Gear three-dimensional structure construction method, system, equipment and medium

Abstract

The invention discloses a gear three-dimensional structure construction method, system and device and a medium, and the method comprises the steps: obtaining at least two groups of regional point cloud data of a to-be-detected gear and gear rotation angles corresponding to the regional point cloud data; the regional point cloud data is obtained by scanning through a laser scanner based on a multi-axis rotary clamping mechanism adjusting gear rotation angles, and the gear rotation angles corresponding to any two groups of regional point cloud data are different; calculating a rotation center axis of the gear to be measured according to the point cloud data of all areas; registering every two adjacent groups of regional point cloud data based on the rotation center axis and the gear rotation angle corresponding to the regional point cloud data to obtain registered regional point cloud data; and constructing a pose map structure of the to-be-detected gear based on the registered regional point cloud data to obtain a three-dimensional structure of the to-be-detected gear, so that a helical tooth surface with a complex curved surface can be comprehensively scanned, and non-contact, efficient and high-precision detection of the gear is realized.

Description

Technical Field

[0001] This application relates to the technical field of gear processing, and in particular, to a method, system, device and medium for constructing a three-dimensional structure of a gear. Background Art

[0002] As an important component in industrial development, spiral bevel gears are subject to high performance requirements and service loads. Among them, the quality inspection of spiral bevel gears is an important process for inspecting the manufacturing quality of gears.

[0003] Traditional quality inspection of spiral bevel gears mainly uses a contact-type coordinate measuring machine for detection. However, this contact-type measurement method is not only cumbersome to operate, but also can only detect a few local points at a time, resulting in incomplete detection data and inability to quickly obtain the global information of the tooth surface of the spiral bevel gear. In addition, traditional quality inspection of spiral bevel gears requires multiple detections to complete the overall inspection, resulting in large test errors, low data reliability, and poor detection efficiency. Summary of the Invention

[0004] The following is an overview of the subject matter described in detail in this document. This overview is not intended to limit the scope of protection of the claims.

[0005] The main purpose of the embodiments of the present disclosure is to provide a method, system, device and storage medium for constructing a three-dimensional structure of a gear, which can quickly and accurately obtain the global three-dimensional point cloud information of the gear and achieve efficient and high-precision data detection.

[0006] The first aspect of the embodiments of this application provides a method for constructing a three-dimensional structure of a gear for a central controller, and the method includes:

[0007] Obtain at least two sets of regional point cloud data of the gear to be measured and the gear rotation angles corresponding to the regional point cloud data. The regional point cloud data is obtained by scanning with a laser scanner based on the adjustment of the gear rotation angle by a multi-axis rotary clamping mechanism, and the gear rotation angles corresponding to any two sets of regional point cloud data are different;

[0008] Calculate the rotation central axis of the gear to be measured according to all the regional point cloud data;

[0009] Register every two adjacent sets of regional point cloud data based on the rotation central axis and the gear rotation angles corresponding to the regional point cloud data to obtain the registered regional point cloud data;

[0010] Construct a pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured.

[0011] An embodiment of the present application provides a method for constructing a three-dimensional structure of a gear. By obtaining at least two sets of regional point cloud data of the gear to be measured and the gear rotation angles corresponding to the regional point cloud data, the regional point cloud data is obtained by scanning with a laser scanner based on the adjustment of the gear rotation angle by a multi-axis rotation clamping mechanism, and the gear rotation angles corresponding to any two sets of regional point cloud data are different; calculating the rotation center axis of the gear to be measured according to all the regional point cloud data; registering every two adjacent sets of regional point cloud data based on the rotation center axis and the gear rotation angles corresponding to the regional point cloud data to obtain the registered regional point cloud data; constructing a pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured, which can comprehensively scan the spiral tooth surface with complex curved surfaces and realize non-contact, efficient, high-precision detection of the gear.

[0012] In some embodiments of the present application, the calculating the rotation center axis of the gear to be measured according to all the regional point cloud data includes:

[0013] Segmenting the gear cone surface area in each of the regional point cloud data to obtain first point cloud data;

[0014] Performing plane fitting on the first point cloud data to obtain the plane normal vector of the first point cloud data;

[0015] Calculating the normal line of the gear to be measured according to the plane normal vector;

[0016] Calculating the rotation center axis of the spiral bevel gear to be measured according to the normal line of the gear to be measured.

[0017] In some embodiments of the present application, the segmenting the gear cone surface area in all the regional point cloud data to obtain first point cloud data includes:

[0018] Performing preprocessing on all the regional point cloud data to obtain first processed point cloud data, where the preprocessing includes at least one of filtering processing and downsampling;

[0019] Extracting the point cloud data of the gear cone surface area from the first processed point cloud data to obtain second processed point cloud data;

[0020] Filtering the second processed point cloud data according to a preset threshold to obtain the first point cloud data.

[0021] In some embodiments of the present application, the registering every two adjacent sets of regional point cloud data based on the rotation center axis and the gear rotation angles corresponding to the regional point cloud data to obtain the registered regional point cloud data includes:

[0022] Perform initial registration on all the regional point cloud data according to the rotation center axis and the gear rotation angle corresponding to the regional point cloud data, and obtain the corresponding roughly registered regional point cloud data;

[0023] Perform secondary registration on every two adjacent groups of roughly registered regional point cloud data to obtain the corresponding registered regional point cloud data.

[0024] In some embodiments of the present application, before constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data, it further includes:

[0025] Calculate the transformation matrix between every two adjacent groups of regional point cloud data according to the roughly registered regional point cloud data and the registered regional point cloud data;

[0026] Calculate the information matrix of the gear to be measured according to the regional point cloud data, the registered regional point cloud data, and the transformation matrix.

[0027] In some embodiments of the present application, constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data includes:

[0028] Construct pose graph nodes of the gear to be measured according to all the second regional point cloud data and the gear rotation angle corresponding to the second regional point cloud data;

[0029] Based on the pose graph nodes and the information matrix, construct the pose graph structure of the gear to be measured according to the transformation matrix.

[0030] In some embodiments of the present application, constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured includes:

[0031] Use a preset graph optimization algorithm to optimize the pose graph structure of the gear to be measured to obtain the three-dimensional structure of the gear to be measured.

[0032] To achieve the above object, a second aspect of the embodiments of the present invention provides a gear three-dimensional structure construction system, and the system includes:

[0033] An acquisition module, configured to acquire at least two groups of regional point cloud data of the gear to be measured and the gear rotation angle corresponding to the regional point cloud data, where the regional point cloud data is obtained by scanning with a laser scanner by adjusting the gear rotation angle through a multi-axis rotation clamping mechanism, and the gear rotation angles corresponding to any two groups of regional point cloud data are different;

[0034] A calculation module, configured to calculate the rotation center axis of the gear to be measured according to all the regional point cloud data;

[0035] A registration module, configured to register every two adjacent sets of regional point cloud data based on the rotation center axis and the gear rotation angle corresponding to the regional point cloud data, so as to obtain registered regional point cloud data;

[0036] A construction module, configured to construct a pose graph structure of the gear to be measured based on the registered regional point cloud data, so as to obtain a three-dimensional structure of the gear to be measured.

[0037] To achieve the above object, a third aspect of the embodiments of the present invention provides an electronic device, including: at least one control processor and a memory communicatively connected to the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor, so that the at least one control processor can execute the above-mentioned method for constructing a three-dimensional structure of a gear.

[0038] To achieve the above object, a fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, which stores computer-executable instructions for causing a computer to execute the above-mentioned method for constructing a three-dimensional structure of a gear.

[0039] It can be understood that the beneficial effects of the above-mentioned second aspect to the fourth aspect compared with the related art are the same as those of the above-mentioned first aspect compared with the related art. For the relevant descriptions, reference can be made to the relevant descriptions in the above-mentioned first aspect, and details will not be repeated here. Description of the Drawings

[0040] The above and / or additional aspects and advantages of the present application will become obvious and easy to understand from the following description of the embodiments in conjunction with the accompanying drawings, where:

[0041] Figure 1 is a schematic flowchart of a method for constructing a three-dimensional structure of a gear provided by an embodiment of the present application;

[0042] Figure 2 is a schematic diagram of a multi-axis rotary clamping mechanism provided by an embodiment of the present application;

[0043] Figure 3 is a schematic diagram of spiral bevel gear point cloud data provided by an embodiment of the present application;

[0044] Figure 4 is a schematic diagram of the result of gear point cloud preprocessing provided by an embodiment of the present application;

[0045] Figure 5 is a gear point cloud preprocessing interaction interface provided by an embodiment of the present application;

[0046] Figure 6 is the data points required for calculating the rotation center provided by an embodiment of the present application;

[0047] Figure 7 It is a schematic diagram of coordinate normalization transformation provided by an embodiment of the present application;

[0048] Figure 8 It is the point cloud data of the rough registration area provided by an embodiment of the present application;

[0049] Figure 9 It is a schematic diagram of optimized fine registration provided by an embodiment of the present application;

[0050] Figure 10 It is a schematic diagram of the three-dimensional construction result of the gear tooth surface provided by an embodiment of the present application;

[0051] Figure 11 It is a schematic diagram of the structure of a three-dimensional gear structure construction system provided by an embodiment of the present application;

[0052] Figure 12 It is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0053] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present application and should not be construed as a limitation to the present application.

[0054] In the description of the present application, if the first, second, etc. are described only for the purpose of distinguishing technical features, it should not be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence of the indicated technical features.

[0055] In the description of the present application, it should be understood that the orientation descriptions such as up and down indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present application.

[0056] In the description of the present application, it should be noted that unless otherwise clearly defined, words such as setting, installation, and connection should be understood in a broad sense. Those skilled in the art can reasonably determine the specific meanings of the above words in the present application in combination with the specific content of the technical solution.

[0057] With the rapid development of the aviation industry, as the core power system of the aviation industry, the key transmission component, the spiral bevel gear, is subject to increasingly high performance requirements and operating loads. Among them, the quality inspection work of spiral bevel gears is an important process for inspecting the manufacturing quality of gears.

[0058] In traditional quality inspection work of spiral bevel gears, a contact-type coordinate measuring machine is used for detection. This contact-type operation measurement method is cumbersome, and only a few local points can be detected at a time, unable to obtain the global information of the gear tooth surface and unable to meet the requirements of high-efficiency quality inspection of spiral bevel gears for aviation engines.

[0059] Moreover, currently, the quality inspection of aviation spiral bevel gears by traditional contact measurement mainly relies on coordinate measuring machines, which have problems such as cumbersome operation, low efficiency, and only being able to detect local information of the tooth surface, and cannot meet the current quality inspection requirements of aviation spiral bevel gears. In addition, traditional quality inspection of spiral bevel gears requires multiple detections to complete the overall inspection, with problems such as large test errors, low data reliability, and poor detection efficiency.

[0060] Based on this, the embodiments of the present application provide a method, system, electronic device, and medium for constructing a three-dimensional structure of a gear, aiming to be able to quickly and accurately obtain the global three-dimensional point cloud information of the gear and achieve high-efficiency and high-precision data detection.

[0061] The method, system, electronic device, and medium for constructing a three-dimensional structure of a gear provided by the embodiments of the present application are specifically described through the following embodiments. First, the method for constructing a three-dimensional structure of a gear in the embodiments of the present application is described.

[0062] The embodiments of the present application can obtain and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0063] The basic technologies of artificial intelligence generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. The software technologies of artificial intelligence mainly include several major directions such as computer vision technology, robotics technology, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0064] The three-dimensional structure construction method of gears provided by the embodiments of the present application relates to the technical field of spiral bevel gear processing. The three-dimensional structure construction method of gears provided by the embodiments of the present application can be applied to a terminal, a server, or software running on a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the three-dimensional structure construction method of gears, etc., but is not limited to the above forms.

[0065] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0066] It should be noted that in each specific embodiment of the present application, when it comes to performing relevant processing based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first. Moreover, the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiments of the present application need to obtain the user's sensitive personal information, the user's separate permission or separate consent will be obtained through methods such as pop-up windows or redirecting to a confirmation page. After clearly obtaining the user's separate permission or separate consent, the necessary user-related data for the normal operation of the embodiments of the present application will be obtained.

[0067] For this purpose, refer to Figure 1, embodiments of the present application provide a method for constructing a three-dimensional structure of a gear. This method is applied to a central controller, which can be a server, an electronic device, a mobile terminal, etc., and no specific limitation is made here. The method includes the following steps S110 to S140.

[0068] Step S110: Obtain at least two groups of regional point cloud data of the gear to be measured and the gear rotation angles corresponding to the regional point cloud data. The regional point cloud data is obtained by scanning with a laser scanner while adjusting the gear rotation angle based on a multi-axis rotating clamping mechanism. The gear rotation angles corresponding to any two groups of regional point cloud data are different.

[0069] In this step, by adjusting the rotation angle of the gear to be measured through a multi-axis rotating clamping mechanism, scanning the gear to be measured with a laser scanner to obtain regional point cloud data, and obtaining the gear rotation angle corresponding to each group of regional point cloud data, the global information of the shape and size of the gear to be measured is obtained, the stability and reliability of the detection of aviation spiral bevel gears are improved, and it can be applied to the three-dimensional reconstruction of various different models of aviation spiral bevel gears within the measurement range.

[0070] In one embodiment, the spatial attitude of the gear is adjusted through a multi-axis rotating clamping mechanism, so that the line laser scanner obtains regional point cloud data from different observation directions. Each group of regional point cloud data corresponds to a specific gear rotation angle (such as 0°, 45°, 90°...), and the gear rotation angle recorded in real time by a high-precision encoder built in the clamping mechanism is obtained, providing an initial pose constraint for subsequent point cloud registration, realizing coverage of the entire tooth surface through multi-viewpoint cloud + angle constraint, avoiding errors caused by traditional missing tooth tip / root data, and meeting the robust registration requirements through the scanning angle difference.

[0071] As Figure 2 shown, the multi-axis rotating clamping mechanism includes: a three-jaw chuck, a rotating slide table, and an inclined slide table. The three-jaw chuck in the multi-axis rotating clamping mechanism fixes the spiral bevel gear during scanning, so as to obtain a suitable scanning perspective by adjusting the inclined slide table and the rotating slide table, and then make the spiral bevel gear rotate at equal angles around the gear center axis, obtain the three-dimensional point cloud data of the tooth surface of the spiral bevel gear from different perspectives, realize the full-round scanning of the gear tooth surface, and obtain the auxiliary information of the gear rotation angle. Among them, the gear rotation angle is recorded in real time by a high-precision encoder built in the multi-axis rotating clamping mechanism (such as triggering a scan every 5° of rotation), providing an initial pose constraint for subsequent point cloud registration.

[0072] Specifically, a high-precision linear motion platform driven by a linear motor, the motion platform includes a motion plate for driving the measured spiral bevel gear to move along the Y direction, and generating a pulse signal for cooperating with the line laser to sample the point cloud, so as to obtain the three-dimensional point cloud of the tooth surface of the bevel gear to be measured.

[0073] Furthermore, high-density three-dimensional point cloud data of the tooth surface of the spiral bevel gear to be measured is obtained through a 2D line laser sensor head. The line laser controller receives the data collected by the sensor head and realizes the transmission of data and instructions with the server. The server receives the point cloud data obtained by the line laser, and sends relevant instructions and deploys and implements the three-dimensional point cloud preprocessing and registration algorithms for the tooth surface of the spiral bevel gear.

[0074] Specifically, the 2D line laser is fixed above the high-precision linear motion platform through a bracket. In addition, the height, position, and angle of the line laser can be adjusted through the bracket. The line laser sensor head is connected to the line laser controller through a signal line, and the controller is connected to the server through a network cable. The server and the line laser controller perform high-speed data transmission through a transmission protocol. The line laser controller receives the encoding signal from the encoder of the high-precision linear motion platform through the signal line and the AB phase of the encoder. The multi-axis rotary clamping mechanism is fixed on the motion platform. When scanning the spiral bevel gear, the clamping mechanism moves along the Y direction together with the moving plate.

[0075] In some embodiments, based on the line laser scanning platform and the multi-axis rotary clamping mechanism, three-dimensional point cloud data of the tooth surface of the spiral bevel gear and its gear rotation angle auxiliary information at different perspectives are scanned and collected to obtain the global point cloud information of the tooth surface of the spiral bevel gear, as Figure 3 shown in the schematic diagram of the point cloud data of the collected spiral bevel gear.

[0076] Among them, the gear rotation angle information is provided by the multi-axis rotary clamping mechanism. When scanning the spiral bevel gear to be measured, the spiral bevel gear to be measured is fixed by the three-jaw chuck of the fixture. Then, the rotary slide table is used to rotate the spiral bevel gear to be measured at equal angles as needed. When rotating one week, the tooth surface point cloud data at different perspectives is obtained simultaneously. Through the combination of multi-axis active rotary scanning and angle encoding feedback, high-precision detection of the entire tooth surface is realized, effectively solving the core problems such as insufficient data coverage and low registration efficiency in the prior art, and having significant patent innovation and industrial practicability.

[0077] Furthermore, by utilizing the advantages of non-contact measurement of the line laser scanning platform and high-precision data acquisition of the multi-axis rotary clamping mechanism, it is possible to efficiently and highly accurately obtain the surface shape and size data of the spiral bevel gear to be measured. Among them, the line laser scanner can obtain tooth surface point clouds with high precision and high density. In addition, based on the multi-axis rotary clamping mechanism, it is possible to perform omnidirectional scanning of the spiral tooth surface with complex curved surfaces and obtain angle information to assist in the efficient three-dimensional reconstruction of the point cloud of the spiral bevel gear to be measured.

[0078] Step S120: Calculate the rotation center axis of the gear to be measured according to all regional point cloud data.

[0079] In this step, the point cloud data of all regions of the gear to be measured is segmented to extract the point cloud data for calculating the rotation center of the gear to be measured, and the first point cloud data is obtained. Furthermore, plane fitting is performed on the first point cloud data to obtain the plane normal vector of the gear to be measured, and the normal line of the gear to be measured is calculated based on the plane normal vector to obtain the rotation center axis of the gear to be measured. Thus, the coordinates of the rotation center axis of the gear to be measured are calculated through the conical surface geometric characteristics (normal vector projection intersection line) of the point cloud data. By directly constructing a reference using the point cloud geometric parameters of the spiral bevel gear to be measured, it can adapt to complex industrial environments, maintain high precision under interference such as oil stains and wear, and improve the detection efficiency and repeatability.

[0080] In some embodiments, before calculating the rotation center axis of the spiral bevel gear to be measured according to the three-dimensional point cloud data in step S120, the following steps are included:

[0081] Preprocess all region point cloud data to obtain the first processed point cloud data. The preprocessing includes at least one of filtering and downsampling;

[0082] Extract the point cloud data of the gear conical surface region from the first processed point cloud data to obtain the second processed point cloud data;

[0083] Screen the second processed point cloud data according to a preset threshold to obtain the first point cloud data.

[0084] As Figure 4 is the three-dimensional point cloud data of the gear to be measured after preprocessing. The preprocessing operation includes filtering the three-dimensional point cloud data of the gear to be measured obtained by scanning at different angles (to remove irrelevant point clouds, such as the bottom plate) and downsampling operation (to achieve accelerated calculation).

[0085] In this embodiment, first, preprocess all region point cloud data to obtain the first processed point cloud data. Preferably: filter each group of region point cloud data; further, downsample each group of region point cloud data, and then obtain the first processed point cloud data after preprocessing.

[0086] Furthermore, extract the point cloud data of the gear conical surface region from the first processed point cloud data. Preferably, use Euclidean clustering to extract the conical surface region to obtain the second processed point cloud data. Then, screen the second processed point cloud data according to a preset threshold. Preferably, screen the conical surface points through normal vector consistency check (angle threshold < 15°) to obtain the first point cloud data, realizing the reduction of detection data error and the improvement of detection accuracy.

[0087] In some embodiments, the three-dimensional point cloud data of the gear to be measured can be filtered by filtering methods such as statistical filtering, radius filtering, pass-through filtering, and voxel grid downsampling. Among them, statistical filtering is a method based on the neighborhood characteristics of points to identify and remove outliers. It calculates the mean and standard deviation of the distances between each point and its neighboring points, and then removes those points with a large deviation from the average distance of the neighboring points. Radius filtering determines whether to retain a point by checking the number of neighbors within a specified radius of each point. If a point does not have enough neighbors in its neighborhood, it is considered an outlier and removed. Pass-through filtering allows users to selectively retain some points in the point cloud according to the coordinate range in a certain axis. For example, threshold ranges are set in the three dimensions of x, y, and z, and only the points that meet the conditions are retained. Voxel grid downsampling divides the space into uniformly sized cubes (voxels), and then uses the centroid or center point of all points within each voxel to represent the entire voxel. This method can significantly reduce the data volume of the point cloud while maintaining its basic structure. Specific selection is based on processing requirements and data characteristics and is not limited here. Usually, multiple filtering techniques are combined to achieve the best effect. For example, statistical filtering is first used to remove obvious noise points, then voxel grid downsampling is applied to reduce the data volume, and finally pass-through filtering is used to focus on the region of interest.

[0088] In some embodiments, in step S120, calculating the rotation center axis of the spiral bevel gear to be measured according to the three-dimensional point cloud data includes the following steps S210 to S240:

[0089] Step S210, segment the gear cone surface region in each regional point cloud data to obtain the first point cloud data;

[0090] Step S220, perform plane fitting on the first point cloud data to obtain the plane normal vector of the first point cloud data;

[0091] Step S230, calculate the normal line of the gear to be measured according to the plane normal vector;

[0092] Step S240, calculate the rotation center axis of the spiral bevel gear to be measured according to the normal line of the gear to be measured.

[0093] In this embodiment, to perform regional segmentation on the gear to be measured, it is necessary to first preprocess the three-dimensional point cloud data, such as Figure 5It is an interactive interface for preprocessing the point cloud of the gear to be measured. Then, based on the actual scanned point cloud (regional point cloud data) of the gear to be measured, the first point cloud data containing only the conical surface feature region is segmented by the region growing algorithm. Among them, the region growing algorithm is an image segmentation technology, mainly used to divide an image into multiple parts or regions. Specifically, it starts from a set of predefined seed points and gradually merges adjacent pixels with similar properties (such as gray value, color, texture, etc.) to these seed points into the same region.

[0094] Specifically, the regional point cloud data scanned from the gear to be measured is filtered, and the conical surface region is extracted by Euclidean clustering to achieve region segmentation. Then, the conical points of the gear to be measured are screened through normal vector consistency check to remove noise from the three-dimensional point cloud data.

[0095] In some embodiments, first, in the preprocessing stage, the gear to be measured is regionally segmented to extract the point cloud data required for calculating the rotation center, such as Figure 6 The data points required for calculating the rotation center, and then through the plane fitting algorithm, the extracted point cloud data is plane-fitted to obtain the normal vector of the plane.

[0096] Furthermore, the normal line perpendicular to the conical surface is calculated using the straight line formula. Specifically, the two conical surface normal lines pointing to the rotation center axis are projected onto the plane of the end face of the gear to be measured (perpendicular to the rotation center axis), and then the intersection points of the two projected lines on the end face of the gear to be measured are calculated. Then, the rotation center axis is calculated using the straight line formula. The normal line is obtained by making the plane of the gear to be measured perpendicular to the conical surface of the gear to be measured.

[0097] In this embodiment, through calculation based on the actual point cloud, the measurement method adapts to any deformation state of the gear to be measured, with a wider applicability. Moreover, the traditional artificial / calibration block dependence mode is replaced by geometric feature drive, and there are significant improvements in adaptability to complex industrial environments, measurement efficiency improvement, and error chain control, laying a core benchmark for later global registration and contact imprint analysis.

[0098] In some embodiments, the regional point cloud data of the gear to be measured is preprocessed. First, data filtering is performed (such as statistical outlier removal, mean K = 50, standard deviation threshold 1.5), then region segmentation is performed, preferably using Euclidean clustering to extract the conical surface region (clustering distance threshold δ = 2 mm), and finally noise removal is performed. Specifically, the conical points are screened through normal vector consistency check (angle threshold < 15°), and finally the first point cloud data after segmentation is obtained. Among them, the selected conical surface region needs to satisfy axial symmetry to ensure that the plane fitting result reflects the geometric symmetry characteristics of the gear to be measured.

[0099] Further, it is preferably to calculate the normal vector direction of the gear to be measured according to the first point cloud data obtained after region segmentation by using the least squares plane fitting algorithm, and then obtain the plane normal vector of the first point cloud data. Among them, the least squares plane fitting is a data fitting technique used to find the best fitting plane for a set of data points. Specifically, it is achieved by minimizing the sum of the squares of the perpendicular distances (errors) from the actual data points to the fitting plane, so as to extract meaningful information from the chaotic data and provide a basis for subsequent processing.

[0100] Further, project the normal lines of each conical fitting plane onto the end face plane of the gear to be measured (usually the XY plane) to generate projection lines, and then solve the rotation center axis through the intersection points of two or more projection lines. Thus, through the dynamic analysis of the geometric features of the gear to be measured, the dependence on external calibration is eliminated, and the reference is directly constructed by using the point cloud geometric parameters of the gear to be measured, realizing the rapid calculation of the rotation center axis and significantly improving the detection efficiency and repeatability.

[0101] Step S130: Register each adjacent two groups of regional point cloud data based on the rotation center axis and the gear rotation angle corresponding to the regional point cloud data to obtain the registered regional point cloud data.

[0102] In this step, first, perform primary registration on all regional point cloud data in sequence according to the rotation center axis and the gear rotation angle corresponding to the regional point cloud data to obtain the corresponding roughly registered regional point cloud data, and then perform secondary registration on each adjacent two groups of roughly registered regional point cloud data to obtain the corresponding registered regional point cloud data.

[0103] Specifically, first, based on the gear rotation angle corresponding to the regional point cloud data and the calculated rotation center axis of the gear to be measured, perform rough registration rotation on the three-dimensional point cloud data of the gear to be measured to achieve the global initial alignment of the point cloud of the gear to be measured. Among them, the gear rotation angle is preferably provided by the gear rotation angle information provided by the multi-axis rotation clamping mechanism, for example, recorded in real time by a high-precision encoder built in the multi-axis rotation clamping mechanism.

[0104] Further, based on the roughly registered regional point cloud data, perform secondary registration on each adjacent two groups of roughly registered regional point cloud data to obtain the corresponding registered regional point cloud data, so as to reduce the data error in the detection process.

[0105] In some embodiments, registering each adjacent two groups of regional point cloud data based on the rotation center axis and the gear rotation angle corresponding to the regional point cloud data in step S130 to obtain the registered regional point cloud data includes the following steps S310 to S320:

[0106] Step S310: Perform primary registration on all regional point cloud data according to the rotation center axis and the gear rotation angle corresponding to the regional point cloud data to obtain the corresponding roughly registered regional point cloud data;

[0107] Step S320: Perform secondary registration on every two adjacent groups of roughly registered regional point cloud data to obtain the corresponding registered regional point cloud data.

[0108] In this embodiment, based on the gear rotation angle information provided by the multi-axis rotation clamping mechanism during the acquisition process and the calculated rotation center axis of the gear to be measured, rough rotation registration is performed on the three-dimensional point cloud data of the gear to be measured.

[0109] Specifically, through the parameter rotation matrix, an equal-angle rotation transformation of the point cloud of the gear to be measured around the rotation center axis is realized according to the three-dimensional point cloud data of the gear to be measured after preprocessing and the scanned fixed angle, and a rough registration result is obtained. As Figure 7 shown as the roughly registered regional point cloud data.

[0110] Furthermore, based on the Robust Kernel-based Iterative Closest Point (Robust ICP) algorithm between every two adjacent groups of roughly registered regional point cloud data, precise registration is performed on the initially aligned three-dimensional point cloud data to obtain the corresponding registered regional point cloud data. Among them, the Robust Kernel-based Iterative Closest Point (Robust ICP) algorithm is an improved version of the traditional ICP algorithm, aiming to improve the stability of point cloud registration in scenarios with noise, outliers, and partial overlaps.

[0111] Specifically, the traditional ICP iteration includes matching point pairs, calculating the transformation, applying the transformation, and convergence judgment. The key improvement of Robust ICP is the error calculation and optimization stage. By introducing an error weighting mechanism, Robust ICP significantly improves the robustness of the traditional ICP without significantly increasing the computational complexity. Among them, the selection of the kernel function (such as Huber / Tukey) and parameter adjustment need to be optimized according to the specific scenario and are not limited here, so as to balance noise suppression and convergence speed.

[0112] Furthermore, the specific steps of Robust ICP include: first, find the nearest neighbor (corresponding point) in the target point cloud for each point in the source point cloud, and then calculate the residual, that is, calculate the geometric error of each corresponding point pair.

[0113] Furthermore, a robust kernel function can also be applied. By applying the kernel function to each residual and then performing weighted optimization, the optimal rotation matrix and translation vector are solved through weighted least squares, where the weight is determined by the derivative of the kernel function (for example, the Huber kernel assigns low weights to large errors). Then, iterative updates are performed. By applying the transformation and repeating the above steps until convergence (such as when the change in transformation parameters is less than the threshold or the residuals are stable), the interference of outliers and noise can be effectively suppressed, and the registration accuracy can be improved.

[0114] In some embodiments, before constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data in step S140, the following steps S410 to S420 are included:

[0115] Step S410: Calculate the transformation matrix between every two adjacent groups of regional point cloud data according to the coarsely registered regional point cloud data and the registered regional point cloud data;

[0116] Step S420: Calculate the information matrix of the gear to be measured according to the regional point cloud data, the registered regional point cloud data, and the transformation matrix.

[0117] In this embodiment, the transformation matrix is obtained through the iterative closest point fine registration system based on the robust sum function, and then the information matrix is calculated according to the three-dimensional point cloud data, the first point cloud data, and the transformation matrix.

[0118] Specifically, the specific methods for obtaining the fine registration transformation matrix and the information matrix are as follows: (The algorithm basic module comes from the point cloud processing code library Open3d). Among them, the transformation matrix is obtained by the ICP fine registration algorithm module based on the robust sum function, specifically implemented through the function registration_icp(). The information matrix is calculated from the source point cloud, the target point cloud, and their transformation matrix, specifically implemented through the function get_information_matrix_from_point_clouds().

[0119] Specifically, the iterative closest point fine registration based on the robust sum function adopts the ICP fine registration with the Huber loss function, and combines the GTSAM graph optimization library (Georgia Tech Smoothing and Mapping) to optimize the pose graph, solving the problem of error accumulation in traditional point cloud stitching.

[0120] Step S140: Construct the pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured.

[0121] In this step, based on the constructed pose graph structure, the graph nodes and edges of the pose graph structure are optimized through a preset global registration system to obtain the three-dimensional point cloud of the gear to be measured.

[0122] In some embodiments, in step S140, a pose graph structure of the gear to be measured is constructed based on the registered regional point cloud data, including the following steps S510 to S520:

[0123] Step S510: Construct pose graph nodes of the gear to be measured according to all the second regional point cloud data and the gear rotation angles corresponding to the second regional point cloud data;

[0124] Step S520: Based on the pose graph nodes and the information matrix, construct the pose graph structure of the gear to be measured according to the transformation matrix.

[0125] In this embodiment, the transformation matrix refers to the relative position and rotation relationship between adjacent point clouds, usually including a rotation matrix and a translation vector. The information matrix reflects the reliability or uncertainty of this transformation. For example, in ICP registration, the information matrix is determined by the overlap degree or matching error of the point clouds.

[0126] Furthermore, a pose graph is constructed based on the transformation matrix and the information matrix. The specific structure can be that each node represents the pose (position and direction) of a point cloud, and the edge represents the transformation relationship and its uncertainty between adjacent nodes. Then, the transformation matrix and the information matrix are used as edges to connect the corresponding nodes, obtaining the pose graph structure of the gear to be measured. Thus, through the global optimization of the pose graph, the overall accuracy and robustness of the registration are significantly improved.

[0127] In some embodiments, in step S140, a pose graph structure of the gear to be measured is constructed based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured, including the following step S610:

[0128] Step S610: Use a preset graph optimization algorithm to optimize the pose graph structure of the gear to be measured to obtain the three-dimensional structure of the gear to be measured.

[0129] In this embodiment, based on the constructed pose graph structure, the graph nodes of the pose graph structure and the edges of the pose graph structure are optimized through a preset global registration system to obtain the three-dimensional structure of the gear to be measured.

[0130] As Figure 9 shown is a schematic diagram of optimized fine registration. Specifically, the specific steps of global optimization registration based on graph optimization include: constructing a pose graph structure of the point cloud of the spiral bevel gear to be measured through a graph structure module, and specifically adding graph nodes through the pose_graph.nodes.append() function.

[0131] Further, the pose_graph.edges.append() function is used to add edges to the graph nodes, and the transformation matrices and information matrices of all the spiral bevel gear point clouds to be registered are used for the construction of the pose graph, obtaining the overall graph structure of the spiral bevel gear point clouds to be measured; then, based on the constructed graph structure of the spiral bevel gear point clouds to be measured, the global registration module is used to globally optimize the graph nodes and edges, specifically implemented through the global_optimization() function, to complete the global fine registration of the three-dimensional point cloud of the spiral bevel gear tooth surface.

[0132] Further, the globally optimized spiral bevel gear point clouds to be measured are fused into a whole, finally realizing the construction of the three-dimensional point cloud structure of the spiral bevel gear tooth surface, as Figure 10 shown in the schematic diagram of the construction result of the three-dimensional structure of the spiral bevel gear tooth surface to be measured.

[0133] In one embodiment, first, the spiral bevel gear to be measured is fixed on a rotary stage ( Figure 1 ), triggering the line laser scanner (wavelength 650 nm, accuracy ±2 μm) to move along the Y-axis, synchronously collecting point cloud data, triggering a scan every 5°, and then receiving the encoder AB phase signals ( Figure 2 ) through the laser controller, generating a point cloud time series and performing noise filtering.

[0134] Further, a pass-through filter is performed on the single-viewpoint point cloud. Among them, the pass-through filter is mainly used to remove the bottom plate and other irrelevant point clouds, and the specific filtering range may need to be determined according to the actual situation. The conical surface area is extracted, and then the RANSAC algorithm is used to fit the normal plane of the conical surface, project the normal to the end face plane of the spiral bevel gear to be measured ( Figure 6 ), and calculate the intersection point to determine the rotation center axis. Among them, RANSAC (Random Sample Consensus) is a model fitting algorithm with strong robustness, specifically estimating the optimal model parameters from the data containing noise and outliers through random sampling and iterative voting.

[0135] Further, according to the rotation center axis (direction vector), a normalized coordinate transformation is performed on all single-viewpoint point clouds (regional point cloud data). The formula for the normalized coordinate transformation is as follows:

[0136] First, according to the rotation center point p o = [x0, y0, z0], a translation transformation is performed on the point cloud data. The source point cloud is p, and after translation, it is pt:

[0137] pt = p - p o ;

[0138] Then, perform a rotation transformation on the point cloud data after the translation transformation. Let the normal vector of the rotation center axis be p normal , and the unit vector of the z-axis z = [0, 0, 1].

[0139] Furthermore, it can be known from the Rodriguez rotation matrix formula that to solve a rotation matrix, two parameters need to be calculated first. One is the rotation axis k of the rotation matrix, and the other is the rotation angle α. Therefore, calculate the rotation axis k and normalize it to be denoted as u. The specific formula is as follows:

[0140] k = p normal ×z;

[0141]

[0142] Further, calculate the rotation angle α:

[0143] α = arccos(p normal ·z);

[0144] Then, apply the Rodriguez formula to calculate the rotation matrix:

[0145] R = I + (sinα)U + (1 - cosα)U 2 ;

[0146] where, I is the identity matrix, and U is the cross product matrix of u.

[0147] Furthermore, after calculating the rotation matrix, perform rotation normalization on the point cloud data after the translation transformation. The normalized point cloud is denoted as p or :

[0148]

[0149] where, R ij represents the elements at different positions in the matrix and is a scalar.

[0150] If the coordinate transformation of the point cloud data is performed in the form of a pose matrix, then the rotation matrix R needs to be converted into a homogeneous matrix T. The transformation relationship is as follows:

[0151]

[0152] where, R is the rotation matrix based on the rotation center axis, T is the transformation matrix, P is the single-viewpoint point cloud, and P ′ is the single-viewpoint point cloud after the coordinate transformation.

[0153] Specifically, perform normalized coordinate transformation on the point cloud data of the gear to be measured according to the rotation center axis, as Figure 6 shown, so that the center of the point cloud of the gear to be measured is zeroed to the world coordinate center.

[0154] Furthermore, based on the encoder angle θ and the coordinates of the rotation center axis, the point cloud is rotated by θ around the center axis for rough registration ( Figure 8 ), and the ICP fine registration with a Huber kernel (loss threshold k = 0.1) is performed on the adjacent view point clouds to generate a transformation matrix and an information matrix.

[0155] When performing rough registration on the point cloud data, the calculation formula of the transformation matrix is the same as that of the coordinate normalization part. Since the point cloud data has been normalized, the rotation axis for calculating the rotation matrix is the z-axis. Specifically, the unit vector of the rotation axis is z = [0, 0, 1], and the rotation angle is θ:

[0156] R robust = I + (sinθ)K + (1 - cosθ)K 2

[0157] where K is the cross product matrix of z. R robust is the rough registration rotation matrix and can be converted into the rough registration pose matrix T robust .

[0158] Furthermore, use the GTSAM library to construct a pose graph ( Figure 9 ), and output the globally optimized three-dimensional reconstructed point cloud ( Figure 10 ). The overall measurement efficiency is improved, the measurement accuracy is optimized, and the measurement robustness is enhanced.

[0159] In this embodiment, by obtaining at least two sets of regional point cloud data of the gear to be measured and the gear rotation angles corresponding to the regional point cloud data, the regional point cloud data is obtained by scanning with a laser scanner based on the adjustment of the gear rotation angle by a multi-axis rotation clamping mechanism, and the gear rotation angles corresponding to any two sets of regional point cloud data are different; calculating the rotation center axis of the gear to be measured according to all the regional point cloud data; registering each adjacent two sets of regional point cloud data based on the rotation center axis and the gear rotation angles corresponding to the regional point cloud data to obtain the registered regional point cloud data; constructing a pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured, which can comprehensively scan the helical tooth surface with a complex curved surface and realize non-contact, efficient and high-precision detection of the gear.

[0160] As Figure 11 shown, some embodiments of the present application provide a gear three-dimensional structure construction system, which includes an acquisition module 1110, a calculation module 1120, a registration module 1130, and a construction module 1140. Specifically:

[0161] An acquisition module 1110 is configured to acquire at least two sets of regional point cloud data of a gear to be measured and the gear rotation angles corresponding to the regional point cloud data. The regional point cloud data is obtained by scanning with a laser scanner while adjusting the gear rotation angle based on a multi-axis rotation clamping mechanism, and the gear rotation angles corresponding to any two sets of regional point cloud data are different.

[0162] A calculation module 1120 is configured to calculate the rotation central axis of the gear to be measured according to all the regional point cloud data.

[0163] A registration module 1130 is configured to register every two adjacent sets of regional point cloud data based on the rotation central axis and the gear rotation angles corresponding to the regional point cloud data, so as to obtain the registered regional point cloud data.

[0164] A construction module 1140 is configured to optimize the pose graph structure according to the gear point cloud map structure, so as to obtain the three-dimensional point cloud of the spiral bevel gear to be measured.

[0165] In some embodiments, the calculation module 1120 may include: segmenting the gear cone surface area in each set of regional point cloud data to obtain first point cloud data.

[0166] In some embodiments, the calculation module 1120 may include: performing plane fitting on the first point cloud data to obtain the plane normal vector of the first point cloud data.

[0167] In some embodiments, the calculation module 1120 may include: calculating the normal line of the gear to be measured according to the plane normal vector.

[0168] In some embodiments, the calculation module 1120 may include: calculating the rotation central axis of the spiral bevel gear to be measured according to the normal line of the gear to be measured.

[0169] In some embodiments, the calculation module 1120 may include: preprocessing all the regional point cloud data to obtain first processed point cloud data, and the preprocessing includes at least one of filtering processing and downsampling.

[0170] In some embodiments, the calculation module 1120 may include: extracting the point cloud data of the gear cone surface area from the first processed point cloud data to obtain second processed point cloud data.

[0171] In some embodiments, the calculation module 1120 may include: screening the second processed point cloud data according to a preset threshold to obtain first point cloud data.

[0172] In some embodiments, the registration module 1130 may include: performing primary registration on all the regional point cloud data according to the rotation central axis and the gear rotation angles corresponding to the regional point cloud data to obtain corresponding roughly registered regional point cloud data.

[0173] In some embodiments, the registration module 1130 may include: performing secondary registration on the point cloud data of every two adjacent sets of roughly registered regions to obtain the corresponding registered regional point cloud data.

[0174] In some embodiments, the registration module 1130 may include: calculating the transformation matrix between every two adjacent sets of regional point cloud data according to the roughly registered regional point cloud data and the registered regional point cloud data.

[0175] In some embodiments, the registration module 1130 may include: calculating the information matrix of the gear to be measured according to the regional point cloud data, the registered regional point cloud data, and the transformation matrix.

[0176] In some embodiments, the registration module 1130 may include: constructing the pose graph nodes of the gear to be measured according to all the second regional point cloud data and the gear rotation angles corresponding to the second regional point cloud data.

[0177] In some embodiments, the registration module 1130 may include: constructing the pose graph structure of the gear to be measured based on the pose graph nodes and the information matrix according to the transformation matrix.

[0178] In some embodiments, the registration module 1130 may include: optimizing the pose graph structure of the gear to be measured by using a preset graph optimization algorithm to obtain the three-dimensional structure of the gear to be measured.

[0179] It should be noted that the gear three-dimensional structure construction system provided in this embodiment and the above-mentioned gear three-dimensional structure construction method are based on the same inventive concept. Therefore, the relevant content of the above-mentioned gear three-dimensional structure construction method also applies to the content of the gear three-dimensional structure construction system. Therefore, it will not be elaborated here.

[0180] To solve the problems that the traditional spiral bevel gear quality inspection needs to be carried out multiple times to complete the overall inspection, there are large test errors, low data reliability, and poor detection efficiency, the system obtains at least two sets of regional point cloud data of the gear to be measured and the gear rotation angles corresponding to the regional point cloud data. The regional point cloud data is obtained by scanning with a laser scanner based on the adjustment of the gear rotation angle by a multi-axis rotary clamping mechanism, and the gear rotation angles corresponding to any two sets of regional point cloud data are different; calculating the rotation central axis of the gear to be measured according to all the regional point cloud data; registering every two adjacent sets of regional point cloud data based on the rotation central axis and the gear rotation angles corresponding to the regional point cloud data to obtain the registered regional point cloud data; constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured. In this way, it is possible to achieve a comprehensive scan of the spiral tooth surface with a complex curved surface and achieve non-contact, high-efficiency, and high-precision detection of the gear.

[0181] An embodiment of the present application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above-mentioned method for constructing a three-dimensional structure of a gear is implemented.

[0182] As Figure 12 , Figure 12 is a schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present application. The electronic device includes:

[0183] At least one battery;

[0184] At least one memory;

[0185] At least one processor;

[0186] At least one program;

[0187] The program is stored in the memory, and the processor executes at least one program to implement a method for constructing a three-dimensional structure of a gear as described above in the present disclosure.

[0188] The electronic device can be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), an in-vehicle computer, etc.

[0189] The electronic device of the embodiment of the present application will be introduced in detail below.

[0190] The processor 1600 can be implemented by using a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present disclosure;

[0191] The memory 1700 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 1700 can store an operating system and other application programs. When implementing the technical solutions provided by the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 1700 and are called by the processor 1600 to execute a method for constructing a three-dimensional structure of a gear in the embodiments of the present disclosure.

[0192] The input / output interface 1800 is used to implement information input and output;

[0193] A communication interface 1900, which is used to implement the communication interaction between this device and other devices. It can achieve communication through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.);

[0194] A bus 2000, which transmits information between various components of the device (such as a processor 1600, a memory 1700, an input / output interface 1800, and a communication interface 1900);

[0195] Among them, the processor 1600, the memory 1700, the input / output interface 1800, and the communication interface 1900 achieve communication connections with each other inside the device through the bus 2000.

[0196] The embodiments of the present disclosure also provide a storage medium, which is a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to make a computer execute the above-mentioned method for constructing a three-dimensional structure of a gear.

[0197] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory can include high-speed random access memory, and can also include non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely provided with respect to the processor, and these remote memories can be connected to the processor through a network. Examples of the above-mentioned network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0198] The embodiments described in the embodiments of the present disclosure are for more clearly explaining the technical solutions of the embodiments of the present disclosure, and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. Those skilled in the art can know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems.

[0199] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present disclosure, and may include more or fewer steps than those shown in the figures, or combine some steps, or different steps.

[0200] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0201] Those of ordinary skill in the art will understand that all or some of the steps in the methods disclosed above, and the functional modules / units in systems and devices, can be implemented as software, firmware, hardware, or a suitable combination thereof.

[0202] As used in the specification of this application and the above accompanying drawings, the terms "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such data may be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order different from those illustrated or described herein. In addition, the terms "comprising" and "having", and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that comprises a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0203] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist simultaneously. Here, A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or a similar expression means any combination of these items, including any combination of single items (ones) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0204] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection to each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be in electrical, mechanical, or other forms.

[0205] The unit described as a separating component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0206] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, can also exist separately as individual physical units, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0207] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0208] The above has specifically described the preferred implementation of the embodiments of the present application, but the embodiments of the present application are not limited to the above implementation manners. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the embodiments of the present application, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the embodiments of the present application.

[0209] The above has described the embodiments of the present application in detail with reference to the accompanying drawings, but the present application is not limited to the above embodiments. Various changes can be made without departing from the purpose of the present application within the knowledge scope of ordinary technical personnel in the technical field to which it belongs.

Claims

1. A method for constructing a three-dimensional structure of a gear, characterized in that, The method includes: Obtaining at least two sets of regional point cloud data of the gear to be measured and the corresponding gear rotation angles of the regional point cloud data. The regional point cloud data is obtained by scanning with a laser scanner after adjusting the gear rotation angle based on a multi-axis rotary clamping mechanism, and the gear rotation angles corresponding to any two sets of regional point cloud data are different; Calculating the rotation central axis of the gear to be measured according to all the regional point cloud data; Registering every two adjacent sets of regional point cloud data based on the rotation central axis and the corresponding gear rotation angles of the regional point cloud data to obtain the registered regional point cloud data; Constructing a pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured.

2. The method for constructing a three-dimensional structure of a gear according to claim 1, wherein The calculating the rotation central axis of the gear to be measured according to all the regional point cloud data includes: Segmenting the gear cone surface area in each of the regional point cloud data to obtain first point cloud data; Performing plane fitting on the first point cloud data to obtain the plane normal vector of the first point cloud data; Calculating the normal line of the gear to be measured according to the plane normal vector; Calculating the rotation central axis of the spiral bevel gear to be measured according to the normal line of the gear to be measured.

3. The gear three-dimensional structure construction method according to claim 2, characterized in that, The segmenting the gear cone surface area in all the regional point cloud data to obtain first point cloud data includes: Performing preprocessing on all the regional point cloud data to obtain first processed point cloud data, and the preprocessing includes at least one of filtering processing and downsampling; Extracting the point cloud data of the gear cone surface area from the first processed point cloud data to obtain second processed point cloud data; Screening the second processed point cloud data according to a preset threshold to obtain the first point cloud data.

4. The gear three-dimensional structure construction method according to claim 3, characterized in that, The registering every two adjacent sets of regional point cloud data based on the rotation central axis and the corresponding gear rotation angles of the regional point cloud data to obtain the registered regional point cloud data includes: Performing primary registration on all the regional point cloud data according to the rotation central axis and the corresponding gear rotation angles of the regional point cloud data to obtain corresponding roughly registered regional point cloud data; Performing secondary registration on every two adjacent sets of roughly registered regional point cloud data to obtain the corresponding registered regional point cloud data.

5. The gear three-dimensional structure construction method according to claim 4, characterized in that, Before constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data, it further includes: Calculating the transformation matrix between every two adjacent sets of regional point cloud data according to the roughly registered regional point cloud data and the registered regional point cloud data; Calculating the information matrix of the gear to be measured according to the regional point cloud data, the registered regional point cloud data, and the transformation matrix.

6. The gear three-dimensional structure construction method according to claim 5, characterized in that The constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data includes: Constructing pose graph nodes of the gear to be measured according to all the second regional point cloud data and the corresponding gear rotation angles of the second regional point cloud data; Constructing the pose graph structure of the gear to be measured based on the pose graph nodes and the information matrix according to the transformation matrix.

7. The gear three-dimensional structure construction method according to claim 1, characterized in that Constructing the pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured, including: Optimizing the pose graph structure of the gear to be measured by using a preset graph optimization algorithm to obtain the three-dimensional structure of the gear to be measured.

8. A three-dimensional structure construction system for gears, characterized in that, The system includes: An acquisition module, configured to acquire at least two groups of regional point cloud data of the gear to be measured and the gear rotation angles corresponding to the regional point cloud data, where the regional point cloud data is obtained by scanning with a laser scanner based on the adjustment of the gear rotation angle by a multi-axis rotation clamping mechanism, and the gear rotation angles corresponding to any two groups of regional point cloud data are different; A calculation module, configured to calculate the rotation central axis of the gear to be measured according to all the regional point cloud data; A registration module, configured to register every two adjacent groups of regional point cloud data based on the rotation central axis and the gear rotation angles corresponding to the regional point cloud data to obtain the registered regional point cloud data; A construction module, configured to construct the pose graph structure of the gear to be measured based on the registered regional point cloud data to obtain the three-dimensional structure of the gear to be measured.

9. An electronic device, characterized in that: It includes at least one control processor and a memory for communicatively connecting with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute a method for constructing a three-dimensional structure of a gear according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to cause a computer to execute a method for constructing a three-dimensional structure of a gear according to any one of claims 1 to 7.

Images