A real-time measurement method for tile flatness based on minimum containing area

Abstract

This application provides a real-time method for measuring the flatness of ceramic tiles based on a minimum containment region, belonging to the field of flatness measurement technology. The method includes: acquiring the original scan dataset of the target object, the original scan dataset including original point cloud data; extracting convex hull endpoints based on the original point cloud data to obtain a convex hull endpoint set, the convex hull endpoint set containing multiple target convex hull endpoints, the convex hull endpoint set being a subset of the original scan dataset; performing containment region detection on the convex hull endpoint set based on a containment region detection model to obtain target containment plane parameters, the target containment plane parameters being used to construct a first target containment plane and a second target containment plane; and calculating the planar distance based on the target containment plane parameters to obtain the target flatness of the target object. This application embodiment can achieve comprehensive detection of the surface flatness of the object to be measured, thereby improving the accuracy and efficiency of flatness detection.

Description

Technical Field

[0001] This application relates to the field of flatness measurement technology, and in particular to a method for real-time measurement of tile flatness based on a minimum containment area. Background Technology

[0002] Flatness testing refers to the process of measuring and evaluating the flatness of an object's surface. In engineering, manufacturing, and construction, flatness testing is used to inspect the surface flatness of objects, and surface flatness is a crucial quality indicator that directly affects product quality and appearance. Therefore, surface flatness testing is a vital step in the quality inspection of large industrial products such as ceramic tiles and steel strips.

[0003] However, current flatness testing technologies typically employ manual methods. For example, the tile industry primarily relies on manual inspection to measure flatness. Manual inspection is inefficient, suitable only for random sampling of tiles, and cannot achieve comprehensive testing of all products. Furthermore, manual inspection often relies on feeler gauges, which can only measure flatness in a few fixed directions, failing to capture complete surface information and potentially leading to incomplete and inaccurate flatness measurements. Based on this, existing technologies have proposed using laser displacement sensors to replace manual inspection. However, this sensor acquisition method is also easily limited by fixed locations, unable to comprehensively inspect the entire tile surface, thus affecting the accuracy of the final flatness measurement. Therefore, how to achieve comprehensive surface flatness testing of the object under test and improve the accuracy and efficiency of flatness detection has become an urgent technical problem to be solved. Summary of the Invention

[0004] The main objective of this application is to propose a real-time measurement method for tile flatness based on the minimum containment area, which can achieve comprehensive detection of the surface flatness of the object under test, thereby improving the accuracy and efficiency of flatness detection.

[0005] To achieve the above objectives, a first aspect of this application proposes a method for real-time measurement of tile flatness based on a minimum containment area, the method comprising:

[0006] Obtain the original scan dataset of the target object, which includes original point cloud data. The original scan dataset is used to represent a set of multiple original point cloud data after scanning the surface of the target object based on a preset multidimensional scanner.

[0007] Convex hull endpoints are extracted based on the original point cloud data to obtain a convex hull endpoint set, which contains multiple target convex hull endpoints and is a subset of the original scan dataset.

[0008] The target convex hull endpoint set is subjected to convex hull endpoint detection based on the convex hull endpoint detection model to obtain target convex hull endpoint parameters. The target convex hull endpoint parameters are used to construct a first target convex hull endpoint and a second target convex hull endpoint. The first target convex hull endpoint and the second target convex hull endpoint are two parallel planes, and the region between the first target convex hull endpoint and the second target convex hull endpoint is the minimum convex hull endpoint. The minimum convex hull endpoint is used to characterize the minimum region that contains all the target convex hull endpoints.

[0009] The flatness of the target object is obtained by calculating the planar distance based on the target containment plane parameters.

[0010] In some embodiments, the step of extracting convex hull endpoints based on the original point cloud data to obtain a convex hull endpoint set includes:

[0011] Randomly obtain first candidate point cloud data and second candidate point cloud data from the original point cloud data;

[0012] A candidate point cloud structure is constructed based on the first candidate point cloud data, and the candidate point cloud structure includes candidate structure surfaces;

[0013] Based on the candidate structural surface, point cloud data is selected from the second candidate point cloud data to obtain the target convex hull endpoint. The target convex hull endpoint is used to characterize the second candidate point cloud data that is farthest from the candidate structural surface.

[0014] The convex hull endpoint set is constructed based on multiple target convex hull endpoints.

[0015] In some embodiments, the step of performing containment region detection on the convex hull endpoint set based on the containment region detection model to obtain target containment plane parameters includes:

[0016] Obtain initial bounding plane parameters and a preset number of iterations. The initial bounding plane parameters are used to construct a first initial bounding plane and a second initial bounding plane, which are two parallel planes.

[0017] Obtain the point cloud coordinates of the endpoint of the target convex hull to obtain the point cloud coordinates of the convex hull endpoint;

[0018] Based on the preset number of iterations, the initial containment plane parameters, and the point cloud coordinates of the convex hull endpoints, a function is constructed to obtain the target optimization function;

[0019] The target optimization function is calculated to determine the parameters of the target encompassing plane.

[0020] In some embodiments, the step of performing function computation on the target optimization function to determine the target encompassing plane parameters includes:

[0021] For the current iteration number, perform function calculation on the target optimization function to obtain the candidate iteration bounding plane parameters;

[0022] Based on the preset number of iterations and the candidate iteration containment plane parameters, adjacent iteration containment plane parameters are obtained. The adjacent iteration containment plane parameters are used to characterize the candidate iteration containment plane parameters obtained in the next iteration number after the current iteration number.

[0023] The flatness difference is calculated based on the candidate iteration bounding plane parameter corresponding to the current iteration number and the adjacent iteration bounding plane parameter to obtain the adjacent flatness difference;

[0024] When the adjacent smoothness difference of the current iteration number is less than the preset smoothness difference threshold, the adjacent smoothness difference is used as a candidate smoothness difference.

[0025] The target containment plane parameters are determined based on the candidate flatness difference.

[0026] In some embodiments, the initial bounding plane parameters include initial normal vector sub-parameters and initial origin distance sub-parameters. The step of constructing a target optimization function based on the preset number of iterations, the initial bounding plane parameters, and the coordinates of the convex hull endpoint point cloud includes:

[0027] Construct an initial objective sub-function based on the initial normal vector sub-parameters;

[0028] Based on the initial normal vector sub-parameters, the convex hull endpoint point cloud coordinates, and the initial origin distance sub-parameters, an initial condition sub-function is constructed;

[0029] Construct an initial objective function based on the initial objective sub-function and the initial condition sub-function;

[0030] The gradient of the initial normal vector sub-parameters is calculated based on the preset number of iterations to obtain the normal vector gradient sub-parameters.

[0031] Based on the preset number of iterations, the gradient of the initial origin distance sub-parameter is calculated to obtain the origin distance gradient sub-parameter;

[0032] Construct an optimization objective function based on the normal vector gradient sub-parameters;

[0033] Based on the normal vector gradient sub-parameter, the convex hull endpoint point cloud coordinates, and the origin distance gradient sub-parameter, an optimization condition sub-function is constructed.

[0034] The objective optimization function is constructed based on the objective sub-function and the condition sub-function.

[0035] In some embodiments, determining the target containment plane parameters based on the candidate flatness differences includes:

[0036] The smallest candidate smoothness difference among the multiple candidate smoothness differences is taken as the target smoothness difference;

[0037] The candidate iterative containment plane parameter corresponding to the target flatness difference is used as the target containment plane parameter.

[0038] In some embodiments, obtaining the original scan dataset of the target object includes:

[0039] The target object is scanned by the multi-dimensional scanner to obtain surface point cloud data.

[0040] Acquire preset transmission point cloud data, which is used to characterize the point cloud data of the transmission platform on which the target object is placed;

[0041] The surface point cloud data is selected based on the preset transmitted point cloud data to obtain the original point cloud data, and the original scan dataset is constructed based on the original point cloud data.

[0042] To achieve the above objectives, a second aspect of this application provides a real-time tile flatness measurement device based on a minimum containment area, the device comprising:

[0043] The acquisition module is used to acquire the original scan dataset of the target object. The original scan dataset includes original point cloud data. The original scan dataset is used to represent a set of multiple original point cloud data after scanning the surface of the target object based on a preset multidimensional scanner.

[0044] The extraction module is used to extract convex hull endpoints based on the original point cloud data to obtain a convex hull endpoint set, which contains multiple target convex hull endpoints and is a subset of the original scan dataset.

[0045] The detection module is used to perform containment region detection on the set of convex hull endpoints based on the containment region detection model to obtain target containment plane parameters. The target containment plane parameters are used to construct a first target containment plane and a second target containment plane. The first target containment plane and the second target containment plane are two parallel planes, and the area between the first target containment plane and the second containment plane is the minimum containment region. The minimum containment region is used to characterize the minimum region that contains all the target convex hull endpoints.

[0046] The calculation module is used to calculate the planar distance based on the target containment plane parameters to obtain the target flatness of the target object.

[0047] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0048] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0049] This application proposes a real-time measurement method for tile flatness based on a minimum containment region. It uses a multi-dimensional scanner to scan the surface of a target object, obtaining a raw scan dataset containing multiple raw point cloud data. The raw point cloud data in the raw scan dataset reflects the complete surface information of the target object, thus achieving comprehensive detection of the surface flatness of the target object under test. Specifically, a set of convex hull endpoints containing multiple target convex hull endpoints is obtained based on the raw point cloud data. This application simplifies computation by extracting convex hull endpoints from the raw point cloud data and then performing subsequent operations on these endpoints, thereby improving the efficiency of flatness detection. A containment region detection model is used to detect the containment region of the convex hull endpoint set, obtaining target containment plane parameters for constructing parallel first and second target containment planes. The region between the first and second target containment planes is the minimum containment region, which is the smallest region containing all target convex hull endpoints. Based on the target containment plane parameters, planar distances are calculated to obtain the target flatness of the target object. Therefore, this application transforms the target flatness into the distance between the parallel first target containment plane and the second target containment plane, and uses the containment region detection model to obtain the target containment plane parameters, thereby calculating the target flatness based on the target containment plane parameters, which can effectively improve the accuracy of flatness detection. Attached Figure Description

[0050] Figure 1 This is a flowchart of a method for real-time measurement of tile flatness based on a minimum containment area, provided in an embodiment of this application.

[0051] Figure 2 yes Figure 1 The flowchart of step S101 in the text;

[0052] Figure 3 yes Figure 1 The flowchart of step S102 in the document;

[0053] Figure 4 yes Figure 1 The flowchart of step S103 in the process;

[0054] Figure 5 yes Figure 4 The flowchart of step S403 in the process;

[0055] Figure 6 yes Figure 4 The flowchart of step S404 in the document;

[0056] Figure 7 yes Figure 6 The flowchart of step S605 in the process;

[0057] Figure 8 This is a schematic diagram of a method for real-time measurement of tile flatness based on a minimum containment area, provided in an embodiment of this application.

[0058] Figure 9 This is a schematic diagram of a real-time tile flatness measurement device based on a minimum containment area, provided in an embodiment of this application.

[0059] Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0061] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0063] First, let's analyze some of the terms used in this application:

[0064] Point cloud: A collection of discrete points in space. These discrete points are typically acquired using multidimensional scanning equipment (such as laser scanners and stereo cameras in 3D scanning equipment), and the acquired point cloud data can be used to construct digital elevation models (DEMs), 3D models, and perform spatial analysis.

[0065] A convex hull is the smallest convex polygon or polyhedron that contains all points in a given set of points. In other words, a convex hull is the smallest convex envelope of a point set. Convex hulls have wide applications in computer graphics, geographic information systems, pattern recognition, and other fields. Methods for solving convex hulls include Graham's scan method, Jarvis's step method, and QuickHull's method.

[0066] Support Vector Machine (SVM) is a machine learning algorithm commonly used in classification and regression analysis. The basic principle of SVM is to find an optimal hyperplane to effectively separate data points of different classes.

[0067] Flatness testing refers to the process of measuring and evaluating the flatness of an object's surface. In engineering, manufacturing, and construction, flatness testing is used to inspect the surface flatness of objects, and surface flatness is a crucial quality indicator that directly affects product quality and appearance. Therefore, surface flatness testing is a vital step in the quality inspection of large industrial products such as ceramic tiles and steel strips.

[0068] However, current flatness testing technologies typically employ manual methods. For example, the tile industry primarily relies on manual inspection to measure flatness. Manual inspection is inefficient, suitable only for random sampling of tiles, and cannot achieve comprehensive testing of all products. Furthermore, manual inspection often relies on feeler gauges, which can only measure flatness in a few fixed directions, failing to capture complete surface information and potentially leading to incomplete and inaccurate flatness measurements. Based on this, existing technologies have proposed using laser displacement sensors to replace manual inspection. However, this sensor acquisition method is also easily limited by fixed locations, unable to comprehensively inspect the entire tile surface, thus affecting the accuracy of the final flatness measurement. Therefore, how to achieve comprehensive flatness testing of the surface of the object being measured, and improve the accuracy and efficiency of flatness testing, has become an urgent technical problem to be solved.

[0069] Based on this, the embodiments of this application provide a method for real-time measurement of tile flatness based on the minimum containment area, which can realize comprehensive detection of the flatness of the surface of the object to be measured, thereby improving the accuracy and efficiency of flatness detection.

[0070] This application provides a method for real-time measurement of tile flatness based on a minimum containment area, relating to the field of flatness measurement technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or 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 communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing a method for real-time measurement of tile flatness based on a minimum containment area, but is not limited to the above forms.

[0071] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This 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. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0072] Figure 1 This is an optional flowchart of a method for real-time measurement of tile flatness based on a minimum containment area, provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S104.

[0073] Step S101: Obtain the original scan dataset of the target object. The original scan dataset includes original point cloud data. The original scan dataset is used to represent a set of multiple original point cloud data after scanning the surface of the target object based on a preset multidimensional scanner.

[0074] Step S102: Extract convex hull endpoints based on the original point cloud data to obtain a convex hull endpoint set. The convex hull endpoint set contains multiple target convex hull endpoints and is a subset of the original scan dataset.

[0075] Step S103: Based on the containment region detection model, the containment region of the convex hull endpoint set is detected to obtain the target containment plane parameters. The target containment plane parameters are used to construct the first target containment plane and the second target containment plane. The first target containment plane and the second target containment plane are two parallel planes, and the area between the first target containment plane and the second target containment plane is the minimum containment region. The minimum containment region is used to characterize the minimum region that contains all the target convex hull endpoints.

[0076] Step S104: Calculate the planar distance based on the target containment plane parameters to obtain the target flatness of the target object.

[0077] In step S101 of some embodiments, the target object refers to the object whose surface flatness is to be detected, and the target object has a surface to be scanned. In different application scenarios, the target object can be a tile, steel strip, etc., and is not specifically limited here. A multi-dimensional scanner refers to a multi-dimensional device used to scan the surface of a target object, such as a 3D scanner, a 4D scanner, etc., and is not specifically limited here. Surface point cloud data refers to the point cloud data obtained after scanning the target object. The original scan dataset is used to characterize the set of multiple original point cloud data obtained after scanning the surface of the target object based on a preset multi-dimensional scanner. Therefore, compared with related technologies that can only measure the flatness of tiles in a few fixed directions, and whose sensor acquisition methods are easily limited by fixed positions, this application scans the surface of the target object based on a multi-dimensional scanner, which can obtain more complete surface information of the object, thereby improving the accuracy of subsequent flatness detection.

[0078] Please see Figure 2 In some embodiments, step S101 may include, but is not limited to, steps S201 to S203:

[0079] Step S201: Scan the target object using a multi-dimensional scanner to obtain surface point cloud data;

[0080] Step S202: Obtain preset teleportation point cloud data. The preset teleportation point cloud data is used to characterize the point cloud data of the teleportation platform where the target object is placed.

[0081] Step S203: Select data from the surface point cloud data according to the preset transmitted point cloud data to obtain the original point cloud data, and construct the original scan dataset based on the original point cloud data.

[0082] In step S201 of some embodiments, when scanning the target object using a multi-dimensional scanner, it is first necessary to set the parameters of the multi-dimensional scanner, including scanning resolution, scanning range, and scanning speed. The selection of these parameters affects the accuracy and efficiency of the scan. The multi-dimensional scanner is positioned at an appropriate location and angle to ensure that the entire surface of the target object can be scanned completely and accurately. Then, the multi-dimensional scanner is started to begin scanning the target object. At this time, the multi-dimensional scanner emits lasers or other sensors to scan the surface of the target object and records the position and features of each scan point. In this way, the point cloud data of each scan point collected by the multi-dimensional scanner is the surface point cloud data of this application, that is, multiple surface point cloud data can be obtained by scanning the target object using a multi-dimensional scanner.

[0083] In step S202 of some embodiments, the transport platform is a platform for placing the target object and capable of transporting the target object from one place to another. Therefore, it is necessary to acquire preset transport point cloud data, which characterizes the point cloud data corresponding to the transport platform where the target object is placed. Furthermore, this preset transport point cloud data can be obtained by scanning the transport platform using a multi-dimensional scanner.

[0084] In step S203 of some embodiments, since the multi-dimensional scanner may also scan the transport platform when scanning the target object, the point cloud data of the transport platform may be mistaken for the point cloud data of the target object, leading to inaccurate flatness measurement results. Therefore, after obtaining multiple surface point cloud data, data selection can be performed on the surface point cloud data based on preset transport point cloud data. Specifically, preset transport point cloud data that affects the accuracy of flatness detection can be removed from the surface point cloud data. In this way, the original point cloud data refers to the point cloud data that only represents the complete surface of the target object.

[0085] It should be noted that data selection refers to the process of filtering and deleting preset transport point cloud data from surface point cloud data. Thus, the original point cloud data refers to the point cloud data remaining after filtering and deleting the preset transport point cloud data from the surface point cloud data, i.e., the original point cloud data used to characterize the point cloud data related to the surface flatness of the target object.

[0086] It should be noted that the raw point cloud data includes the position coordinates of the corresponding scanned points. For example, let the raw point cloud data be denoted as I, and let the raw point cloud data contain a total of N data points, I∈R. N×3 At this point, the three-dimensional coordinates of the i-th data point can be denoted as x. i , i = 1, 2, ..., N.

[0087] This embodiment uses a multi-dimensional scanner to scan the surface of the target object, obtaining multiple raw point cloud data containing all surface information of the target object. Compared to related technologies that can only measure flatness from a few angles, this embodiment obtains the raw point cloud data of the target object by filtering and deleting preset transmitted point cloud data from the surface point cloud data, enabling comprehensive detection of the flatness of the object's surface. In other words, in the actual flatness detection of large industrial products, this application can obtain the surface information of large industrial products through a multi-dimensional scanner, achieving more accurate results than manual measurement of surface flatness. Furthermore, the large-scale point cloud data obtained from large industrial products often contains millions of points. In order to enable the production line of these industrial products to achieve real-time flatness detection, an algorithm is needed to quickly and accurately calculate flatness from the large-scale point cloud data. Therefore, combined with the model constructed in this application, large-scale point cloud data of the target object can be obtained quickly and accurately in real time, resulting in a more accurate target flatness. For example, when the target object is a tile, a structured light-based multi-dimensional scanner can be used to obtain the complete surface information of the tile. The obtained raw point cloud data is a large-scale point cloud. Processing large-scale point clouds requires corresponding flatness measurement methods to achieve fast and accurate estimation of flatness based on point cloud data.

[0088] In step S102 of some embodiments, the target convex hull endpoint refers to the convex hull endpoint obtained by extracting convex hull endpoints from the original point cloud data. The convex hull endpoint set refers to a collection of multiple target convex hull endpoints.

[0089] Please see Figure 3 In some embodiments, step S102 may include, but is not limited to, steps S301 to S304:

[0090] Step S301: Randomly obtain the first candidate point cloud data and the second candidate point cloud data from the original point cloud data;

[0091] Step S302: Construct a candidate point cloud structure based on the first candidate point cloud data. The candidate point cloud structure includes candidate structure surfaces.

[0092] Step S303: Select point cloud data from the second candidate point cloud data based on the candidate structure surface to obtain the target convex hull endpoint. The target convex hull endpoint is used to characterize the second candidate point cloud data that is farthest from the candidate structure surface.

[0093] Step S304: Construct a convex hull endpoint set based on multiple target convex hull endpoints.

[0094] In step S301 of some embodiments, the first candidate point cloud data refers to the original point cloud data used to construct the candidate point cloud structure. The second candidate point cloud data refers to the original point cloud data used to determine the endpoints of the target convex hull. The second candidate point cloud data can be any original point cloud data in the original scan dataset, or any original point cloud data remaining after removing the first candidate point cloud data from the original scan dataset. Therefore, both the first and second candidate point cloud data are subsets of the original point cloud data.

[0095] In step S302 of some embodiments, the candidate point cloud structure is a structure formed by connecting the points corresponding to the first candidate point cloud data. A candidate structure face is any face that makes up the candidate point cloud structure. For example, if the number of first candidate point cloud data obtained is 4, then a cone-shaped candidate point cloud structure can be constructed based on these 4 first candidate point cloud data (i.e., four arbitrary original point cloud data). In this case, the candidate point cloud structure contains four candidate structure faces, and each candidate structure face is triangular in shape.

[0096] In step S303 of some embodiments, point cloud data selection refers to the process of selecting the point cloud data that is furthest from the candidate structural surface from the second candidate point cloud data. The selected second candidate point cloud data is then used as the endpoint of the target convex hull.

[0097] It should be noted that the point cloud data farthest from the candidate structure surface may be the first candidate point cloud data that makes up the candidate point cloud structure. In this case, the second candidate point cloud data includes the first candidate point cloud data.

[0098] In step S304 of some embodiments, the endpoint connection line refers to the connection line connecting different target convex hull endpoints. Therefore, by connecting different target convex hull endpoints in the convex hull endpoint set using endpoint connection lines, this application can obtain the convex hull of the target object, which is a shell-shaped structure representing the outermost point of the target object. For example, the convex hull endpoint set includes N h The target convex hull endpoints will be N h By connecting the endpoints of the target convex hull with endpoint connectors, the convex hull of the target object can be obtained, denoted as .

[0099] It should be noted that a target convex hull endpoint can be obtained through steps S301 to S304. In practical applications, after obtaining a target convex hull endpoint, by updating the combination of the first candidate point cloud data and the second candidate point cloud data obtained from the original point cloud data each time, and repeating steps S302 to S304, all target convex hull endpoints can be obtained. Each time a new target convex hull endpoint is found, the old target convex hull endpoints within the current candidate point cloud structure can be deleted to avoid the data already identified as target convex hull endpoints affecting the efficiency of point cloud data selection. After determining multiple target convex hull endpoints, all target convex hull endpoints can be connected by endpoint connection lines to obtain the convex hull.

[0100] In one specific embodiment, the embodiments of this application can calculate the convex hull of the original point cloud data according to a set convex hull solving method (such as the QuickHull algorithm). Specifically, firstly, any four first candidate point cloud data are selected from the original point cloud data. Then, a candidate point cloud structure with a cone structure can be constructed based on the points corresponding to these four first candidate point cloud data. Then, for each candidate structural face of the constructed candidate point cloud structure, the point farthest from the candidate structural face is selected as the latest target convex hull endpoint. When a new target convex hull endpoint is found, if an old target convex hull endpoint is inside the candidate point cloud structure, the old target convex hull endpoint is deleted. This process is iterated until no new target convex hull endpoint can be found based on each candidate structural face. After obtaining all the target convex hull endpoints, the convex hull of the target object can be obtained.

[0101] It should be noted that since the endpoint of the target convex hull is used to represent the second candidate point cloud data that is farthest from the candidate structure surface, the selected second candidate point cloud data can be outside the candidate point cloud structure, on the endpoint connection line of the constructed candidate point cloud structure, or the first candidate point cloud data of the constructed candidate point cloud structure. It can be flexibly set according to actual needs, and no specific limitation is made here.

[0102] This embodiment, through the above steps, first identifies the outermost points in the original point cloud data, which are the points furthest from each candidate structural surface. The selected second candidate point cloud data is then used as the target convex hull endpoint. The connection lines between the convex hull endpoint set and the target convex hull endpoint form the convex hull, which is the outermost structure of the original point cloud data. Because the original point cloud data is large-scale, but the minimum containment region method algorithms in related technologies are designed for small-scale point clouds obtained by contact probes, directly applying them to the large-scale point clouds of this application would result in excessively long computation times. Therefore, this embodiment introduces the concept of a "convex hull" into point cloud data processing. First, a fast convex hull algorithm is applied to obtain the convex hull of the original point cloud data, and then the target convex hull endpoints are used for subsequent calculations. This greatly simplifies the computation and enables rapid and accurate determination of flatness based on the original point cloud data.

[0103] In step S103 of some embodiments, the target containing plane parameters are the parameters used to construct the first and second target containing planes. The minimum containing region is the region including the first and second initial containing planes. The containing region detection model can be an improved support vector machine model, or other machine learning models or deep learning models, etc., without specific limitations here. Containing region detection refers to using the containing region detection model to calculate and obtain the target containing plane parameters. In other words, embodiments of this application use an improved support vector machine model to obtain the parameters of the two planes that make up the minimum containing region.

[0104] It should be noted that the Improved Vector Machine Model (ISVM) typically refers to a model that optimizes and improves upon the traditional Support Vector Machine (SVM) model. The improved vector machine model achieves classification by finding a hyperplane in a high-dimensional space that maximizes the margin between two classes.

[0105] Please see Figure 4 In some embodiments, step S103 may include, but is not limited to, steps S401 to S404:

[0106] Step S401: Obtain the initial bounding plane parameters and the preset number of iterations. The initial bounding plane parameters are used to construct the first initial bounding plane and the second initial bounding plane. The first initial bounding plane and the second initial bounding plane are two parallel planes.

[0107] Step S402: Obtain the point cloud coordinates of the endpoint of the target convex hull, and get the point cloud coordinates of the endpoint of the convex hull;

[0108] Step S403: Based on the preset number of iterations, initial bounding plane parameters, and convex hull endpoint point cloud coordinates, a function is constructed to obtain the target optimization function;

[0109] Step S404: Perform function calculation on the target optimization function to determine the target bounding plane parameters.

[0110] In step S401 of some embodiments, the initial containment plane parameters are the parameters used to construct the first and second initial containment planes. The preset number of iterations is a pre-set number of iterations.

[0111] In step S402 of some embodiments, the point cloud coordinates of the convex hull endpoint are the point cloud coordinates corresponding to the target convex hull endpoint. Since the target convex hull endpoint is a subset of the original point cloud data, the point cloud coordinates of the o-th convex hull endpoint are denoted as x. o o = 1, 2, ..., N h .

[0112] In step S403 of some embodiments, the target optimization function is a function constructed based on a preset number of iterations, initial bounding plane parameters, and the coordinates of the endpoint point cloud of the convex hull. The initial bounding plane parameters include initial normal vector sub-parameters and initial origin distance sub-parameters. Specifically, if the initial normal vector sub-parameters are denoted as A, B, and C, and the initial origin distance sub-parameters are denoted as D1 and D2, then the initial bounding plane parameters include A, B, C, D1, and D2. Therefore, if the first initial bounding plane is denoted as H1, then H1 can be expressed as Ax + By + Cz + D1 = 0; if the second initial bounding plane is denoted as H2, then H2 can be expressed as Ax + By + Cz + D2 = 0. D1 is not equal to D2.

[0113] This embodiment first constructs two initial parallel planes, namely a first initial enclosing plane and a second initial enclosing plane. By solving for the initial enclosing plane parameters of the first and second initial enclosing planes, the distance between the first and second initial enclosing planes is minimized. At this point, the first and second initial enclosing planes form the minimum enclosing region. This embodiment cleverly transforms the calculation of the surface smoothness of the target object into the calculation of the distance between the first and second initial enclosing planes through the above steps. This embodiment transforms the calculation of smoothness into the distance between two parallel planes. By setting the initial enclosing plane parameters, constructing a target optimization function, and calculating the target enclosing plane parameters based on the target optimization function, the distance between the first and second target enclosing planes constructed based on the target enclosing plane parameters is then the surface smoothness of the target object.

[0114] Please see Figure 5 In some embodiments, step S403 may also include, but is not limited to, steps S501 to S508:

[0115] Step S501: Construct an initial objective sub-function based on the initial normal vector sub-parameters;

[0116] Step S502: Construct the initial condition sub-function based on the initial normal vector sub-parameter, the coordinates of the convex hull endpoint point cloud, and the initial origin distance sub-parameter;

[0117] Step S503: Construct the initial objective function based on the initial objective sub-function and the initial condition sub-function;

[0118] Step S504: Calculate the gradient of the initial normal vector sub-parameters based on a preset number of iterations to obtain the normal vector gradient sub-parameters;

[0119] Step S505: Calculate the gradient of the initial origin distance sub-parameter based on the preset number of iterations to obtain the origin distance gradient sub-parameter;

[0120] Step S506: Construct the optimization objective sub-function based on the normal vector gradient sub-parameters;

[0121] Step S507: Construct the optimization condition sub-function based on the normal vector gradient sub-parameter, the coordinates of the convex hull endpoint point cloud, and the origin distance gradient sub-parameter;

[0122] Step S508: Construct the objective optimization function based on the objective sub-function and the optimization condition sub-function.

[0123] In step S501 of some embodiments, the initial target sub-function is a function constructed based on the initial normal vector sub-parameters.

[0124] In one embodiment, the initial objective function can be the maximum squared value of the initial normal vector sub-parameters. In this case, the initial objective function can be expressed as max(A) 2 +B 2 +C 2 ).

[0125] In another embodiment, to facilitate numerical comparison, the initial objective sub-function can be a function obtained by weighting the maximum squared values ​​of the initial normal vector sub-parameters. In this case, the initial objective sub-function can be (A 2 +B 2 +C 2 The initial objective function is multiplied by a coefficient to ensure the calculated result meets expectations. This coefficient can be flexibly adjusted based on actual circumstances. For example, if the weighting coefficient of the initial objective function is set to 1 / 2, then the initial objective function can be expressed as:

[0126] In step S502 of some embodiments, the initial condition sub-function is a condition function of the initial objective sub-function. The initial condition sub-function is a condition function constructed based on the initial normal vector sub-parameter, the coordinates of the convex hull endpoint point cloud, and the initial origin distance sub-parameter. Specifically, "st" is an abbreviation for "such that," used to represent a certain condition or restriction. Therefore, the initial condition sub-function can be expressed as the following formula (1):

[0127]

[0128] In step S503 of some embodiments, the initial objective function is a function constructed based on the initial objective subfunction and the initial condition subfunction. Therefore, combining with the above formula (1), the initial objective function can be expressed as shown in the following formula (2):

[0129] max(A 2 +B 2 +C 2 )

[0130]

[0131] Following step S503 in some embodiments, this application can perform function optimization based on the gradient descent algorithm, a preset number of iterations, initial normal vector sub-parameters, initial origin distance sub-parameters, and convex hull endpoint point cloud coordinates to obtain the target optimization function. The specific calculation process is described in steps S504 to S508 below.

[0132] In step S504 of some embodiments, the normal vector gradient sub-parameter is a parameter obtained by gradient calculation based on a preset number of iterations and the initial normal vector sub-parameter. For example, the preset number of iterations can be denoted as k, and the initial normal vector sub-parameter of the k-th iteration can be denoted as A. k B k C k In each iteration, the gradients of the initial normal vector subparameters can be denoted as ΔA, ΔB, and ΔC.

[0133] In step S505 of some embodiments, the origin distance gradient sub-parameter refers to the parameter obtained by gradient calculation based on a preset number of iterations and an initial origin distance sub-parameter. Assume the initial origin distance sub-parameter for the k-th iteration is D. k1 and D k2 The gradient of the initial origin distance sub-parameter, calculated in each iteration, can be denoted as ΔD. k1 and ΔD k2 .

[0134] In step S506 of some embodiments, the optimization objective function is a function constructed based on the normal vector gradient sub-parameters. The optimization objective function can be expressed as the following formula (3):

[0135]

[0136] In step S507 of some embodiments, the optimization condition sub-function refers to a condition function constructed based on the normal vector gradient sub-parameter, the coordinates of the convex hull endpoint point cloud, and the origin distance gradient sub-parameter. The optimization condition sub-function can be expressed as the following formula (4):

[0137]

[0138] In step S508 of some embodiments, the objective optimization function refers to a function constructed based on the objective sub-function and the condition sub-function. Therefore, combining the above formulas (3) and (4), the objective optimization function can be expressed as the following formula (5):

[0139]

[0140]

[0141] It should be noted that this objective function can be transformed into a linear programming problem, which can be solved using the interior point method.

[0142] In step S404 of some embodiments, function calculation refers to calculating the target optimization function corresponding to the above formula (5) to obtain the target containment plane parameters that meet the requirements.

[0143] Please see Figure 6 In some embodiments, step S404 includes, but is not limited to, steps S601 to S605:

[0144] Step S601: For the current iteration number, perform function calculation on the target optimization function to obtain the candidate iteration bounding plane parameters;

[0145] Step S602: Obtain adjacent iteration bounding plane parameters based on the preset number of iterations and candidate iteration bounding plane parameters. The adjacent iteration bounding plane parameters are used to characterize the candidate iteration bounding plane parameters obtained in the next iteration number after the current iteration number.

[0146] Step S603: Calculate the flatness difference based on the candidate iteration bounding plane parameters corresponding to the current iteration number and the adjacent iteration bounding plane parameters to obtain the adjacent flatness difference.

[0147] Step S604: When the adjacent smoothness difference of the current iteration number is less than the preset smoothness difference threshold, the adjacent smoothness difference is used as a candidate smoothness difference.

[0148] Step S605: Determine the target containment plane parameters based on the candidate flatness difference.

[0149] In step S601 of some embodiments, the current iteration number is the number of the current iteration number. The candidate iteration encompassing plane parameters are parameters obtained by function calculation based on the objective optimization function. Assuming the current iteration number is k, the candidate iteration encompassing plane parameters can be expressed as the following formula (6):

[0150]

[0151] In step S602 of some embodiments, the adjacent iteration encompassing plane parameter refers to the candidate iteration encompassing plane parameter obtained corresponding to the next iteration number of the current iteration number, which is obtained based on a preset number of iterations and the candidate iteration encompassing plane parameters. For example, if the current iteration number can be k, then the next iteration number is k+1. Because D k2 D k1 D is a constant term. k2 -D k1 The value of D does not change with the iteration number. Therefore, we can assume that D... k2 -D k1 =2. The adjacent iteration bounding plane parameter can be expressed as the following formula (7):

[0152]

[0153] It should be noted that the last iteration is the case where there is no next iteration. In the case of the last iteration, the bounding plane parameter of the adjacent iteration can be recorded as 0.

[0154] In step S603 of some embodiments, the adjacent smoothness difference refers to the difference calculated based on the candidate iteration bounding plane parameter corresponding to the current iteration number and the adjacent iteration bounding plane parameter. The adjacent smoothness difference is denoted as δ, and the calculation process can be specifically expressed as the following formula (8):

[0155]

[0156] In step S604 of some embodiments, the smoothness difference threshold refers to a preset value. A candidate smoothness difference is the adjacent smoothness difference δ of the current iteration number that is less than the preset smoothness difference threshold.

[0157] In this embodiment, the model parameters of two consecutive iterations are iterated until the difference between adjacent smoothness values ​​is less than a preset smoothness difference threshold. This difference between adjacent smoothness values ​​is then used as a candidate smoothness difference value.

[0158] In step S605 of some embodiments, the target encompassing plane parameter is a plane parameter determined based on the candidate flatness difference.

[0159] Please see Figure 7 In some embodiments, step S605 may include, but is not limited to, steps S701 to S702:

[0160] Step S701: Select the candidate smoothness difference with the smallest value among multiple candidate smoothness differences as the target smoothness difference.

[0161] Step S702: The candidate iterative bounding plane parameters corresponding to the target flatness difference are used as the target bounding plane parameters.

[0162] In step S701 of some embodiments, the target smoothness difference is the smallest candidate smoothness difference among multiple candidate smoothness differences. Specifically, the candidate iteration containment plane parameter for each current iteration number and the adjacent iteration containment plane parameter for the next iteration number are calculated respectively. Multiple adjacent smoothness differences whose adjacent smoothness differences obtained based on the candidate iteration containment plane parameters and adjacent iteration containment plane parameters are less than a preset smoothness difference threshold are used as multiple candidate smoothness differences. At this time, the first initial containment plane is the first target containment plane, and the second initial containment plane is the second target containment plane.

[0163] In one embodiment, the adjacent smoothness difference corresponding to each iteration number is obtained. Assuming that when the current iteration number is k, the adjacent smoothness difference δ is less than a preset smoothness difference threshold, and the adjacent smoothness difference corresponding to the current iteration number k is the smallest, then the adjacent smoothness difference when the current iteration number k is taken as the target smoothness difference.

[0164] In another embodiment, when the first occurrence of an adjacent smoothness difference δ is less than a preset smoothness difference threshold, the candidate smoothness difference of that occurrence is taken as the target smoothness difference.

[0165] In step S702 of some embodiments, the target enclosing plane parameter refers to the parameters of the final first initial enclosing plane and the second initial enclosing plane determined based on the candidate smoothness difference δ. For example, if the current iteration number corresponding to the target smoothness difference is k, the target enclosing plane parameter can be A. k B k C k D k1 and D k2 .

[0166] In step S104 of some embodiments, the target flatness refers to the data obtained by calculating the planar distance based on the target encompassing plane parameters. The target flatness can be denoted as d. The planar distance calculation refers to calculating the target flatness according to the definition of the minimum encompassing area method, that is, the distance between the first target encompassing plane and the second target encompassing plane is the target flatness d. Because D k2-D k1 =2, so the calculation process of the target flatness d can be expressed as the following formula (9):

[0167]

[0168] After step S104 in some embodiments, if the target flatness is greater than a certain value, it indicates that the target object has a flatness defect. At this time, a signal can be sent in a timely manner via SMS or to a reminder device to remind maintenance personnel to repair the equipment and solve the flatness defect problem.

[0169] This application transforms the target flatness into the distance between a parallel first target enclosing plane and a second target enclosing plane, and uses an enclosing region detection model to obtain the target enclosing plane parameters. The target flatness is then calculated based on these parameters, effectively improving the accuracy of flatness detection. Embodiments of this application can be used for flatness detection of ceramic tiles, improving the accuracy and speed of flatness detection during manufacturing, reducing scrap rates and the cost of manual inspection, and improving enterprise production quality.

[0170] In one specific embodiment, such as Figure 8 As shown, firstly, in this embodiment of the application, the target convex hull endpoints and the convex hull can be obtained through a pre-set fast convex hull algorithm. The fast convex hull algorithm corresponds to the convex hull endpoint extraction method in step S102 above. Specifically, the original point cloud data I∈R of the target object is first calculated using a fast convex hull algorithm (such as the QuickHull algorithm). N×3 Corresponding convex shell The fast convex hull algorithm first initializes by selecting any four points from the original point cloud data and constructing an initial tetrahedron using these four points. Then, for each face of the initial tetrahedron, the point furthest from that face on the outside of the tetrahedron is selected as the latest target convex hull endpoint. When a new target convex hull endpoint is found, if an old target convex hull endpoint exists inside the initial tetrahedron, the old target convex hull endpoint is deleted. This process is iterated until no new target convex hull endpoints can be found on any face of the initial tetrahedron. After obtaining all target convex hull endpoints, the convex hull of the target object can be obtained. Specific implementation details can be found in steps S301 to S304 above.

[0171] Secondly, after obtaining the convex hull of the target object, this embodiment constructs a minimum containment region plane for the endpoints of the target convex hull using an improved support vector machine model, thereby obtaining the minimum containment region plane. Specifically, this embodiment obtains the containment convex hull using an improved support vector machine model. The minimum containment area. This application embodiment considers N on the convex hull. hGiven several endpoints, find two parallel planes that can contain these endpoints: the first target containing plane and the second target containing plane. The first target containing plane and the second target containing plane constitute the minimum containing region plane. We can define D... k2 =D+1,D k1 =D-1. The first target encompassing plane can be denoted as H1: Ax+By+Cz+D-1=0, and the second target encompassing plane can be denoted as H2: Ax+By+Cz+D+1=0. Specific implementation methods can refer to steps S401 to S404 above.

[0172] Next, the flatness value, i.e., the target flatness of the target object, is calculated based on the obtained minimum containment area. This minimum containment area calculation corresponds to the planar distance calculation in step S104 above. Specifically, the flatness is calculated according to the definition of the minimum containment area method. The distance between the minimum containment area planes H1 and H2 is the flatness d: For specific implementation methods, please refer to step S104 above.

[0173] Please see Figure 9 This application also provides a real-time tile flatness measurement device based on a minimum containment area, which can realize the above-mentioned real-time tile flatness measurement method based on a minimum containment area. The device includes:

[0174] The acquisition module is used to acquire the original scan dataset of the target object. The original scan dataset includes original point cloud data. The original scan dataset is used to represent a set of multiple original point cloud data after scanning the surface of the target object based on a preset multidimensional scanner.

[0175] The extraction module is used to extract convex hull endpoints based on the original point cloud data, and obtain a convex hull endpoint set. The convex hull endpoint set contains multiple target convex hull endpoints and is a subset of the original scan dataset.

[0176] The detection module is used to perform containment region detection on the set of convex hull endpoints based on the containment region detection model to obtain target containment plane parameters. The target containment plane parameters are used to construct the first target containment plane and the second target containment plane. The first target containment plane and the second target containment plane are two parallel planes, and the area between the first target containment plane and the second target containment plane is the minimum containment region. The minimum containment region is used to characterize the minimum region that contains all target convex hull endpoints.

[0177] The calculation module is used to calculate the planar distance based on the containment plane parameters to obtain the target flatness of the target object.

[0178] The specific implementation of this real-time tile flatness measurement device based on the minimum containment area is basically the same as the specific embodiment of the real-time tile flatness measurement method based on the minimum containment area described above, and will not be repeated here.

[0179] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for real-time measurement of tile flatness based on a minimum containment area. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0180] Please see Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0181] The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0182] The memory 1002 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1002 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and called and executed by the processor 1001 to execute the real-time measurement method for tile flatness based on the minimum containment area of ​​this application embodiment.

[0183] Input / output interface 1003 is used to implement information input and output;

[0184] The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0185] Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004);

[0186] The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0187] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for real-time measurement of tile flatness based on a minimum containment area.

[0188] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0189] This application provides a real-time measurement method for tile flatness based on the minimum containment region. It obtains an original scan dataset containing multiple original point cloud data by scanning the surface of a target object using a multi-dimensional scanner. The original point cloud data in the original scan dataset reflects the complete surface information of the target object, thus achieving comprehensive detection of the surface flatness of the target object. Specifically, a set of convex hull endpoints containing multiple target convex hull endpoints is obtained based on the original point cloud data. This application simplifies the computation by extracting convex hull endpoints from the original point cloud data and then performing subsequent operations on these endpoints. In other words, this application first utilizes convex hull theory to reduce the computational load when using the minimum containment region method to obtain flatness, thereby improving the efficiency of flatness detection. The containment region detection model is used to detect the containment region of the convex hull endpoint set, obtaining target containment plane parameters for constructing parallel first and second target containment planes. The region between the first and second target containment planes is the minimum containment region, which is the smallest region containing all target convex hull endpoints. The target flatness of the target object is obtained by calculating the planar distance based on the target containment plane parameters. Therefore, this application transforms the target flatness into the distance between a parallel first target enclosing plane and a second target enclosing plane, and uses an enclosing region detection model to obtain the target enclosing plane parameters, thereby calculating the target flatness based on the target enclosing plane parameters. In other words, embodiments of this application can effectively improve the accuracy of flatness detection by combining the fast convex hull method and the enclosing region detection model to jointly determine the large-scale point cloud data of the target object (such as a tile).

[0190] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0191] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0192] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0193] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0194] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0195] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: 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.

[0196] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0197] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0198] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0199] If the integrated unit is implemented as 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 this application, in essence, or the part that contributes to the prior art, or all or part of the 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 to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0200] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for real-time measurement of tile flatness based on a minimum containment area, characterized in that, The method includes: Obtain the original scan dataset of the target object, which includes original point cloud data. The original scan dataset is used to represent a set of multiple original point cloud data after scanning the surface of the target object based on a preset multidimensional scanner. Convex hull endpoints are extracted based on the original point cloud data to obtain a convex hull endpoint set, which contains multiple target convex hull endpoints and is a subset of the original scan dataset. The target convex hull endpoint set is subjected to convex hull endpoint detection based on the convex hull endpoint detection model to obtain target convex hull endpoint parameters. The target convex hull endpoint parameters are used to construct a first target convex hull endpoint and a second target convex hull endpoint. The first target convex hull endpoint and the second target convex hull endpoint are two parallel planes, and the region between the first target convex hull endpoint and the second target convex hull endpoint is the minimum convex hull endpoint. The minimum convex hull endpoint is used to characterize the minimum region that contains all the target convex hull endpoints. The flatness of the target object is obtained by calculating the planar distance based on the target containment plane parameters. The step of performing containment region detection on the convex hull endpoint set based on the containment region detection model to obtain target containment plane parameters includes: Obtain initial bounding plane parameters and a preset number of iterations. The initial bounding plane parameters are used to construct a first initial bounding plane and a second initial bounding plane, which are two parallel planes. Obtain the point cloud coordinates of the endpoint of the target convex hull to obtain the point cloud coordinates of the convex hull endpoint; Based on the preset number of iterations, the initial containment plane parameters, and the point cloud coordinates of the convex hull endpoints, a function is constructed to obtain the target optimization function; The target optimization function is calculated to determine the parameters of the target encompassing plane; The step of performing function calculation on the target optimization function to determine the target encompassing plane parameters includes: For the current iteration number, perform function calculation on the target optimization function to obtain the candidate iteration bounding plane parameters; Based on the preset number of iterations and the candidate iteration containment plane parameters, adjacent iteration containment plane parameters are obtained. The adjacent iteration containment plane parameters are used to characterize the candidate iteration containment plane parameters obtained in the next iteration number after the current iteration number. The flatness difference is calculated based on the candidate iteration bounding plane parameter corresponding to the current iteration number and the adjacent iteration bounding plane parameter to obtain the adjacent flatness difference; When the adjacent smoothness difference of the current iteration number is less than the preset smoothness difference threshold, the adjacent smoothness difference is used as a candidate smoothness difference. The target containment plane parameters are determined based on the candidate flatness difference.

2. The method according to claim 1, characterized in that, The extraction of convex hull endpoints based on the original point cloud data yields a set of convex hull endpoints, including: Randomly obtain first candidate point cloud data and second candidate point cloud data from the original point cloud data; A candidate point cloud structure is constructed based on the first candidate point cloud data, and the candidate point cloud structure includes candidate structure surfaces; Based on the candidate structural surface, point cloud data is selected from the second candidate point cloud data to obtain the target convex hull endpoint. The target convex hull endpoint is used to characterize the second candidate point cloud data that is farthest from the candidate structural surface. The convex hull endpoint set is constructed based on multiple target convex hull endpoints.

3. The method according to claim 1, characterized in that, The initial bounding plane parameters include initial normal vector sub-parameters and initial origin distance sub-parameters. The function construction based on the preset number of iterations, the initial bounding plane parameters, and the coordinates of the convex hull endpoint point cloud yields the target optimization function, including: Construct an initial objective sub-function based on the initial normal vector sub-parameters; Based on the initial normal vector sub-parameters, the convex hull endpoint point cloud coordinates, and the initial origin distance sub-parameters, an initial condition sub-function is constructed; Construct an initial objective function based on the initial objective sub-function and the initial condition sub-function; The gradient of the initial normal vector sub-parameters is calculated based on the preset number of iterations to obtain the normal vector gradient sub-parameters. Based on the preset number of iterations, the gradient of the initial origin distance sub-parameter is calculated to obtain the origin distance gradient sub-parameter; Construct an optimization objective function based on the normal vector gradient sub-parameters; Based on the normal vector gradient sub-parameter, the convex hull endpoint point cloud coordinates, and the origin distance gradient sub-parameter, an optimization condition sub-function is constructed. The objective optimization function is constructed based on the objective sub-function and the condition sub-function.

4. The method according to claim 1, characterized in that, Determining the target containment plane parameters based on the candidate flatness differences includes: The smallest candidate smoothness difference among the multiple candidate smoothness differences is taken as the target smoothness difference; The candidate iterative containment plane parameter corresponding to the target flatness difference is used as the target containment plane parameter.

5. The method according to claim 1, characterized in that, The process of obtaining the original scan dataset of the target object includes: The target object is scanned by the multi-dimensional scanner to obtain surface point cloud data. Acquire preset transmission point cloud data, which is used to characterize the point cloud data of the transmission platform on which the target object is placed; The surface point cloud data is selected based on the preset transmitted point cloud data to obtain the original point cloud data, and the original scan dataset is constructed based on the original point cloud data.

6. A real-time measurement device for tile flatness based on a minimum containment area, characterized in that, The device includes: The acquisition module is used to acquire the original scan dataset of the target object. The original scan dataset includes original point cloud data. The original scan dataset is used to represent a set of multiple original point cloud data after scanning the surface of the target object based on a preset multidimensional scanner. The extraction module is used to extract convex hull endpoints based on the original point cloud data to obtain a convex hull endpoint set, which contains multiple target convex hull endpoints and is a subset of the original scan dataset. The detection module is used to perform containment region detection on the set of convex hull endpoints based on the containment region detection model to obtain target containment plane parameters. The target containment plane parameters are used to construct a first target containment plane and a second target containment plane. The first target containment plane and the second target containment plane are two parallel planes, and the area between the first target containment plane and the second containment plane is the minimum containment region. The minimum containment region is used to characterize the minimum region that contains all the target convex hull endpoints. The calculation module is used to calculate the planar distance based on the target containment plane parameters to obtain the target flatness of the target object; The device is further configured to obtain initial containment plane parameters and a preset number of iterations, wherein the initial containment plane parameters are used to construct a first initial containment plane and a second initial containment plane, the first initial containment plane and the second initial containment plane being two parallel planes; obtain the point cloud coordinates of the endpoint of the target convex hull, thereby obtaining the point cloud coordinates of the convex hull endpoint; and construct a function based on the preset number of iterations, the initial containment plane parameters and the point cloud coordinates of the convex hull endpoint, thereby obtaining the target optimization function; The target optimization function is calculated to determine the parameters of the target encompassing plane; The apparatus is further configured to: perform function calculation on the target optimization function for the current iteration number to obtain candidate iteration encompassing plane parameters; obtain adjacent iteration encompassing plane parameters based on the preset number of iterations and the candidate iteration encompassing plane parameters, wherein the adjacent iteration encompassing plane parameters are used to characterize the candidate iteration encompassing plane parameters obtained corresponding to the next iteration number of the current iteration number; calculate the smoothness difference based on the candidate iteration encompassing plane parameters corresponding to the current iteration number and the adjacent iteration encompassing plane parameters to obtain adjacent smoothness difference; when the adjacent smoothness difference of the current iteration number is less than a preset smoothness difference threshold, use the adjacent smoothness difference as a candidate smoothness difference; and determine the target encompassing plane parameters based on the candidate smoothness difference.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the method for real-time measurement of tile flatness based on the minimum containment area as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a method for real-time measurement of tile flatness based on a minimum containment area, as described in any one of claims 1 to 5.

Images