect solutions in the obtained second matrix set, causing useful data to be erroneously excluded and reducing the integrity of the data in the three-dimensional point cloud set. The implementation process of this embodiment is as follows:

[0107] First, use a tree data structure to pre-construct the indexes of all points in the first fused point cloud; traverse all indexes to search for potential matching pairs between the three-dimensional point cloud in the three-dimensional point cloud set and the first fused point cloud. Among them, the search points are points in the three-dimensional point cloud, and the first target points are points in the first fused point cloud. A potential matching pair can be formed by a search point and the nearest first target point. Then, there may be multiple search points forming potential matching pairs with one first target point. At this time, use the search condition that for search points with the same index, a target matching pair is formed between the first target point with the minimum Euclidean distance and the search point to remove multiple potential matching pairs with a common first target point and screen out the target matching pairs. Finally, on the condition that the distance between the search point and the first target point in the target matching pair is the smallest, use algorithms such as linearized least squares method to obtain the second matrix set.

[0108] Through this embodiment, the accuracy of calculating the second matrix set can be further improved, the occurrence of useful data being erroneously excluded can be reduced, and the integrity of the data in the three-dimensional point cloud set can be guaranteed.

[0109] In some of the embodiments, as Figure 5 shown, determining the potential matching pairs between the three-dimensional point cloud in the three-dimensional point cloud set and the first fused point cloud based on the index of the pre-constructed first fused point cloud in step S410 includes the following steps:

[0110] Step S411: Construct the indexes of each first target point in the first fused point cloud;

[0111] Step S412: Determine the nearest point of the three-dimensional point in the three-dimensional point cloud in the first fused point cloud;

[0112] Step S413, based on a preset distance threshold, eliminate the abnormal points in the three-dimensional points to obtain non-abnormal three-dimensional points and the corresponding proportion.

[0113] Step S414, based on the non-abnormal three-dimensional points, determine potential matching pairs.

[0114] Specifically, use a binary tree (which can be a KD tree (K-Dimensional Tree)) to construct the index of each first target point in the first fused point cloud, which can be applied to the search for key data in multi-dimensional space; traverse the index to search for the nearest points in the first fused point cloud of the three-dimensional points in the three-dimensional point cloud corresponding to each index. Since there may be noise in the three-dimensional points in the three-dimensional point cloud, based on a preset distance threshold, eliminate the abnormal points in the three-dimensional points, that is, eliminate the point set greater than the distance threshold, so as to obtain non-abnormal three-dimensional points and the corresponding proportion ratio, thereby reducing the influence of noise on registration. Among them, the proportion of non-abnormal three-dimensional points can be used to evaluate the overall quality of this frame of three-dimensional point cloud to determine whether this frame of three-dimensional point cloud is deleted or retained.

[0115] At this time, the nearest points in the first fused point cloud of each three-dimensional point in the three-dimensional point cloud have been determined, and non-abnormal three-dimensional points have been obtained; based on this, the distances of all non-abnormal three-dimensional points on the section where the corresponding nearest points are located can be calculated, and the mean square error corresponding to the distance can also be obtained at this time; among them, the section where the nearest point is located refers to the section of the plane of the nearest point in the first fused point cloud. Then select the normal three-dimensional points with a distance less than the preset distance threshold, and form potential matching point pairs with the corresponding first target points. Among them, the distance threshold can be set by the user according to the usage scenario, and there is no limitation on this.

[0116] Through this embodiment, using the KD tree, effectively identify noise, and thus quickly and accurately calculate potential matching point pairs.

[0117] In some of these embodiments, the scanning method of the three-dimensional scanner further includes the following steps:

[0118] After obtaining the second matrix set, based on the second matrix set, convert the three-dimensional point clouds in each frame of the three-dimensional point cloud set into three-dimensional updated point clouds;

[0119] Respectively determine the mean square errors of the distances of the three-dimensional points in the three-dimensional updated point cloud and the three-dimensional point cloud on the section where the corresponding nearest points are located, and obtain the first mean square error and the second mean square error;

[0120] According to the first mean square error and the second mean square error, determine whether the current registration meets the convergence condition; the convergence condition is that the change amplitude between the first matrix set and the second matrix set is less than the preset amplitude threshold;

[0121] If the convergence condition is satisfied, the three-dimensional updated point cloud replaces the three-dimensional point cloud in the three-dimensional point cloud set;

[0122] If the convergence condition is not satisfied, the three-dimensional updated point cloud replaces the three-dimensional point cloud in the three-dimensional point cloud set; according to the number and proportion of target matching pairs, the updated three-dimensional updated point cloud in each frame of the three-dimensional point cloud set is re-registered with the first fused point cloud.

[0123] Specifically, in the process of optimizing and stitching the matrix set through registration, the updated second matrix set needs to ensure that the reprojection error of the marked points is as small as possible, and at the same time, the distance between the matching point pairs is as small as possible. Therefore, after obtaining the second matrix set, it is necessary to judge whether the calculation of the second matrix set is accurate, that is, whether the convergence condition is satisfied; thus avoiding the influence of the calculation error of the second matrix set. Then the specific implementation process is as follows:

[0124] Each three-dimensional point cloud in each frame of the three-dimensional point cloud set has a corresponding transformation matrix in the second matrix set; based on the second matrix set, the three-dimensional point cloud in each frame of the three-dimensional point cloud set is converted into a three-dimensional updated point cloud frame by frame or in parallel; calculate the first mean square error of the distance of the three-dimensional points in the three-dimensional updated point cloud on the corresponding nearest point plane; calculate the second mean square error of the distance of the three-dimensional points in the three-dimensional point cloud on the corresponding nearest point plane.

[0125] If the first mean error is less than the second mean square error, it is considered that the change amplitude between the first matrix set and the second matrix set is less than the preset amplitude threshold; the current registration satisfies the convergence condition, then the calculation of the second matrix set is accurate, and the three-dimensional updated point cloud replaces the three-dimensional point cloud in the three-dimensional point cloud set; then return to step S230 for execution.

[0126] If the first mean square error is greater than or equal to the second mean square error, it is considered that the change range between the first matrix set and the second matrix set is greater than or equal to the preset amplitude threshold; if the current registration does not meet the convergence condition, then according to the number and proportion of target matching pairs, the three-dimensional updated point cloud is used to replace the three-dimensional point cloud in the three-dimensional point cloud set; the updated three-dimensional updated point cloud in each frame of the three-dimensional point cloud set is re-registered with the first fused point cloud. Among them, replacing the three-dimensional point cloud in the three-dimensional point cloud set with the three-dimensional updated point cloud according to the number and proportion of target matching pairs can be divided into multiple situations: 1. If the number of target matching point pairs is 0 or the second matrix set contains non-numeric values, then discard this frame of three-dimensional point cloud; 2. If the proportion ratio of target matching pairs < 0.5, it is considered that this frame of three-dimensional point cloud belongs to edge data and can be discarded; 3. If the proportion ratio of target matching pairs >= 0.5, it is considered that this frame of three-dimensional point cloud belongs to central data, which is relatively important and should be retained and cannot be discarded. In other embodiments, if the convergence condition is still not met after reaching the preset number of iterations, then end, and use the matrix set corresponding to the smallest mean square error as the configured matrix set.

[0127] Through this embodiment, on the premise of reducing point cloud noise and filling data holes, the utilization rate of effective data is improved and data loss is reduced.

[0128] In some of these embodiments, the scanning method of the three-dimensional scanner further includes the following steps:

[0129] Based on the second matrix set, convert the first target point to the second target point; and calculate the average values of the distances from the first target point and the second target point to the corresponding tangent plane where the nearest point is located, respectively, to obtain the first target average value and the second target average value;

[0130] Compare the first target average value and the second target average value;

[0131] If the first target average value is greater than the second target average value, then the second matrix set meets the configuration requirements;

[0132] If the first target average value is less than or equal to the second target average value, then re-estimate the second matrix set.

[0133] Specifically, each three-dimensional point cloud in each frame of the three-dimensional point cloud set has a corresponding transformation matrix in the second matrix set; based on the second matrix set, convert the first target point to the second target point frame by frame or in parallel; then calculate the average value of the distance from the first target point to the corresponding tangent plane where the nearest point is located to obtain the first target average value; calculate the average value of the distance from the second target point to the corresponding tangent plane where the nearest point is located to obtain the second target average value.

[0134] If the first target average value is greater than the second target average value, the second matrix set meets the configuration requirements; the second matrix set replaces the first matrix set. If the first target average value is less than or equal to the second target average value, the second matrix set does not meet the configuration requirements, and the second matrix set is re-estimated by means of a pre-trained neural network model, a non-linear optimization algorithm, etc.

[0135] Through this embodiment, the average value is used to quickly evaluate whether the second matrix set meets the configuration requirements.

[0136] In some of these embodiments, the scanning method of the 3D scanner further includes the following steps:

[0137] When the scanning object is a ball gauge and at least one marking point is set beside the ball gauge, based on the target fused point cloud, determine the ball diameter of the ball gauge;

[0138] Based on the ball diameter of the ball gauge, determine the scanning accuracy.

[0139] As Figure 6 shown, in this embodiment, a tracking scanner is used to scan two ball gauges (ball gauge 1 and ball gauge 2), and at least one marking point (marking point 1 and marking point 2) is set beside each ball gauge. Among them, the tracking scanner has an image acquisition device with a large wide angle. In the embodiment of the method with markings, start scanning from ball gauge 1 in the direction of ball gauge 2 to obtain the target fused point clouds corresponding to ball gauge 1 and ball gauge 2 respectively; the target fused point cloud is a set of three-dimensional point clouds on the surface of the ball gauge. Select the two farthest three-dimensional points in the target fused point cloud and take the average to obtain the ball diameter of the ball gauge. Then compare the ball diameter of the ball gauge with the standard size of the ball gauge to determine the scanning accuracy; then the scanning accuracies of ball gauge 1 and ball gauge 2 can be determined to judge whether the scanning accuracy will change as the scanning progresses. The scanning accuracy can also be combined with the scanning angle, the number of iterations, etc. to display the current scanning accuracy in real time.

[0140] In some of these embodiments, the scanning method of the 3D scanner further includes the following steps:

[0141] Taking obtaining a new set of transformation matrices as one iteration, when the number of iterations meets a preset number threshold, obtain the target transformation matrix.

[0142] Specifically, obtaining a new set of transformation matrices each time is regarded as one iteration, which can be considered as steps S230 and S240 being executed once. For example: input the result of step S240 into step S230 to calculate a new set of transformation matrices; then return to step S240 for registration until the number of iterations meets the preset number threshold to obtain the target transformation matrix to further improve the quality of point cloud fusion.

[0143] Generally speaking, steps S230 and S240 only need to be executed once. Moreover, during the process of point cloud fusion and registration, local point cloud sets and corresponding matrix sets in the three-dimensional point cloud set can be selected for processing, which can also achieve the purpose of improving efficiency and quality, and thus will not be repeated here.

[0144] It should be noted that the steps shown in the above process or the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0145] In this embodiment, a scanning device of a three-dimensional scanner is also provided. This device is used to implement the above-mentioned embodiment and preferred implementation manners, and those that have been described will not be repeated. The following terms such as "module", "unit", "sub-unit", etc. can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0146] Figure 7 is the structural block diagram of the scanning device of the three-dimensional scanner in this embodiment. As Figure 7 shown, the device includes: an acquisition module 210, a first point cloud fusion module 220, a registration module 230, and a second point cloud fusion module 240;

[0147] The acquisition module 210 is used to acquire a three-dimensional point cloud set and a first matrix set for each frame based on the scanning of the scanning object; the first matrix set is a set of transformation matrices obtained by transforming each frame of the three-dimensional point cloud set into a preset third-party coordinate system;

[0148] The first point cloud fusion module 220 is used to perform point cloud fusion on the three-dimensional point cloud set based on the first matrix set to obtain a first fused point cloud;

[0149] The registration module 230 is used to register the three-dimensional points in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set; the second matrix set is a set of transformation matrices for adjusting the positions of each frame of the three-dimensional point cloud set;

[0150] The second point cloud fusion module 240 is used to perform point cloud fusion on the three-dimensional point cloud set based on the second matrix set to obtain a target fused point cloud.

[0151] With the above device, the problem in the related art that the RT matrix calculated for a single frame is not accurate enough, resulting in a decrease in the quality of the final scanned data, is solved. Registration is used to improve the accuracy of matrix calculation, and the second matrix set is used to replace the first matrix set to perform point cloud fusion on the three-dimensional point cloud set, thereby improving the quality of the scanned data.

[0152] In some of these embodiments, the first point cloud fusion module 220 is further configured to convert the three-dimensional point cloud set to a third-party coordinate system based on the first matrix set; in the third-party coordinate system, perform point cloud fusion on the three-dimensional point cloud set at a preset resolution to obtain a first fused point cloud;

[0153] Or, downsample the three-dimensional point cloud set to obtain a downsampled point cloud set at a preset resolution;

[0154] Based on the first matrix set, convert the downsampled point cloud set to the third-party coordinate system for point cloud fusion to obtain a first fused point cloud.

[0155] In some of these embodiments, the registration module 230 is further configured to traverse the three-dimensional point cloud set and the first matrix set, and perform ICP registration on the three-dimensional point cloud in each frame of the three-dimensional point cloud set and the first fused point cloud to obtain a second matrix set.

[0156] In some of these embodiments, the acquisition module 210 is further configured to obtain the global data of the marked points based on the scanning of the marked points on the scanned object; at least one marked point is provided on the scanned object;

[0157] Based on the scanning of the scanned object, obtain the three-dimensional point cloud set, the first matrix set, and the marked point data for each frame; and determine the matching relationship between the marked point data and the global marked point data.

[0158] In some of these embodiments, the registration module 230 is further configured to traverse the three-dimensional point cloud set, the first matrix set, and the marked point data, and perform registration on the three-dimensional point cloud in each frame of the three-dimensional point cloud set and the first fused point cloud to obtain a second matrix set.

[0159] In some of these embodiments, the registration module 230 is further configured to perform ICP registration on the three-dimensional point cloud in each frame of the three-dimensional point cloud set and the first fused point cloud with the coincidence of the section where each frame of the three-dimensional point cloud set is located and the section where the first fused point cloud is located, and the marked points remaining fixed as the constraint conditions to obtain a second matrix set.

[0160] In some of these embodiments, the registration module 230 is further configured to determine potential matching pairs between the three-dimensional point cloud in the three-dimensional point cloud set and the first fused point cloud based on the pre-constructed index of the first fused point cloud;

[0161] Based on preset search conditions, target matching pairs are screened out from potential matching pairs; the search conditions are that for search points with the same index, target matching pairs are formed between the first target point with the smallest Euclidean distance and the search point; the search points are points in the three-dimensional point cloud; the first target points are points in the first fused point cloud;

[0162] With the condition that the distance between the search point and the first target point in the target matching pair is the smallest, a second matrix set is obtained.

[0163] In some of these embodiments, the registration module 230 is further configured to construct the index of each first target point in the first fused point cloud;

[0164] Determine the nearest point of the three-dimensional points in the three-dimensional point cloud in the first fused point cloud;

[0165] Based on a preset distance threshold, abnormal points in the three-dimensional points are removed to obtain non-abnormal three-dimensional points and the corresponding proportion;

[0166] Based on the non-abnormal three-dimensional points, potential matching pairs are determined.

[0167] In some of these embodiments, the scanning device of the three-dimensional scanner further includes a convergence module;

[0168] The convergence module is configured to, after obtaining the second matrix set, based on the second matrix set, convert the three-dimensional point cloud in each frame of the three-dimensional point cloud set into a three-dimensional updated point cloud;

[0169] Respectively determine the mean square error of the distances of the three-dimensional points in the three-dimensional updated point cloud and the three-dimensional point cloud in the corresponding nearest point's tangent plane, to obtain a first mean square error and a second mean square error;

[0170] According to the first mean square error and the second mean square error, determine whether the current registration meets the convergence condition; the convergence condition is that the change amplitude between the first matrix set and the second matrix set is less than a preset amplitude threshold;

[0171] If the convergence condition is met, the three-dimensional point cloud in the three-dimensional point cloud set is replaced with the three-dimensional updated point cloud;

[0172] If the convergence condition is not met, then according to the number and proportion of the target matching pairs, the three-dimensional point cloud in the three-dimensional point cloud set is replaced with the three-dimensional updated point cloud; the updated three-dimensional updated point cloud in each frame of the three-dimensional point cloud set and the first fused point cloud are re-registered.

[0173] In some of these embodiments, the scanning device of the three-dimensional scanner further includes a configuration requirement comparison module;

[0174] A configuration requirement comparison module is used to convert a first target point into a second target point based on a second matrix set; and calculate the average values of the distances from the first target point and the second target point to the corresponding nearest point's tangent plane respectively, to obtain a first target average value and a second target average value;

[0175] Compare the first target average value and the second target average value;

[0176] If the first target average value is greater than the second target average value, the second matrix set meets the configuration requirements;

[0177] If the first target average value is less than or equal to the second target average value, re - estimate the second matrix set.

[0178] In some of these embodiments, the scanning device of the 3D scanner further includes: an accuracy calculation module;

[0179] The accuracy calculation module is used to determine the ball diameter of the ball gauge based on the target fused point cloud when the scanning object is a ball gauge and at least one marking point is set beside the ball gauge;

[0180] Determine the scanning accuracy based on the ball diameter of the ball gauge.

[0181] In some of these embodiments, the scanning device of the 3D scanner further includes: an iteration module;

[0182] The iteration module is used to take obtaining a new set of transformation matrices as one iteration, and obtain the target transformation matrix when the number of iterations meets a preset number threshold.

[0183] It should be noted that the above - mentioned various modules can be functional modules or program modules, and can be implemented either by software or by hardware. For the modules implemented by hardware, the above - mentioned various modules can be located in the same processor; or the above - mentioned various modules can also be located in different processors in any combined form.

[0184] In this embodiment, a computer device is also provided, including a memory and a processor. A computer program is stored in the memory, and the processor is set to run the computer program to execute the steps in any one of the above - mentioned method embodiments.

[0185] Optionally, the above - mentioned computer device may further include a transmission device and input - output devices. Among them, the transmission device is connected to the above - mentioned processor, and the input - output devices are connected to the above - mentioned processor.

[0186] Optionally, in this embodiment, the above - mentioned processor can be set to execute the following steps through the computer program:

[0187] S1. Based on the scanning of the object to be scanned, obtain the three-dimensional point cloud set and the first matrix set for each frame; the first matrix set is the set of transformation matrices for transforming the three-dimensional point cloud set of each frame into a preset third-party coordinate system;

[0188] S2. Based on the first matrix set, perform point cloud fusion on the three-dimensional point cloud set to obtain the first fused point cloud;

[0189] S3. Register the three-dimensional point clouds in the three-dimensional point cloud set of each frame with the first fused point cloud to obtain the second matrix set; the second matrix set is the set of transformation matrices for adjusting the positions of the three-dimensional point cloud set of each frame;

[0190] S4. Based on the second matrix set, perform point cloud fusion on the three-dimensional point cloud set to obtain the target fused point cloud.

[0191] It should be noted that specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation manners, and will not be elaborated in this embodiment.

[0192] In addition, in combination with the scanning method of the three-dimensional scanner provided in the above embodiments, a storage medium can also be provided to implement this in this embodiment. A computer program is stored on the storage medium; when the computer program is executed by a processor, it implements any one of the scanning methods of the three-dimensional scanner in the above embodiments.

[0193] It should be understood that the specific embodiments described here are only used to explain this application, rather than to limit it. According to the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of this application.

[0194] Obviously, the drawings are only some examples or embodiments of this application. For those of ordinary skill in the art, this application can also be applied to other similar situations according to these drawings without creative work. In addition, it can be understood that although the work done during the development process here may be complex and time-consuming, for those of ordinary skill in the art, certain design, manufacturing, or production changes based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient disclosure of this application.

[0195] The term "embodiment" in this application means that the specific features, structures, or characteristics described in combination with the embodiment may be included in at least one embodiment of this application. The phrase appears in various positions in the specification does not necessarily mean the same embodiment, nor does it mean being independent or alternative to other embodiments and mutually exclusive. Those of ordinary skill in the art can clearly or implicitly understand that the embodiments described in this application can be combined with other embodiments without conflict.

[0196] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of patent protection. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A scanning method of a three-dimensional scanner, characterized in that: include: Based on scanning the scanned object, a three-dimensional point cloud set and a first matrix set of each frame are obtained; The first matrix set is a set of transformation matrices for transforming the three-dimensional point cloud set in each frame into a preset third-party coordinate system; Based on the first matrix set, performing point cloud fusion on the three-dimensional point cloud set to obtain a first fused point cloud; The three-dimensional point cloud in each frame of the three-dimensional point cloud set is registered with the first fused point cloud to obtain a second matrix set; the second matrix set is a set of transformation matrices for position adjustment of the three-dimensional point cloud set in each frame; Based on the second matrix set, point cloud fusion is performed on the three-dimensional point cloud set to obtain a target fused point cloud.

2. The scanning method of the three-dimensional scanner according to claim 1, characterized in that: Based on the first matrix set, performing point cloud fusion on the three-dimensional point cloud set to obtain a first fused point cloud, including: Based on the first matrix set, the three-dimensional point cloud set is converted to the third-party coordinate system; in the third-party coordinate system, point cloud fusion is performed on the three-dimensional point cloud set at a preset resolution to obtain a first fused point cloud; Or, down-sampling the three-dimensional point cloud set to obtain a down-sampled point cloud set at a preset resolution; Based on the first matrix set, the downsampled point cloud set is converted to the third-party coordinate system for point cloud fusion to obtain a first fused point cloud.

3. The scanning method of the three-dimensional scanner according to claim 1, characterized in that: Based on scanning the scanned object, a three-dimensional point cloud set and a first matrix set of each frame are obtained, including: Based on scanning the marking point on the scanned object, obtaining the marking point global data; at least one marking point is set on the scanned object; Based on scanning the scanned object, a three-dimensional point cloud set, a first matrix set and marker point data of each frame are obtained; and a matching relationship between the marker point data and the marker point global data is determined.

4. The scanning method of the three-dimensional scanner according to claim 3, characterized in that: The three-dimensional point cloud in each frame of the three-dimensional point cloud set is registered with the first fused point cloud to obtain a second matrix set, including: The three-dimensional point cloud set, the first matrix set and the marking point data are traversed, and the three-dimensional point cloud in each frame of the three-dimensional point cloud set is registered with the first fused point cloud to obtain a second matrix set.

5. The scanning method of the three-dimensional scanner according to claim 4, characterized in that: The three-dimensional point cloud in each frame of the three-dimensional point cloud set is registered with the first fused point cloud to obtain a second matrix set, including: With the overlap of the section where the three-dimensional point cloud set in each frame is located and the section where the first fused point cloud is located, and the fixed marking points as constraints, the three-dimensional point cloud in the three-dimensional point cloud set in each frame is aligned with the first fused point cloud to obtain a second matrix set.

6. The scanning method of the three-dimensional scanner according to any one of claims 1 or 5, characterized in that: The three-dimensional point cloud in each frame of the three-dimensional point cloud set is registered with the first fused point cloud to obtain a second matrix set, including: Determining potential matching pairs between the three-dimensional point cloud in a three-dimensional point cloud set and the first fused point cloud based on the pre-constructed index of the first fused point cloud; Based on a preset search condition, a target matching pair is screened out from the potential matching pairs; the search condition is that for search points with the same index, a target matching pair is formed between a first target point with the smallest Euclidean distance and the search point; the search point is a point in the three-dimensional point cloud; the first target point is a point in the first fused point cloud; The second matrix set is obtained on the condition that the distance between the search point and the first target point in the target matching pair is minimized.

7. The scanning method of the three-dimensional scanner according to claim 6, characterized in that: Determining potential matching pairs between the three-dimensional point cloud in a three-dimensional point cloud set and the first fused point cloud based on the pre-constructed index of the first fused point cloud includes: Constructing an index of each first target point in the first fused point cloud; Determine the closest point of the three-dimensional point in the three-dimensional point cloud in the first fused point cloud; Based on a preset distance threshold, outliers in the three-dimensional points are eliminated to obtain non-outlier three-dimensional points and corresponding proportions; Based on non-anomalous 3D points, potential matching pairs are determined.

8. The scanning method of the three-dimensional scanner according to claim 7, characterized in that: The method further comprises: After obtaining the second matrix set, based on the second matrix set, converting the three-dimensional point cloud in each frame of the three-dimensional point cloud set into a three-dimensional update point cloud; Determine the mean square error of the distance between the three-dimensional updated point cloud and the three-dimensional point in the three-dimensional point cloud at the corresponding tangent plane where the nearest point is located, to obtain a first mean square error and a second mean square error; According to the first mean square error and the second mean square error, judging whether the current registration satisfies a convergence condition; the convergence condition is that the change amplitude between the first matrix set and the second matrix set is less than a preset amplitude threshold; If the convergence condition is met, replacing the three-dimensional point cloud in the three-dimensional point cloud set with the three-dimensional updated point cloud; If the convergence condition is not met, the three-dimensional point cloud in the three-dimensional point cloud set is replaced by the three-dimensional updated point cloud according to the number of the target matching pairs and the proportion; and the updated three-dimensional updated point cloud in the three-dimensional point cloud set of each frame is realigned with the first fused point cloud.

9. The scanning method of the three-dimensional scanner according to claim 6, characterized in that: The method further comprises: Based on the second matrix set, the first target point is converted into a second target point; and the average values ​​of the distances from the first target point and the second target point to the corresponding tangent planes of the nearest points are respectively calculated to obtain the first target average value and the second target average value; comparing the first target average value and the second target average value; If the first target average value is greater than the second target average value, then the second matrix set meets the configuration requirement; If the first target average value is less than or equal to the second target average value, the second matrix set is re-estimated.

10. The scanning method of the three-dimensional scanner according to claim 6, characterized in that: The method further comprises: In a case where the scanning object is a spherical gauge and at least one marking point is set next to the spherical gauge, determining the spherical diameter of the spherical gauge based on the target fused point cloud; Based on the spherical diameter of the spherical gauge, a scanning accuracy is determined.

11. A scanning device of a three-dimensional scanner, characterized in that: include: An acquisition module, a first point cloud fusion module, a registration module, and a second point cloud fusion module; The acquisition module is used to acquire a three-dimensional point cloud set and a first matrix set of each frame based on scanning the scanned object; the first matrix set is a set of transformation matrices for converting the three-dimensional point cloud set of each frame into a preset third-party coordinate system; The first point cloud fusion module is used to perform point cloud fusion on the three-dimensional point cloud set based on the first matrix set to obtain a first fused point cloud; The registration module is used to register the three-dimensional point cloud in each frame of the three-dimensional point cloud set with the first fused point cloud to obtain a second matrix set; the second matrix set is a set of transformation matrices for position adjustment of the three-dimensional point cloud set in each frame; The second point cloud fusion module is used to perform point cloud fusion on the three-dimensional point cloud set based on the second matrix set to obtain a target fused point cloud.

12. A computer device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps of the scanning method of the three-dimensional scanner according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the scanning method of the three-dimensional scanner according to any one of claims 1 to 10 are implemented.

Images