Method and device for measuring the twist of refractory bricks based on three-dimensional point cloud

Abstract

This application discloses a method for measuring the torsion of refractory bricks based on three-dimensional point clouds, belonging to the field of machine vision technology. The method includes: acquiring a three-dimensional point cloud of the surface of the refractory brick to be measured; preprocessing the three-dimensional point cloud to obtain a planar point cloud; establishing an object coordinate system based on principal component analysis; obtaining a target depth image and an index map based on the planar point cloud; performing X-shaped sampling on the target depth image within the measurement area, and combining it with the index map to obtain a sampled point cloud; and obtaining the torsion of the refractory brick based on the sampled point cloud using a torsion evaluation method. This method improves the efficiency and accuracy of refractory brick torsion measurement.

Description

Technical Field

[0001] This application belongs to the field of machine vision technology, and in particular relates to a method and device for measuring the torsion of refractory bricks based on three-dimensional point clouds. Background Technology

[0002] Refractory bricks are crucial high-temperature equipment materials in industries such as metallurgy, building materials, and chemicals. The precision of their shape and dimensions directly affects the service life and safety of these equipment. With the increasing demand for high-temperature materials in industrial equipment, refractory bricks often deform during production and use due to factors such as high temperatures and mechanical stress, particularly exhibiting torsional deformation. This phenomenon poses a threat to the safe operation of equipment. Therefore, the degree of torsion of refractory bricks has become a key indicator for evaluating their quality and performance.

[0003] Existing distortion detection methods directly project point cloud data onto a 2D plane to generate depth images, lacking effective management of the point cloud index, making it difficult to quickly locate the original point cloud data in subsequent processing. Furthermore, existing methods suffer from high memory consumption and computational complexity when processing large-scale point cloud data, resulting in low online detection efficiency. Simultaneously, due to the complex distortion deformation of refractory brick surfaces, there is currently no unified standard for surface distortion detection using 3D point clouds, making it difficult to guarantee the consistency and reliability of measurement results. Therefore, the application of related technologies in distortion measurement scenarios suffers from problems such as complexity, low efficiency, and low accuracy. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a method and apparatus for measuring the torsion of refractory bricks based on three-dimensional point clouds, which improves the efficiency and accuracy of refractory brick torsion measurement.

[0005] In a first aspect, this application provides a method for measuring the torsion of refractory bricks based on three-dimensional point clouds, the method comprising:

[0006] A three-dimensional point cloud is collected from the surface of the refractory brick to be tested, and the three-dimensional point cloud is preprocessed to obtain a planar point cloud.

[0007] Establish the object coordinate system based on principal component analysis;

[0008] Based on the planar point cloud, a target depth image and an index map are obtained;

[0009] X-shaped sampling is performed on the target depth image within the area to be tested, and the sampling point cloud is obtained by combining it with the index map;

[0010] Based on the sampling point cloud, the torsion of the refractory brick is obtained through a torsion evaluation method.

[0011] According to one embodiment of this application, obtaining the torsion of the refractory brick based on the sampled point cloud using a torsion evaluation method includes:

[0012] A plane fit is performed on the three-dimensional point cloud of the refractory brick surface to be tested to obtain a plane formula;

[0013] The spatial topological relationship of the point cloud is established based on the K-dimensional tree, and the centroid coordinates of the points in the sampled point cloud are obtained.

[0014] The z-coordinates of all sampling points in the sampling point cloud are obtained based on the aforementioned planar formula.

[0015] Calculate the relative z-coordinates of all sampling points in the sampling point cloud;

[0016] The torsion of the refractory brick is calculated based on the relative z-coordinates of all the sampling points.

[0017] According to one embodiment of this application, establishing the object coordinate system based on the principal component analysis method includes:

[0018] Based on the planar point cloud, construct a point cloud vector matrix;

[0019] The point cloud vector matrix is ​​decentered to obtain a decentralized point cloud vector matrix;

[0020] Based on the decentralized point cloud vector matrix, the covariance matrix is ​​obtained;

[0021] Based on the covariance matrix, the rotation matrix and homogeneous transformation matrix of the object coordinate system relative to the camera coordinate system are obtained;

[0022] Based on the rotation matrix and the homogeneous transformation matrix, an object coordinate system is established.

[0023] According to one embodiment of this application, obtaining the target depth image and index map based on the planar point cloud includes:

[0024] Traverse the planar point cloud to obtain the maximum and minimum values ​​of the planar point cloud in the X, Y, and Z directions;

[0025] Based on the maximum and minimum values ​​of the planar point cloud in the X, Y, and Z directions, the minimum bounding box of the planar point cloud is obtained.

[0026] Based on the resolution and range of the planar point cloud in the X and Y directions, an 8-bit depth image is allocated, and a 24-bit index image of the same size is generated simultaneously.

[0027] The coordinate values ​​of the planar point cloud in the X and Y directions are mapped to the pixel positions of the depth image, and the gray values ​​of the pixels in the depth image are calculated based on the coordinate values ​​of the planar point cloud in the Z direction to generate the target depth image.

[0028] Based on the position of the planar point cloud in the target depth image, the index of the planar point cloud is stored in the index map of the corresponding position.

[0029] According to one embodiment of this application, the step of performing X-shaped sampling on the target depth image within the region to be tested, and combining it with an index map to obtain a sampled point cloud, includes:

[0030] Subpixel edges of the target depth image are extracted based on an edge detection algorithm;

[0031] The sub-pixel edges are fitted to straight lines l1, l2, l3, and l4 using the weighted least squares method;

[0032] Based on the lines l1, l2, l3 and l4, calculate the intersection points P1, P2, P3 and P4 of adjacent lines;

[0033] The domain ReduceRegion of the target depth image is obtained based on the boundary cropping distance;

[0034] Connect the diagonal points P1, P2 and P3, P4 to obtain the linear regions LineRegion1 and LineRegion2;

[0035] Calculate the intersection region of LineRegion1, LineRegion2 and ReduceRegion, and store the pixel coordinates of the intersection region into the sampling point cloud pixel set;

[0036] Based on the pixel set of the sampled point cloud and the index map, the sampled point cloud is obtained.

[0037] According to one embodiment of this application, calculating the torsion of the refractory brick based on the relative z-coordinates of all the sampling points includes:

[0038] Iterate through the relative z-coordinates of all sampling points to obtain the maximum and minimum z-coordinate values;

[0039] Based on the maximum and minimum values ​​of the z-coordinate, the torsion of the refractory brick is calculated using the following formula:

[0040] f = d max -d min

[0041] Where f is the torsion of the refractory brick, and d max d is the maximum value of the z-coordinate. minIt is the minimum value of the z-coordinate.

[0042] According to one embodiment of this application, the preprocessing of the three-dimensional point cloud includes:

[0043] The three-dimensional point cloud is filtered using a pass-through filtering algorithm to obtain the three-dimensional point cloud of the target region.

[0044] Clustering and denoising are performed on the three-dimensional point cloud of the target region to obtain a planar point cloud.

[0045] Secondly, this application provides a refractory brick torsion measurement device based on three-dimensional point cloud, the device comprising:

[0046] The acquisition module is used to acquire a three-dimensional point cloud of the refractory brick surface to be tested, and to preprocess the three-dimensional point cloud to obtain a planar point cloud.

[0047] A module is created to establish an object coordinate system based on the principal component analysis method.

[0048] The first processing module is used to obtain a target depth image and an index map based on the planar point cloud;

[0049] The second processing module is used to perform X-shaped sampling on the target depth image within the area to be tested, and combine it with the index map to obtain the sampling point cloud;

[0050] The third processing module is used to obtain the torsion degree of the refractory brick based on the sampling point cloud using a torsion degree evaluation method.

[0051] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the refractory brick torsion measurement method based on three-dimensional point cloud as described in the first aspect above.

[0052] Fourthly, this application provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for measuring the torsion of refractory bricks based on three-dimensional point clouds as described in the first aspect above.

[0053] Fifthly, this application provides a chip, which includes a processor and a communication interface, the communication interface being coupled to the processor, and the processor being used to run programs or instructions to implement the method for measuring the torsion of refractory bricks based on three-dimensional point clouds as described in the first aspect.

[0054] In a sixth aspect, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method for measuring the torsion of refractory bricks based on three-dimensional point clouds as described in the first aspect above.

[0055] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application.

[0056] This invention provides a method for measuring the torsion of refractory bricks based on three-dimensional point clouds, which has the following advantages over existing technologies:

[0057] Beneficial effects:

[0058] (1) This invention obtains a planar point cloud by collecting three-dimensional point cloud data of the surface of refractory bricks and performing preprocessing. The principal component analysis method is used to establish the object coordinate system, eliminating the tilt error in the Z direction, which makes the subsequent depth image generation and distortion assessment more accurate. The target depth image and index map are generated by the planar point cloud. The two-dimensional depth map is sampled by X-shaped pixels, and the sampled point cloud is obtained by the index map. The distortion assessment method is used to analyze the sampled point cloud, which reduces the accuracy loss problem when generating point clouds from conventional depth images. The accuracy and efficiency of refractory brick distortion measurement are improved by efficient addressing access to the point cloud index, realizing the rapid and accurate measurement of refractory brick distortion and reducing the measurement complexity.

[0059] (2) This invention obtains a planar formula by performing planar fitting on a three-dimensional point cloud and establishes a spatial topological relationship of the point cloud based on a K-dimensional tree, which can obtain the centroid coordinates in the sampled point cloud. By calculating the z-coordinates of all sampled points based on the planar formula and further calculating the relative z-coordinates, the torsion of the refractory brick is calculated based on the relative z-coordinates of all sampled points. This effectively simulates the measurement process of the standard manual straightedge method, and can efficiently and accurately measure the torsion of the refractory brick, reducing measurement errors.

[0060] (3) This invention establishes an object coordinate system based on principal component analysis, which can align the object coordinate system with the camera coordinate system. By constructing a point cloud vector matrix, performing decentralization processing, calculating the covariance matrix, and obtaining the rotation matrix and homogeneous transformation matrix, the object coordinate system is established, eliminating the tilt error of the measurement plane in the Z direction. This makes the subsequent depth image generation and distortion assessment more accurate, improving the accuracy and efficiency of refractory brick distortion measurement. Attached Figure Description

[0061] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0062] Figure 1 This is one of the flowcharts illustrating the method for measuring the torsion of refractory bricks based on three-dimensional point clouds provided in the embodiments of this application;

[0063] Figure 2This is a schematic diagram of the structure for collecting measurements on the surface of refractory bricks provided in an embodiment of this application;

[0064] Figure 3 This is a schematic diagram of the measurement of refractory brick torsion provided in an embodiment of this application;

[0065] Figure 4 This is a schematic diagram of the establishment of the object coordinate system provided in the embodiments of this application;

[0066] Figure 5 This is a schematic diagram of depth image storage provided in an embodiment of this application;

[0067] Figure 6 This is a schematic diagram of X-shaped sampling provided in an embodiment of this application;

[0068] Figure 7 This is the second flowchart of the method for measuring the torsion of refractory bricks based on three-dimensional point clouds provided in the embodiments of this application;

[0069] Figure 8 This is a schematic diagram of the refractory brick torsion measurement device based on three-dimensional point cloud provided in the embodiments of this application;

[0070] Figure 9 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0071] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0072] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0073] The following description, in conjunction with the accompanying drawings, details the method for measuring the torsion of refractory bricks based on three-dimensional point clouds, the device for measuring the torsion of refractory bricks based on three-dimensional point clouds, the electronic equipment, and the readable storage medium provided in this application, through specific embodiments and application scenarios.

[0074] Among them, the method for measuring the torsion of refractory bricks based on three-dimensional point clouds can be applied to the terminal, specifically by the hardware or software in the terminal.

[0075] The terminal includes, but is not limited to, portable communication devices such as mobile phones or tablets with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads). It should also be understood that, in some embodiments, the terminal may not be a portable communication device, but rather a desktop computer with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads).

[0076] The following embodiments describe a terminal including a display and a touch-sensitive surface. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, mouse, and joystick.

[0077] The refractory brick torsion measurement method based on three-dimensional point cloud provided in this application embodiment can be executed by an electronic device or a functional module or entity in an electronic device that can implement the refractory brick torsion measurement method based on three-dimensional point cloud. The electronic devices mentioned in this application embodiment include, but are not limited to, mobile phones, tablets, computers, cameras and wearable devices. The following uses an electronic device as the execution subject to illustrate the refractory brick torsion measurement method based on three-dimensional point cloud provided in this application embodiment.

[0078] In industries such as metallurgy, building materials, and chemicals, refractory bricks are key materials for high-temperature equipment, and the accuracy of their shape and dimensions directly affects the service life and safety of the equipment. However, during production and use, refractory bricks are prone to twisting and deformation due to factors such as high temperature and mechanical stress. Therefore, the degree of twisting of refractory bricks is an important indicator in the quality evaluation of refractory bricks. Currently, the measurement of the twisting degree of refractory bricks mainly relies on manual sampling. The specific operation steps are as follows: place the refractory brick naturally on a refined flat plate, place a steel straightedge along the diagonal direction of the refractory brick surface, then find the position of the maximum gap between the steel straightedge and the surface of the refractory brick being measured, and gently insert a feeler gauge perpendicular to the steel straightedge until the feeler gauge can not be inserted any further. At this point, record the maximum thickness value of the feeler gauge insertion; this value represents the degree of twisting of the refractory brick. However, traditional manual measurement methods have problems such as low detection efficiency, susceptibility to subjective factors in detection results, and difficulty in automation, which cannot meet the urgent needs of modern industry for high-precision and high-efficiency detection.

[0079] With the development of 3D vision technology, 3D point cloud data provides a new solution for the quantitative assessment of refractory brick distortion. 3D point cloud data contains height information of the refractory brick surface, and distortion essentially reflects the inconsistency of this height information. Therefore, by analyzing the elevation information of point cloud data, accurate measurement of refractory brick distortion can be achieved. However, traditional methods typically project point cloud data directly onto a 2D plane to generate a depth image. This method lacks effective management of the point cloud index, making it difficult to quickly locate the original point cloud data in subsequent processing. Furthermore, existing methods suffer from high memory consumption and computational complexity when processing large-scale point cloud data, resulting in low online detection efficiency. Simultaneously, due to the complex distortion and deformation of refractory brick surfaces, a unified standard for surface distortion detection using 3D point clouds has not yet been established, making it difficult to guarantee the consistency and reliability of measurement results.

[0080] Figure 1 This is one of the flowcharts illustrating the method for measuring the torsion of refractory bricks based on three-dimensional point clouds provided in this application embodiment, such as... Figure 1 As shown, the method for measuring the torsion of refractory bricks based on three-dimensional point clouds includes steps 110, 120, 130, 140, and 150.

[0081] Step 110: Collect the three-dimensional point cloud of the refractory brick surface to be tested, and preprocess the three-dimensional point cloud to obtain the planar point cloud;

[0082] Figure 2 This is a schematic diagram of the structure for collecting measurements on the surface of refractory bricks provided in an embodiment of this application, as shown below. Figure 2 As shown, a three-dimensional point cloud is obtained by acquiring data from the surface of the refractory brick using an acquisition device. The acquisition device includes a transport track for transporting the refractory brick, a laser scanner, a three-dimensional vision sensor, and a data processor.

[0083] In some embodiments, the preprocessing of the 3D point cloud includes:

[0084] The three-dimensional point cloud is filtered using a pass-through filtering algorithm to obtain the three-dimensional point cloud of the target region.

[0085] Clustering and denoising are performed on the three-dimensional point cloud of the target region to obtain a planar point cloud.

[0086] It is easy to understand that after obtaining the 3D point cloud, in the camera coordinate system, the redundant point cloud data outside the target area is quickly removed by the pass-through filtering algorithm according to the height range of the 3D point cloud in the Z direction. Then, the point cloud data of the target area is clustered and denoised according to the point cloud density to remove the noise of abnormally flying single and continuous point clouds, so as to obtain the planar point cloud.

[0087] In this embodiment, filtering the 3D point cloud based on the pass-through filtering algorithm can effectively remove irrelevant or noisy data. Further clustering and denoising processing is then performed to obtain a planar point cloud, making subsequent depth image generation and distortion assessment more accurate.

[0088] Step 120: Establish the object coordinate system based on the principal component analysis method;

[0089] It should be noted that, since the camera's optical axis is not perpendicular to the refractory brick measurement surface, the Z-axis coordinate values ​​of the point cloud at the same height in the measurement plane are inconsistent under the camera coordinate system. To facilitate analysis, it is necessary to transform the refractory brick point cloud data in the camera coordinate system to the object coordinate system. The camera coordinate system is transformed into the object coordinate system by using the principal component analysis method. The Z-axis direction of the object coordinate system is the normal vector along the fitted plane of the measurement plane point cloud.

[0090] Step 130: Based on the planar point cloud, obtain the target depth image and index map;

[0091] After obtaining the planar point cloud, a depth image is generated based on the planar point cloud. First, the boundary range of the planar point cloud is calculated. Then, all point cloud data are traversed, and the minimum and maximum values ​​of the point cloud data in the X, Y, and Z directions are calculated to obtain a bounding box that can cover all point cloud data, as shown below:

[0092] {[X min ,X max ],[Y min ,Y max ],[Z min Z max ]}

[0093] The X and Y coordinate values ​​of all point cloud data in the planar point cloud are mapped to the pixel positions of the depth image, and the gray value of the pixel is calculated based on the Z coordinate value to obtain the target depth image.

[0094] Step 140: Perform X-shaped sampling on the target depth image within the area to be tested, and combine it with the index map to obtain the sampling point cloud;

[0095] Furthermore, the sub-pixel edges of the target depth image are first extracted using an edge detection algorithm, and then the weighted least squares method is used to fit the sub-pixel edges into a rectangle. The X-shaped region where the diagonal of the rectangle is located is sampled to obtain the sampled point cloud.

[0096] Step 150: Based on the sampling point cloud, obtain the torsion degree of the refractory brick through the torsion degree evaluation method.

[0097] Finally, the torsion of the refractory brick is obtained by the difference between the relative z coordinates in the sampled point cloud.

[0098] The method for measuring the torsion of refractory bricks based on three-dimensional point clouds provided in this application involves acquiring three-dimensional point cloud data of the refractory brick surface and preprocessing it to obtain a planar point cloud. Principal component analysis is used to establish an object coordinate system, eliminating tilt errors in the Z direction and making subsequent depth image generation and torsion assessment more accurate. A target depth image and index map are generated from the planar point cloud. X-shaped pixel sampling is performed on the two-dimensional depth map, and the sampled point cloud is obtained from the index map. The torsion assessment method is used to analyze the sampled point cloud, reducing the accuracy loss problem when generating point clouds from conventional depth images. Efficient addressing and access to the point cloud index improves the accuracy and efficiency of refractory brick torsion measurement, achieving rapid and accurate measurement of refractory brick torsion and reducing measurement complexity.

[0099] In some embodiments, obtaining the torsion of the refractory brick based on the sampled point cloud using a torsion evaluation method includes:

[0100] A plane fit is performed on the three-dimensional point cloud of the refractory brick surface to be tested to obtain a plane formula;

[0101] The spatial topological relationship of the point cloud is established based on the K-dimensional tree, and the centroid coordinates of the points in the sampled point cloud are obtained.

[0102] The z-coordinates of all sampling points in the sampling point cloud are obtained based on the aforementioned planar formula.

[0103] Calculate the relative z-coordinates of all sampling points in the sampling point cloud;

[0104] The torsion of the refractory brick is calculated based on the relative z-coordinates of all the sampling points.

[0105] In some embodiments, the process of obtaining the torsion degree of refractory bricks through a torsion degree evaluation method is as follows:

[0106] (1) Figure 3 This is a schematic diagram of the refractory brick torsion measurement provided in an embodiment of this application, as shown below. Figure 3 As shown, a plane fit is performed on the 3D point cloud to obtain the plane formula. The calculation formula for the plane formula is shown below:

[0107] Ax + By + Cz + D = 0

[0108] Where A is the first parameter, B is the second parameter, C is the third parameter, and D is the fourth parameter.

[0109] (2) Establish the spatial topological relationship of the point cloud through a K-dimensional tree, set the radius r of the sphere for neighborhood search, and take the sampling point m as an example. i Perform a neighborhood search on the seed node to obtain the sampling point m. i neighborhood point cloud {p i|i=1,2...n}, where n is the number of neighboring point clouds, calculate the centroid of the points in the neighborhood point cloud cluster. The calculation formula is as follows:

[0110]

[0111] (3) Calculate the z-coordinate of each sampling point in the sampling point cloud in the plane according to the plane formula. The calculation formula is as follows:

[0112]

[0113] (4) Calculate the relative z-coordinate of each sampling point. The calculation formula is shown below:

[0114]

[0115] In some embodiments, calculating the torsion of the refractory brick based on the relative z-coordinates of all the sampling points includes:

[0116] Iterate through the relative z-coordinates of all sampling points to obtain the maximum and minimum z-coordinate values;

[0117] Based on the maximum and minimum values ​​of the z-coordinate, the torsion of the refractory brick is calculated using the following formula:

[0118] f = d max -d min

[0119] Where f is the torsion of the refractory brick, and d max d is the maximum value of the z-coordinate. min It is the minimum value of the z-coordinate.

[0120] In this embodiment, by traversing the relative z-coordinates of all sampling points, the maximum and minimum z-coordinates are obtained, and the torsion of the refractory brick is calculated based on these maximum and minimum values, which reduces the measurement complexity and improves the measurement accuracy and efficiency of the refractory brick torsion.

[0121] In this embodiment, a planar formula is obtained by fitting a three-dimensional point cloud to a plane, and a spatial topological relationship of the point cloud is established based on a K-dimensional tree, enabling the acquisition of the centroid coordinates in the sampled point cloud. The z-coordinates of all sampled points are calculated based on the planar formula, and the relative z-coordinates are further calculated. The torsion of the refractory brick is then calculated based on the relative z-coordinates of all sampled points, effectively simulating the standard manual straightedge measurement process. This allows for efficient and accurate measurement of the refractory brick's torsion, reducing measurement errors.

[0122] In some embodiments, establishing the object coordinate system based on the principal component analysis method includes:

[0123] Based on the planar point cloud, construct a point cloud vector matrix;

[0124] The point cloud vector matrix is ​​decentered to obtain a decentralized point cloud vector matrix;

[0125] Based on the decentralized point cloud vector matrix, the covariance matrix is ​​obtained;

[0126] Based on the covariance matrix, the rotation matrix and homogeneous transformation matrix of the object coordinate system relative to the camera coordinate system are obtained;

[0127] Based on the rotation matrix and the homogeneous transformation matrix, an object coordinate system is established.

[0128] For example, Figure 4 This is a schematic diagram of the establishment of the object coordinate system provided in the embodiments of this application, such as... Figure 4 As shown, the process of establishing the object coordinate system using the principal component analysis method is as follows:

[0129] (1) For a target object with n point cloud data, arrange the coordinates of each point cloud into a 3×n matrix A. The expression of matrix A is as follows:

[0130]

[0131] in, It is a single point cloud vector.

[0132] (2) Decentralize all point cloud data. First, calculate the mean of each dimension in all point cloud data. The calculation formula is as follows:

[0133]

[0134] Next, subtracting the center data from matrix A yields the decentralized point cloud vector matrix B, as shown in the following expression:

[0135]

[0136] (3) Based on the decentralized point cloud vector matrix, the covariance matrix is ​​obtained;

[0137] Specifically, let C = B·B T The covariance matrix C is a 3×3 symmetric matrix that describes the distribution of the point cloud along the three coordinate axes. Eigenvalue decomposition is performed on the covariance matrix C to obtain its eigenvalues ​​λ0, λ1, λ2 and corresponding eigenvectors ν0, ν1, ν2. The formulas for calculating its eigenvalues ​​and eigenvectors are shown below:

[0138] C·v i =λ i vi

[0139] i∈{0,1,2}

[0140] eigenvalue λ i This indicates that the point cloud is represented by its corresponding feature vector v i The degree of dispersion of the distribution along the direction. Here, let 0 < λ0 < λ1 < λ2, according to the eigenvalue λ i The magnitude of the eigenvalue determines the orientation of the object's coordinate system axes. The eigenvector ν2 corresponding to the largest eigenvalue λ2 is the X' axis orientation vector in the object's coordinate system O′-X′Y′Z′, the eigenvector ν1 corresponding to the second largest eigenvalue λ1 is the Y' axis orientation vector in the object's coordinate system O′-X′Y′Z′, and the eigenvector ν0 corresponding to the smallest eigenvalue λ0 is the Z' axis orientation vector in the object's coordinate system O′-X′Y′Z′. ν0 also represents a smoothed estimate of the surface normal of the point cloud fitting.

[0141] The expression for the direction of the object coordinate system O′-X′Y′Z′ is as follows:

[0142] {v2,v1,v0}

[0143] The origin of the coordinate system is the geometric center of all point cloud data.

[0144] (4) The expression for the rotation matrix of the object coordinate system relative to the camera coordinate system is as follows:

[0145]

[0146] The expression for the homogeneous transformation matrix corresponding to the rotation matrix is ​​as follows:

[0147]

[0148] The homogeneous transformation matrix reveals the transformation of the object coordinate system relative to the camera coordinate system. This allows us to transform the object point cloud from the camera coordinate system to the object coordinate system. The calculation formula is shown below:

[0149]

[0150] In this embodiment, by establishing an object coordinate system based on principal component analysis, the object coordinate system can be aligned with the camera coordinate system. By constructing a point cloud vector matrix, performing decentralization processing, calculating the covariance matrix, and obtaining the rotation matrix and homogeneous transformation matrix, the object coordinate system is established, eliminating the tilt error of the measurement plane in the Z direction. This makes the subsequent depth image generation and distortion assessment more accurate, improving the accuracy and efficiency of refractory brick distortion measurement.

[0151] In some embodiments, obtaining the target depth image and index map based on the planar point cloud includes:

[0152] Traverse the planar point cloud to obtain the maximum and minimum values ​​of the planar point cloud in the X, Y, and Z directions;

[0153] Based on the maximum and minimum values ​​of the planar point cloud in the X, Y, and Z directions, the minimum bounding box of the planar point cloud is obtained.

[0154] Based on the resolution and range of the planar point cloud in the X and Y directions, an 8-bit depth image is allocated, and a 24-bit index image of the same size is generated simultaneously.

[0155] The coordinate values ​​of the planar point cloud in the X and Y directions are mapped to the pixel positions of the depth image, and the gray values ​​of the pixels in the depth image are calculated based on the coordinate values ​​of the planar point cloud in the Z direction to generate the target depth image.

[0156] Based on the position of the planar point cloud in the target depth image, the index of the planar point cloud is stored in the index map of the corresponding position.

[0157] In some embodiments, Figure 5 This is a schematic diagram of depth image storage provided in an embodiment of this application, such as... Figure 5 As shown, based on the X and Y ranges of the point cloud and the width and height of the depth image, the physical distance of each pixel in the X and Y directions is calculated using the following formula:

[0158]

[0159] Where fx is the physical distance in the X direction, fy is the physical distance in the Y direction, W is the width of the depth image, and H is the height of the depth image.

[0160] Traverse the point cloud to determine the position of point P in the depth image. The calculation formula is as follows:

[0161]

[0162] Among them, (i p ,j p ) represents the location of point cloud P in the depth image.

[0163] The Z-value of the point cloud is mapped to a grayscale range of 0–255 to generate the pixel value corresponding to each pixel in the depth image. The calculation formula is as follows:

[0164]

[0165] Where value is the pixel value corresponding to point cloud P.

[0166] Furthermore, the X and Y coordinates of each point in the planar point cloud are converted into pixel coordinates in the depth image, and the pixel values ​​are written to the depth image buffer. Each pixel in the depth image is allocated 4 bytes of memory space. This is achieved by storing the depth image data and the point cloud index in the same memory buffer, which is divided into two regions: the first byte stores the pixel value of the depth image, and the last 3 bytes store the point cloud index, used to record the point cloud data index corresponding to that pixel. Each index number is an integer (usually of type size_t, occupying 4 or 8 bytes). Using 3 bytes to store the index saves memory space, and the index range that 3 bytes can represent is 0 to 2. 24 -1 (i.e., 16,777,215) can meet the indexing needs of most point cloud data, save memory space, facilitate the mapping between depth images and point cloud data, and can be looked up and accessed through pointer operations.

[0167] In this embodiment, the visualization and storage of planar point clouds are achieved by obtaining the target depth image through planar point clouds and storing the point cloud index. This reduces memory fragmentation and data redundancy, improves the processing efficiency and memory storage rate of planar point clouds, optimizes data access speed, makes subsequent distortion assessment more accurate, and improves the accuracy and efficiency of refractory brick distortion measurement.

[0168] In some embodiments, the step of performing X-shaped sampling on the target depth image within the region to be measured, and combining it with an index map to obtain a sampled point cloud, includes:

[0169] Subpixel edges of the target depth image are extracted based on an edge detection algorithm;

[0170] The sub-pixel edges are fitted to straight lines l1, l2, l3, and l4 using the weighted least squares method;

[0171] Based on the lines l1, l2, l3 and l4, calculate the intersection points P1, P2, P3 and P4 of adjacent lines;

[0172] The domain ReduceRegion of the target depth image is obtained based on the boundary cropping distance;

[0173] Connect the diagonal points P1, P2 and P3, P4 to obtain the linear regions LineRegion1 and LineRegion2;

[0174] Calculate the intersection region of LineRegion1, LineRegion2 and ReduceRegion, and store the pixel coordinates of the intersection region into the sampling point cloud pixel set;

[0175] Based on the pixel set of the sampled point cloud and the index map, the sampled point cloud is obtained.

[0176] In some embodiments, Figure 6 This is a schematic diagram of X-shaped sampling provided in an embodiment of this application, as shown below. Figure 6 As shown, simulating the straightedge method in manual measurement, several feature points in an "X" shape are sampled from the point cloud area on the upper surface of the refractory brick. The specific process is as follows:

[0177] Subpixel edges of the target depth image are extracted using the Canny edge detection algorithm;

[0178] The subpixel edges are fitted as straight lines l1, l2, l3, and l4 using the Tukey weighted least squares method, and then the intersection points P1, P2, P3, and P4 of adjacent straight lines are calculated.

[0179] Set the boundary clipping distance to reduce the influence of missing edges and flash on the distortion measurement of bricks, and obtain the domain ReduceRegion of the depth image of the inward target;

[0180] Connect the diagonal points P1, P2 and P3, P4 to form straight line regions LineRegion1 and LineRegion2. Calculate the intersection region between LineRegion1, LineRegion2 and ReduceRegion, and store the pixel coordinates of the intersection region in the sampled point cloud pixel set G = {p1, p2, ..., p...}. n};

[0181] The 3D point cloud index corresponding to the pixel coordinates in G is obtained from the storage index, and then the coordinate set G′={m1,m2,…,m n}

[0182] Figure 7 This is the second flowchart illustrating the method for measuring the torsion of refractory bricks based on three-dimensional point clouds provided in this application embodiment. Figure 7 As shown, the original point cloud of the refractory brick surface is first collected, and the original point cloud is preprocessed to obtain the target point cloud. Then, the object coordinate system corresponding to the target point cloud is established by the principal component analysis method, the target point cloud is used to generate a depth image, the depth image is sampled and processed, and finally the distortion of the refractory brick is obtained by the distortion evaluation method.

[0183] In this embodiment, by performing image processing operations such as edge detection and line fitting on the depth image, point cloud data of the diagonal region on the upper surface of the refractory brick can be quickly extracted. X-shaped sampling of the depth image can effectively extract key feature points in the depth image, obtain high-quality sampled point cloud, improve the accuracy and precision of point cloud data sampling, make subsequent distortion assessment more accurate, and improve the accuracy and efficiency of refractory brick distortion measurement.

[0184] The refractory brick torsion measurement method based on three-dimensional point clouds provided in this application can be executed by a refractory brick torsion measurement device based on three-dimensional point clouds. This application uses the execution of the refractory brick torsion measurement method based on three-dimensional point clouds by the refractory brick torsion measurement device as an example to illustrate the refractory brick torsion measurement device based on three-dimensional point clouds provided in this application.

[0185] This application also provides a device for measuring the torsion of refractory bricks based on three-dimensional point clouds, such as... Figure 8 As shown, the refractory brick torsion measurement device based on three-dimensional point cloud includes: acquisition module 810, establishment module 820, first processing module 830, second processing module 840 and third processing module 850.

[0186] The acquisition module 810 is used to acquire a three-dimensional point cloud of the surface of the refractory brick to be tested, and to preprocess the three-dimensional point cloud to obtain a planar point cloud.

[0187] Module 820 is established to create an object coordinate system based on the principal component analysis method.

[0188] The first processing module 830 is used to obtain a target depth image and an index map based on the planar point cloud;

[0189] The second processing module 840 is used to perform X-shaped sampling on the target depth image in the area to be tested, and combine it with the index map to obtain the sampling point cloud;

[0190] The third processing module 850 is used to obtain the torsion of the refractory brick based on the sampling point cloud using a torsion evaluation method.

[0191] The method for measuring the torsion of refractory bricks based on three-dimensional point clouds provided in this application involves acquiring three-dimensional point cloud data of the refractory brick surface and preprocessing it to obtain a planar point cloud. Principal component analysis is used to establish an object coordinate system, eliminating tilt errors in the Z direction and making subsequent depth image generation and torsion assessment more accurate. A target depth image and index map are generated from the planar point cloud. X-shaped pixel sampling is performed on the two-dimensional depth map, and the sampled point cloud is obtained from the index map. The torsion assessment method is used to analyze the sampled point cloud, reducing the accuracy loss problem when generating point clouds from conventional depth images. Efficient addressing and access to the point cloud index improves the accuracy and efficiency of refractory brick torsion measurement, achieving rapid and accurate measurement of refractory brick torsion and reducing measurement complexity.

[0192] The refractory brick torsion measurement device based on three-dimensional point cloud provided in this application embodiment can achieve... Figures 1 to 7 The various processes implemented in the embodiment of the refractory brick torsion measurement method based on three-dimensional point cloud will not be described in detail here to avoid repetition.

[0193] In some embodiments, such as Figure 9 As shown, this application embodiment also provides an electronic device 900, including a processor 901, a memory 902, and a computer program stored in the memory 902 and executable on the processor 901. When the program is executed by the processor 901, it implements the various processes of the above-described embodiment of the method for measuring the torsion of refractory bricks based on three-dimensional point clouds and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0194] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0195] This application also provides a non-transitory computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described embodiment of the method for measuring the torsion of refractory bricks based on three-dimensional point clouds and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0196] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0197] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for measuring the torsion of refractory bricks based on three-dimensional point clouds.

[0198] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0199] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described embodiment of the refractory brick torsion measurement method based on three-dimensional point cloud, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0200] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a device-level chip, device chip, chip device, or on-chip device chip, etc.

[0201] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0202] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the refractory brick torsion measurement method based on three-dimensional point cloud of the various embodiments of this application.

[0203] In the description of this application, "first feature" and "second feature" may include one or more of the features.

[0204] In the description of this application, "multiple" means two or more.

[0205] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0206] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0207] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for measuring the twist of a refractory brick based on a three-dimensional point cloud, characterized in that, The method includes: A three-dimensional point cloud is collected from the surface of the refractory brick to be tested, and the three-dimensional point cloud is preprocessed to obtain a planar point cloud. Establish the object coordinate system based on principal component analysis; Based on the planar point cloud, a target depth image and an index map are obtained; The target depth image is sampled in an X-shape within the region to be tested, and the sampled point cloud is obtained by combining it with the index map, including: Subpixel edges of the target depth image are extracted based on an edge detection algorithm; fitting the sub-pixel edge as a straight line using a weighted least squares method , , and ; Based on the straight line , , and , the intersection point of adjacent straight lines is calculated , , and ; The domain ReduceRegion of the target depth image is obtained based on the boundary cropping distance; Connect the diagonal points , and , This yields the linear regions LineRegion1 and LineRegion2; Calculate the intersection region of LineRegion1, LineRegion2 and ReduceRegion, and store the pixel coordinates of the intersection region into the sampling point cloud pixel set; Based on the sampled point cloud pixel set and the index map, the sampled point cloud is obtained; Based on the sampled point cloud, the torsion of the refractory brick is obtained through a torsion evaluation method, including: A plane fit is performed on the three-dimensional point cloud of the refractory brick surface to be tested to obtain a plane formula; The spatial topological relationship of the point cloud is established based on the K-dimensional tree, and the centroid coordinates of the points in the sampled point cloud are obtained. The z-coordinates of all sampling points in the sampling point cloud are obtained based on the aforementioned planar formula. Calculate the relative z-coordinates of all sampling points in the sampling point cloud; The torsion of the refractory brick is calculated based on the relative z-coordinates of all the sampling points.

2. The three-dimensional point cloud-based refractory brick twist measurement method according to claim 1, characterized by, The establishment of the object coordinate system based on the principal component analysis method includes: Based on the planar point cloud, construct a point cloud vector matrix; The point cloud vector matrix is ​​decentered to obtain a decentralized point cloud vector matrix; Based on the decentralized point cloud vector matrix, the covariance matrix is ​​obtained; Based on the covariance matrix, the rotation matrix and homogeneous transformation matrix of the object coordinate system relative to the camera coordinate system are obtained; Based on the rotation matrix and the homogeneous transformation matrix, an object coordinate system is established.

3. The method for measuring the torsion of refractory bricks based on three-dimensional point clouds according to claim 1, characterized in that, The process of obtaining the target depth image and index map based on the planar point cloud includes: Traverse the planar point cloud to obtain the maximum and minimum values ​​of the planar point cloud in the X, Y, and Z directions; Based on the maximum and minimum values ​​of the planar point cloud in the X, Y, and Z directions, the minimum bounding box of the planar point cloud is obtained. Based on the resolution and range of the planar point cloud in the X and Y directions, an 8-bit depth image is allocated, and a 24-bit index image of the same size is generated simultaneously. The coordinate values ​​of the planar point cloud in the X and Y directions are mapped to the pixel positions of the depth image, and the gray values ​​of the pixels in the depth image are calculated based on the coordinate values ​​of the planar point cloud in the Z direction to generate the target depth image. Based on the position of the planar point cloud in the target depth image, the index of the planar point cloud is stored in the index map of the corresponding position.

4. The method for measuring the torsion of refractory bricks based on three-dimensional point clouds according to claim 1, characterized in that, The calculation of the torsion of the refractory brick based on the relative z-coordinates of all the sampling points includes: Iterate through the relative z-coordinates of all sampling points to obtain the maximum and minimum z-coordinate values; Based on the maximum and minimum values ​​of the z-coordinate, the torsion of the refractory brick is calculated using the following formula: ； wherein f is a degree of twist of the firebrick, is a maximum value of the z coordinate, is a minimum value of the z coordinate.

5. The three-dimensional point cloud-based refractory brick twist measurement method according to claim 1, characterized by, The preprocessing of the 3D point cloud includes: The three-dimensional point cloud is filtered using a pass-through filtering algorithm to obtain the three-dimensional point cloud of the target region. Clustering and denoising are performed on the three-dimensional point cloud of the target region to obtain a planar point cloud.

6. A device for measuring the torsion of refractory bricks based on three-dimensional point clouds, implemented using the method for measuring the torsion of refractory bricks based on three-dimensional point clouds as described in any one of claims 1 to 5, characterized in that, The device includes: The acquisition module is used to acquire a three-dimensional point cloud of the refractory brick surface to be tested, and to preprocess the three-dimensional point cloud to obtain a planar point cloud. A module is created to establish an object coordinate system based on the principal component analysis method. The first processing module is used to obtain a target depth image and an index map based on the planar point cloud; The second processing module is used to perform X-shaped sampling on the target depth image within the region to be measured, and combine it with the index map to obtain a sampled point cloud, including: Subpixel edges of the target depth image are extracted based on an edge detection algorithm; fitting the sub-pixel edge as a straight line using a weighted least squares method , , and ; Based on the straight line , , and , the intersection point of adjacent straight lines is calculated , , and ; The domain ReduceRegion of the target depth image is obtained based on the boundary cropping distance; connecting the opposite corners , and , , resulting in line regions LineRegion1 and LineRegion2; Calculate the intersection region of LineRegion1, LineRegion2 and ReduceRegion, and store the pixel coordinates of the intersection region into the sampling point cloud pixel set; Based on the sampled point cloud pixel set and the index map, the sampled point cloud is obtained; The third processing module is used to obtain the torsion of the refractory brick based on the sampled point cloud using a torsion evaluation method, including: A plane fit is performed on the three-dimensional point cloud of the refractory brick surface to be tested to obtain a plane formula; The spatial topological relationship of the point cloud is established based on the K-dimensional tree, and the centroid coordinates of the points in the sampled point cloud are obtained. The z-coordinates of all sampling points in the sampling point cloud are obtained based on the aforementioned planar formula. Calculate the relative z-coordinates of all sampling points in the sampling point cloud; The torsion of the refractory brick is calculated based on the relative z-coordinates of all the sampling points.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for measuring the torsion of refractory bricks based on three-dimensional point clouds as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for measuring the torsion of refractory bricks based on three-dimensional point clouds as described in any one of claims 1 to 5.

Images