Screen automatic welding trajectory point detection method and device

Abstract

The present invention provides a method and device for automatic weld trajectory point detection of a screen mesh. The method comprises: obtaining a three-dimensional screen point cloud corresponding to a screen mesh to be detected; based on local field statistics, filtering out wild points and noise points in the three-dimensional screen point cloud to obtain a filtered three-dimensional point cloud; projecting the screen point cloud along a preset Z axis and performing gridding processing to obtain a two-dimensional image, and performing expansion processing on the screen point cloud to obtain a screen gap connection image; performing region growing processing on the screen gap connection image to obtain a region growing image and a screen mesh set and a mesh corner point set corresponding to each screen mesh; determining a two-dimensional coordinate set of weld points based on the transverse corner point direction of the screen mesh set and the mesh spacing of the screen mesh; and constraining the two-dimensional coordinates of each weld point to approximately the same spacing and angle of X-shaped neighbors to obtain a constrained result. The present invention can determine the two-dimensional coordinates of weld points based on the corner point set of each screen mesh, and the process is fully automatic, thereby improving the efficiency of locating weld points.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence image processing technology, and in particular to a method and device for detecting automatic welding trajectory points of a screen. Background Art

[0002] Currently, common automated screen welding systems primarily utilize modular systems for fixing and positioning the screen, a robotic welding subsystem for performing welding operations based on the positioning, and a 2D vision subsystem for identifying and locating weld points. This modular system typically requires uniformly securing and clamping the screen, placing high demands on the screen type and structure. Furthermore, the weld point identification based on image data is affected by the brightness of the visible light source, making 2D camera-based visual detection ineffective in dark nights. Therefore, the existing technology needs improvement. Summary of the Invention

[0003] The present invention provides a method and device for detecting automatic welding trajectory points of a screen mesh, which can identify welding points based on point cloud projection transformation without the need for human intervention throughout the process.

[0004] In a first aspect, the present invention provides a method for detecting automatic welding trajectory points of a screen mesh, the method comprising:

[0005] Obtaining a three-dimensional point cloud of the sieve corresponding to the sieve to be tested;

[0006] Based on local area statistics, outliers and noise points in the mesh three-dimensional point cloud are screened out to obtain a screened three-dimensional point cloud;

[0007] Projecting the filtered three-dimensional point cloud along a preset Z axis and performing gridding processing to obtain a two-dimensional image, and expanding the two-dimensional image based on a preset graphic mask to obtain a screen gap connection image;

[0008] Processing the mesh gap connectivity image based on region growing to obtain a region growing image, a mesh set in the region growing image, and a mesh corner point set corresponding to each mesh in the mesh set;

[0009] Determining a two-dimensional coordinate set of welding points according to the transverse corner point directions of the screen mesh set and the mesh spacing of the screen mesh;

[0010] The two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set are constrained so that the spacing and angle of the X-shaped neighbors are approximately consistent, and a constraint result is obtained.

[0011] In a second aspect, the present invention further provides a device for detecting automatic mesh welding trajectory points, the device comprising:

[0012] A three-dimensional point cloud acquisition module is used to obtain a three-dimensional point cloud of the screen corresponding to the screen to be tested;

[0013] A three-dimensional point cloud screening module is used to screen out wild points and noise points in the three-dimensional point cloud of the screen based on local field statistics to obtain a screened three-dimensional point cloud;

[0014] a two-dimensional projection module, configured to project the three-dimensional point cloud after screening along a preset Z axis and perform gridding processing to obtain a two-dimensional image, and to dilate the two-dimensional image based on a preset graphic mask to obtain a screen gap connection image;

[0015] a screen mesh acquisition module, configured to process the screen gap connectivity image based on region growing to obtain a region growing image, a screen mesh set in the region growing image, and a mesh corner point set corresponding to each screen mesh in the screen mesh set;

[0016] a welding point identification module, configured to determine a two-dimensional coordinate set of welding points according to the transverse corner point directions of the mesh set of the screen and the mesh spacing of the meshes of the screen;

[0017] The welding point checking module is used to constrain the two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set to be approximately consistent with the spacing and angle of the X-shaped neighbors to obtain a constraint result.

[0018] In the third aspect, the present invention also provides a processing device, including a processor and a memory, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, it executes the steps in any one of the screen automatic welding trajectory point detection methods provided in the embodiments of the present application.

[0019] In a fourth aspect, the present invention also provides a computer-readable storage medium, which stores a plurality of instructions suitable for loading by a processor to execute the steps in any one of the screen automatic welding trajectory point detection methods provided in the embodiments of the present application.

[0020] From the above content, it can be concluded that the present invention realizes the acquisition of the screen mesh set in the screen to be tested based on the two-dimensional projection transformation of the three-dimensional point cloud, and determines the two-dimensional coordinates of the welding points based on the corner point set of each screen mesh in the screen mesh set. The whole process is automatically executed without human intervention, thereby improving the efficiency of locating the welding points in the screen to be tested. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a flow chart of the automatic welding trajectory point detection method for screen mesh in this application;

[0023] Figure 2a It is a schematic diagram of the screen to be detected in the automatic welding trajectory point detection method of the screen in this application;

[0024] Figure 2b It is a schematic diagram of the region growth image in the automatic welding trajectory point detection method of the screen mesh in this application;

[0025] Figure 3 It is a schematic diagram of the mesh holes and mesh corner point sets in the automatic mesh welding trajectory point detection method in this application;

[0026] Figure 4 This is a schematic diagram of the method for detecting automatic welding trajectory points of screen mesh in this application, in which the welding points are constrained to be approximately consistent in spacing and angle with their X-shaped neighbors;

[0027] Figure 5 This is a structural diagram of the automatic welding trajectory point detection device for screen mesh in this application;

[0028] Figure 6 It is a structural diagram of the processing equipment of this application. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0030] In the following description, the specific embodiments of the present application will be described with reference to the steps and symbols performed by one or more computers, unless otherwise stated. Therefore, these steps and operations will be mentioned several times as being performed by a computer, and the computer execution referred to in the embodiments of the present application includes the operation of a computer processing unit that represents an electronic signal of data in a structured form. This operation converts the data or maintains it at a location in the memory system of the computer, which can be reconfigured or otherwise change the operation of the computer in a manner familiar to testers in the field. The data structure maintained by the data is a physical location in the memory, which has specific characteristics defined by the data format. However, the principles of the present application are described in the above text, which does not represent a limitation, and testers in the field will understand that the various steps and operations described below can also be implemented in hardware.

[0031] The principles of the present application may be implemented using many other general-purpose or specific-purpose computing and communication environments or configurations. Examples of well-known computing systems, environments, and configurations suitable for use with the present application include, but are not limited to, handheld phones, personal computers, servers, multiprocessor systems, microcomputer-based systems, mainframe computers, and distributed computing environments, including any of the aforementioned systems or devices.

[0032] The terms "first", "second", and "third" in this application are used to distinguish different objects rather than to describe a specific order. In addition, the terms "include", "have", and any variations thereof are intended to cover non-exclusive inclusions.

[0033] First, before introducing the embodiments of the present application, the following is an introduction to the application background of the present application.

[0034] The executor of the automatic welding trajectory point detection method for the screen provided in the present application can be the device provided in the present application, or a processing device such as a server device, a physical host or a user equipment (UE) that integrates the device, wherein the device can be implemented in hardware or software, and the UE can specifically be a terminal device such as a smart phone, a tablet computer, a laptop computer, a PDA, a desktop computer or a personal digital assistant (PDA).

[0035] Next, we will introduce the automatic welding trajectory point detection method for the screen provided by this application.

[0036] See Figure 1 , Figure 1 A schematic flow chart of the automatic welding trajectory point detection method for screen mesh of the present application is shown, and the method is applied to a server. The embodiment of the present application takes the server as the execution subject as an example. The method provided by the present application may specifically include the following steps:

[0037] S101: Obtain a three-dimensional point cloud of a sieve corresponding to the sieve to be detected.

[0038] In the embodiment of the present application, first Figure 2a The screen to be detected is placed under the laser radar (for example, the upper surface of the screen to be detected is aligned with the laser radar, and the back of the screen to be detected is in contact with the support surface), and then the light formed by the 2D laser line emitted by the laser radar can be referred to as Figure 2a In the vertical direction shown, the 2D laser line is as follows Figure 2aScanning from left to right, as shown, a multi-frame 3D point cloud of the screen under inspection is ultimately acquired in a time-sequential manner. The resulting 3D point cloud of the screen is ultimately uploaded to a server for further processing. This demonstrates that acquiring a 3D point cloud of the screen using a 2D laser line offers the advantages of obtaining highly accurate distance information and being unrestricted by lighting conditions.

[0039] Among them, the acquisition of the three-dimensional point cloud of the screen to be detected based on the laser radar is essentially laser three-dimensional imaging. More specifically, the distance information (Z axis) is obtained by measuring the flight time of the light pulse or modulated signal between the radar and the target, and the orientation information (X, Y axis) in the plane perpendicular to the light beam direction is obtained through scanning or multi-point corresponding measurement.

[0040] Because the screen is characterized by its multiple meshes and multiple weld points, when the 2D laser line hits the mesh area, it exceeds the effective laser range and no points are formed in the 3D point cloud of the screen. However, when the 2D laser line hits the screen area, it loses the effective laser range and can obtain a multi-frame 3D point cloud of the screen to be inspected.

[0041] S102 : Based on local area statistics, outliers and noise points in the mesh three-dimensional point cloud are screened out to obtain a screened three-dimensional point cloud.

[0042] In an embodiment of the present application, after the initial sieve three-dimensional point cloud is obtained, there will be some outliers (which can be understood as wild points) and some noise points. At this time, the local field statistics technology can be used to filter out the wild points and noise points in the sieve three-dimensional point cloud to obtain the filtered three-dimensional point cloud.

[0043] In one embodiment, step S102 includes:

[0044] Obtaining a domain smoothing range corresponding to the local domain statistics; wherein the domain smoothing range is a size of 3*3;

[0045] Each pixel in the mesh three-dimensional point cloud is locally smoothed based on the domain smoothing range to obtain a screened three-dimensional point cloud.

[0046] Among them, the local area statistics method can also be understood as a local statistics method, which is a point operation method for the spatial domain, and its function is to perform image enhancement. The local statistics method specifically includes the local smoothing method, which replaces the original grayscale value of each pixel in the image with the grayscale average value of each pixel in its neighborhood to achieve image smoothing. For example, it is common to use a pixel as the center point and replace the original grayscale value of the pixel with the grayscale average value of the remaining 8 pixels in a 3*3 size around it. It can be seen that based on the above method, the image is finally smoothed and the wild points and noise in the image are removed.

[0047] S103 , projecting the screened three-dimensional point cloud along a preset Z axis and performing gridding processing to obtain a two-dimensional image, and expanding the two-dimensional image based on a preset graphic mask to obtain a screen gap connection image.

[0048] In the embodiments of the present application, projecting the filtered 3D point cloud along a preset Z-axis is a transformation that projects the 3D point cloud onto a 2D plane. Specifically, the transformation involves multiplying the coordinates of each 3D point cloud in the filtered 3D point cloud by a transformation matrix and then projecting it onto a 2D plane, thereby achieving a dimensionality reduction transformation. After completing the 2D projection and performing image expansion processing, the screen rebar pixels can be rendered connected; otherwise, each screen mesh appears as an independent block.

[0049] In one embodiment, the step S103 of projecting the filtered three-dimensional point cloud along a preset Z axis and performing gridding processing to obtain a two-dimensional image includes:

[0050] Projecting the filtered three-dimensional point cloud along the Z axis to obtain a Z-axis projection image;

[0051] The Z-axis projection image is gridded with 1 / 8 of the diameter of the cross section of each steel bar in the screen to be detected as the length and width of a single pixel to obtain a two-dimensional image.

[0052] In an embodiment of the present application, to achieve two-dimensional projection, a transformation matrix corresponding to the Z-axis projection is first obtained. The coordinates of each 3D point cloud in the filtered 3D point cloud (generally three-dimensional coordinates) are then multiplied by the transformation matrix and projected onto a two-dimensional plane (after projection onto the two-dimensional plane, the coordinates of each 3D point cloud become two-dimensional coordinates) to obtain a Z-axis projection image. More specifically, after obtaining the coordinates of each 3D point cloud in the filtered 3D point cloud, the Z-axis coordinates of each 3D point cloud are set to zero and only the X-axis and Y-axis coordinates are retained. This allows the two-dimensional coordinates corresponding to the coordinates of each 3D point cloud in the filtered 3D point cloud to be obtained, thereby obtaining a Z-axis projection image.

[0053] Once the Z-axis projection image is obtained, it needs to be gridded. Specifically, the Z-axis projection image is gridded with a pixel length and width of 1 / 8 the diameter of each steel bar section in the screen to be inspected, resulting in a two-dimensional image. Furthermore, during the gridding process, if a point is in the grid, the pixel is set to 1, otherwise it is set to 0.

[0054] In order to ensure that the mesh holes in the Z-axis projection image are connected rather than independent blocks, the Z-axis projection image needs to be further expanded so that the mesh steel bar pixels appear connected.

[0055] S104 , processing the screen gap connectivity image based on region growing to obtain a region growing image, a screen mesh set in the region growing image, and a mesh corner point set corresponding to each screen mesh in the screen mesh set.

[0056] In the embodiment of the present application, after obtaining the screen gap connection image, in order to better identify the screen mesh, the screen gap connection image can be processed based on region growth to obtain a region growth image. Since the region growth image makes the screen mesh area clearer, please refer to Figure 2b Therefore, the sieve mesh set in the region growing image and the mesh corner point set corresponding to each sieve mesh in the sieve mesh set can be obtained based on the region growing image.

[0057] In one embodiment, step S104 includes:

[0058] Aggregate the pixels or sub-regions in the mesh gap connected image to satisfy the conditions corresponding to the region growing process to obtain a region growing image, a mesh set in the region growing image, and a mesh corner point set corresponding to each mesh in the mesh set.

[0059] Region growing is the process of aggregating pixels or sub-regions into larger regions based on a pre-defined criterion, and it is necessary to ensure that the segmented regions meet the following conditions 1) to 5):

[0060] 1)U(I i )=I;

[0061] 2) Ii is a connected region, i = 1, 2, 3, ... K;

[0062] 3)I i ∩I j = empty set, for any i, j; i≠j;

[0063] 4)P(I i )=True, for i=1,2,……K;where P(I i )=True means the prismatic mesh P(I i ) is not incremented and marked as True;

[0064] 5)I(P i UP j )=False, i≠j; where I(P i UP j ) represents the mesh P i and mesh P j After merging growth, mark it as False;

[0065] Wherein, I represents the mesh gap connected image, which can be divided into I1, I2, ..., I K These sub-regions have no intersecting parts. Therefore, sub-regions are selected in sequence in the mesh gap connection image I for region growth until all mesh pixels have completed growth.

[0066] S105 , determining a two-dimensional coordinate set of welding points according to the transverse corner point directions of the mesh set of the screen and the mesh spacing of the screen meshes.

[0067] In the embodiment of the present application, when the following Figure 2b The region growth image shown in FIG. 1 is shown in FIG. 2 , and the mesh set and the mesh corner point set corresponding to each mesh in the mesh set are known (for example, Figure 2b Each diamond-shaped blank area in the figure represents a mesh hole of the screen, and each mesh hole of the screen has 4 mesh corner points. The two-dimensional coordinates of the weld point can be determined from top to bottom and from left to right (the direction from left to right corresponds to the direction of the transverse corner points). Specifically, a center point determined by two adjacent mesh holes in the same transverse direction can be regarded as a weld point. More specifically, the center point determined by the left corner point of the mesh hole on the left and the right corner point of the mesh hole on the right of the two adjacent mesh holes can be regarded as a weld point. It can be seen that by determining the weld point based on the center points of adjacent mesh holes, the specific positions of multiple weld points can be obtained more quickly and accurately.

[0068] In one embodiment, step S105 includes:

[0069] If there is a mesh corner point set in the mesh set with the smallest distance between the mesh center point coordinates and the origin, obtain the corresponding target mesh, and record the target mesh as the first row and first column mesh P 1，1 , and get the first row and first column of the screen mesh P 1，1 The corresponding mesh corner point set; among them, the first row and first column of the screen mesh P 1，1 Located at row 1 and column 1 in the region growing image;

[0070] In the region growth image, search from left to right in the horizontal corner direction for the mesh P of the i-th row and j-th column. i , j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i ,j + 1; where the initial values ​​of i and j are both 1, and the value range of i is [1, M], and the value range of j is [1, N], where M is a positive integer and N is a positive integer;

[0071] If it is determined that j+1 does not exceed N, based on the mesh size P of the i-th row and j-th column i，j The left corner coordinates and the mesh size P of the sieve in the i-th row and j+1th column i，j+1 The center point of the right corner coordinates obtains the two-dimensional coordinates S of the welding point in row i and column j i，j ;

[0072] Increment j by 1;

[0073] If it is determined that j does not exceed N, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i , i+1 Steps;

[0074] If it is determined that j exceeds N, increment i by 1 and reset j to 1;

[0075] If it is determined that i does not exceed M, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 Steps;

[0076] If it is determined that i exceeds M, obtain the two-dimensional coordinates S of the welding point in row 1 and column 1 1，1 To the two-dimensional coordinate S of the welding point in row M and column N M，N Compose the two-dimensional coordinate set of welding points.

[0077] In the examples of this application, please refer to Figure 3 If the point in the upper left corner of the region growing image is regarded as the origin, it is first necessary to obtain the sieve mesh closest to the upper left corner origin in the sieve mesh set as the target sieve mesh, and record the target sieve mesh as the first row and first column sieve mesh P 1，1 , where the first row and first column of the screen mesh size P 1，1 Located in the 1st row and 1st column of the region growth image. Since each mesh hole is considered as a diamond shape, the mesh hole P in the first row and first column is 1，1 The corresponding mesh corner point set includes 4 corner points (recorded as left corner point, upper corner point, right corner point and lower corner point respectively). 1，1 Then, it is necessary to locate the adjacent mesh of the screen that is also in the first row, specifically, the mesh of the screen P in the first row and the second column. 1，2 When the first row and first column of the screen mesh P is located 1，1 and the first row and second column of the screen mesh P 1，2 Then, based on the first row and first column of the screen mesh P 1，1The left corner point and the first row and second column mesh P 1，2 The center point of the right corner of the right corner is used to determine the two-dimensional coordinate S of the welding point in the first row and first column 1，1 Similarly, get the mesh size P in the first row and third column 1，3 , based on the first row and second column of the screen mesh P 1，2 The left corner point and the first row and third column of the sieve mesh P 1，3 The center point of the right corner of the 2D coordinate S of the welding point in row 1 and column 2 is determined by 1，2 Similarly, after determining the 2D coordinates of all weld points in the first row, we proceed to the second row, starting from the leftmost mesh of the second row and working our way from left to right to determine the 2D coordinates of all weld points in the second row. The 2D coordinates of the weld points in the second row and all subsequent rows are determined using the same method as the 2D coordinates of the weld points in the first row. This indicates that, based on this row-by-row and column-by-column inspection method, the 2D coordinates of all weld points can be accurately determined, thus forming a 2D coordinate set of weld points.

[0078] S106 , constraining the two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set so that the spacing and angle of the X-shaped neighbors are approximately consistent to obtain a constraint result.

[0079] In this embodiment, after all welding points in the region growing image are obtained, the two-dimensional coordinates of each welding point need to be constrained so that the spacing and angle of the X-shaped neighbors are approximately consistent to obtain a constrained result.

[0080] For example, reference Figure 4 , with the two-dimensional coordinates S of the welding point in the second row and second column 2，3 As an example of the consistency constraint, obtain the two-dimensional coordinates S of the welding point in the second row and third column 2，3 The two-dimensional coordinates S of the welding point in the first row and second column of the X-type neighbor 1，2 , the first row and third column two-dimensional coordinates of the welding point S 1，3 , the two-dimensional coordinates S of the welding point in the third row and second column 3，2 And the two-dimensional coordinates S of the welding point in the third row and third column 3，3 . With S 1，2 and S 2，3 Get the first vector X1, with S 1，3 and S 2，3 Get the second vector X2, with S 3，2 and S 2，3 Get the third vector X3, with S 3，3 and S 2，3Obtain the fourth vector X4. If the angle between the first vector X1 and the second vector X2 is 90°, the angle between the second vector X2 and the third vector X4 is 90°, the angle between the fourth vector X4 and the third vector X3 is 90°, the angle between the third vector X3 and the first vector X1 is 90°, the angle between the first vector X1 and the fourth vector X4 is 180°, and the angle between the second vector X2 and the third vector X3 is 180°, then it can be determined that the two-dimensional coordinates of the above five welding points meet the X-type nearest neighbor spacing and angle approximate consistency constraints and no adjustment is required. If the above conditions are not met, appropriate adjustments are required to meet the X-type nearest neighbor spacing and angle approximate consistency constraints.

[0081] When the two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set are constrained to be approximately consistent with the spacing and angle of X-shaped neighbors to obtain a constraint result, the two-dimensional coordinates of each welding point included in the constraint result are adjusted to form the welding point trajectory data corresponding to the screen to be detected.

[0082] It can be seen that the embodiment of the present application realizes the acquisition of the screen mesh set in the screen to be inspected based on the two-dimensional projection transformation of the three-dimensional point cloud, and determines the two-dimensional coordinates of the welding points based on the corner point set of each screen mesh in the screen mesh set. The whole process is automatically executed without the need for human intervention, thereby improving the efficiency of locating the welding points in the screen to be inspected.

[0083] In order to facilitate better implementation of the method of the present application, the embodiment of the present application also provides a screen automatic welding trajectory point detection device 100.

[0084] See also Figure 5 , Figure 5 This is a structural schematic diagram of the screen automatic welding trajectory point detection device 100 of the present application, wherein the screen automatic welding trajectory point detection device 100 may specifically include the following structures: a three-dimensional point cloud acquisition module 101, a three-dimensional point cloud screening module 102, a two-dimensional projection module 103, a screen mesh acquisition module 104, a welding point recognition module 105 and a welding point verification module 106.

[0085] The three-dimensional point cloud acquisition module 101 is used to acquire the three-dimensional point cloud of the sieve corresponding to the sieve to be detected.

[0086] In the embodiment of the present application, first Figure 2a The screen to be detected is placed under the laser radar (for example, the upper surface of the screen to be detected is aligned with the laser radar, and the back of the screen to be detected is in contact with the support surface), and then the light formed by the 2D laser line emitted by the laser radar can be referred to as Figure 2a In the vertical direction shown, the 2D laser line is as follows Figure 2aIf the direction shown is scanned from left to right in sequence, multiple frames of three-dimensional point cloud of the screen to be detected can be finally obtained in time sequence. Specifically, it can be understood that at time t1, a frame of three-dimensional point cloud is obtained when the 2D laser line scans the surface of the screen to be detected, a frame of three-dimensional point cloud is obtained when the 2D laser line scans the surface of the screen to be detected, and at time t2, a frame of three-dimensional point cloud is obtained when the 2D laser line scans the surface of the screen to be detected, and ..., at time tn, a frame of three-dimensional point cloud is obtained when the 2D laser line scans the surface of the screen to be detected. If the scanning of the entire upper surface of the screen to be detected is completed after the end of time tn, the above multiple frames of three-dimensional point clouds are merged to obtain the three-dimensional point cloud of the screen corresponding to the screen to be detected. Moreover, the obtained three-dimensional point cloud of the screen is finally uploaded to the server for further processing. It can be seen that the use of 2D laser lines to obtain the three-dimensional point cloud of the screen has the advantages of obtaining high-precision distance information and not being restricted by lighting conditions.

[0087] Among them, the acquisition of the three-dimensional point cloud of the screen to be detected based on the laser radar is essentially laser three-dimensional imaging. More specifically, the distance information (Z axis) is obtained by measuring the flight time of the light pulse or modulated signal between the radar and the target, and the orientation information (X, Y axis) in the plane perpendicular to the light beam direction is obtained through scanning or multi-point corresponding measurement.

[0088] Because the screen is characterized by its multiple meshes and multiple weld points, when the 2D laser line hits the mesh area, it exceeds the effective laser range and no points are formed in the 3D point cloud of the screen. However, when the 2D laser line hits the screen area, it loses the effective laser range and can obtain a multi-frame 3D point cloud of the screen to be inspected.

[0089] The three-dimensional point cloud screening module 102 is used to screen out outliers and noise points in the mesh three-dimensional point cloud based on local area statistics to obtain a screened three-dimensional point cloud.

[0090] In an embodiment of the present application, after the initial sieve three-dimensional point cloud is obtained, there will be some outliers (which can be understood as wild points) and some noise points. At this time, the local field statistics technology can be used to filter out the wild points and noise points in the sieve three-dimensional point cloud to obtain the filtered three-dimensional point cloud.

[0091] In one embodiment, the 3D point cloud screening module 102 is specifically configured to:

[0092] Obtaining a domain smoothing range corresponding to the local domain statistics; wherein the domain smoothing range is a size of 3*3;

[0093] Each pixel in the mesh three-dimensional point cloud is locally smoothed based on the domain smoothing range to obtain a screened three-dimensional point cloud.

[0094] Among them, the local area statistics method can also be understood as a local statistics method, which is a point operation method for the spatial domain, and its function is to perform image enhancement. The local statistics method specifically includes the local smoothing method, which replaces the original grayscale value of each pixel in the image with the grayscale average value of each pixel in its neighborhood to achieve image smoothing. For example, it is common to use a pixel as the center point and replace the original grayscale value of the pixel with the grayscale average value of the remaining 8 pixels in a 3*3 size around it. It can be seen that based on the above method, the image is finally smoothed and the wild points and noise in the image are removed.

[0095] The two-dimensional projection module 103 is used to project the three-dimensional point cloud after screening along a preset Z axis and perform grid processing to obtain a two-dimensional image, and to expand the two-dimensional image based on a preset graphic mask to obtain a screen gap connection image.

[0096] In the embodiments of the present application, projecting the filtered 3D point cloud along a preset Z-axis is a transformation that projects the 3D point cloud onto a 2D plane. Specifically, the transformation involves multiplying the coordinates of each 3D point cloud in the filtered 3D point cloud by a transformation matrix and then projecting it onto a 2D plane, thereby achieving a dimensionality reduction transformation. After completing the 2D projection and performing image expansion processing, the screen rebar pixels can be rendered connected; otherwise, each screen mesh appears as an independent block.

[0097] In one embodiment, the two-dimensional projection module 103 is specifically configured to:

[0098] Projecting the filtered three-dimensional point cloud along the Z axis to obtain a Z-axis projection image;

[0099] The Z-axis projection image is gridded with 1 / 8 of the diameter of the cross section of each steel bar in the screen to be detected as the length and width of a single pixel to obtain a two-dimensional image.

[0100] In an embodiment of the present application, to achieve two-dimensional projection, a transformation matrix corresponding to the Z-axis projection is first obtained. The coordinates of each 3D point cloud in the filtered 3D point cloud (generally three-dimensional coordinates) are then multiplied by the transformation matrix and projected onto a two-dimensional plane (after projection onto the two-dimensional plane, the coordinates of each 3D point cloud become two-dimensional coordinates) to obtain a Z-axis projection image. More specifically, after obtaining the coordinates of each 3D point cloud in the filtered 3D point cloud, the Z-axis coordinates of each 3D point cloud are set to zero and only the X-axis and Y-axis coordinates are retained. This allows the two-dimensional coordinates corresponding to the coordinates of each 3D point cloud in the filtered 3D point cloud to be obtained, thereby obtaining a Z-axis projection image.

[0101] Once the Z-axis projection image is obtained, it needs to be gridded. Specifically, the Z-axis projection image is gridded with a pixel length and width of 1 / 8 the diameter of each steel bar section in the screen to be inspected, resulting in a two-dimensional image. Furthermore, during the gridding process, if a point is in the grid, the pixel is set to 1, otherwise it is set to 0.

[0102] In order to ensure that the mesh holes in the Z-axis projection image are connected rather than independent blocks, the Z-axis projection image needs to be further expanded so that the mesh steel bar pixels appear connected.

[0103] The screen mesh acquisition module 104 is configured to process the screen gap connectivity image based on region growing to obtain a region growing image, a screen mesh set in the region growing image, and a mesh corner point set corresponding to each screen mesh in the screen mesh set.

[0104] In the embodiment of the present application, after obtaining the screen gap connection image, in order to better identify the screen mesh, the screen gap connection image can be processed based on region growth to obtain a region growth image. Since the region growth image makes the screen mesh area clearer, please refer to Figure 2b Therefore, the sieve mesh set in the region growing image and the mesh corner point set corresponding to each sieve mesh in the sieve mesh set can be obtained based on the region growing image.

[0105] In one embodiment, the screen mesh size acquisition module 104 is specifically configured to:

[0106] Aggregate the pixels or sub-regions in the mesh gap connected image to satisfy the conditions corresponding to the region growing process to obtain a region growing image, a mesh set in the region growing image, and a mesh corner point set corresponding to each mesh in the mesh set.

[0107] Region growing is the process of aggregating pixels or sub-regions into larger regions based on a pre-defined criterion, and it is necessary to ensure that the segmented regions meet the following conditions 1) to 5):

[0108] 1)U(I i )=I;

[0109] 2) Ii is a connected region, i = 1, 2, 3, ... K;

[0110] 3)I i ∩I j = empty set, for any i, j; i≠j;

[0111] 4)P(I i)=True, for i=1,2,……K;where P(I i )=True means the prismatic mesh P(I i ) is not incremented and marked as True;

[0112] 5)I(P i UP j )=False, i≠j; where I(P i UP j ) represents the mesh P i and mesh P j After merging growth, mark it as False;

[0113] Wherein, I represents the mesh gap connected image, which can be divided into I1, I2, ..., I K These sub-regions have no intersecting parts. Therefore, sub-regions are selected in sequence in the mesh gap connection image I for region growth until all mesh pixels have completed growth.

[0114] The welding point identification module 105 is configured to determine a two-dimensional coordinate set of welding points according to the transverse corner point directions of the mesh set of the sieve and the mesh spacing of the sieve mesh.

[0115] In the embodiment of the present application, when the following Figure 2b The region growth image shown in FIG. 1 is shown in FIG. 2 , and the mesh set and the mesh corner point set corresponding to each mesh in the mesh set are known (for example, Figure 2b Each diamond-shaped blank area in the figure represents a mesh hole of the screen, and each mesh hole of the screen has 4 mesh corner points. The two-dimensional coordinates of the weld point can be determined from top to bottom and from left to right (the direction from left to right corresponds to the direction of the transverse corner points). Specifically, a center point determined by two adjacent mesh holes in the same transverse direction can be regarded as a weld point. More specifically, the center point determined by the left corner point of the mesh hole on the left and the right corner point of the mesh hole on the right of the two adjacent mesh holes can be regarded as a weld point. It can be seen that by determining the weld point based on the center points of adjacent mesh holes, the specific positions of multiple weld points can be obtained more quickly and accurately.

[0116] In one embodiment, the welding point identification module 105 is specifically configured to:

[0117] If there is a mesh corner point set in the mesh set with the smallest distance between the mesh center point coordinates and the origin, obtain the corresponding target mesh, and record the target mesh as the first row and first column mesh P 1，1 , and get the first row and first column of the screen mesh P 1，1The corresponding mesh corner point set; among them, the first row and first column of the screen mesh P 1，1 Located at row 1 and column 1 in the region growing image;

[0118] In the region growth image, search from left to right in the horizontal corner direction for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 ; Wherein, the initial values ​​of i and j are both 1, and the value range of i is [1, M], the value range of j is [1, N], M is a positive integer and N is a positive integer;

[0119] If it is determined that j+1 does not exceed N, based on the mesh size P of the i-th row and j-th column i，j The left corner coordinates and the mesh size P of the sieve in the i-th row and j+1th column i，j+1 The center point of the right corner coordinates obtains the two-dimensional coordinates S of the welding point in row i and column j i，j ;

[0120] Increment j by 1;

[0121] If it is determined that j does not exceed N, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 Steps;

[0122] If it is determined that j exceeds N, increment i by 1 and reset j to 1;

[0123] If it is determined that i does not exceed M, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 Steps;

[0124] If it is determined that i exceeds M, obtain the two-dimensional coordinates S of the welding point in row 1 and column 1 1，1 To the two-dimensional coordinate S of the welding point in row M and column N M，N Compose the two-dimensional coordinate set of welding points.

[0125] In the examples of this application, please refer to Figure 3 If the point in the upper left corner of the region growing image is regarded as the origin, it is first necessary to obtain the sieve mesh closest to the upper left corner origin in the sieve mesh set as the target sieve mesh, and record the target sieve mesh as the first row and first column sieve mesh P 1，1 , where the first row and first column of the screen mesh size P 1，1Located in the 1st row and 1st column of the region growth image. Since each mesh hole is considered as a diamond shape, the mesh hole P in the first row and first column is 1，1 The corresponding mesh corner point set includes 4 corner points (such as the left corner point, the upper corner point, the right corner point and the lower corner point). 1，1 Then, it is necessary to locate the adjacent mesh of the screen that is also in the first row, specifically, the mesh of the screen P in the first row and the second column. 1，2 When the first row and first column of the screen mesh P is located 1，1 and the first row and second column of the screen mesh P 1，2 Then, based on the first row and first column of the screen mesh P 1，1 The left corner point and the first row and second column mesh P 1，2 The center point of the right corner of the right corner is used to determine the two-dimensional coordinate S of the welding point in the first row and first column 1，1 Similarly, get the mesh size P in the first row and third column 1，3 , based on the first row and second column of the screen mesh P 1，2 The left corner point and the first row and third column of the sieve mesh P 1，3 The center point of the right corner of the 2D coordinate S of the welding point in row 1 and column 2 is determined by 1，2 Similarly, after determining the 2D coordinates of all weld points in the first row, we proceed to the second row, starting from the leftmost mesh of the second row and working our way from left to right to determine the 2D coordinates of all weld points in the second row. The 2D coordinates of the weld points in the second row and all subsequent rows are determined using the same method as the 2D coordinates of the weld points in the first row. This indicates that, based on this row-by-row and column-by-column inspection method, the 2D coordinates of all weld points can be accurately determined, thus forming a 2D coordinate set of weld points.

[0126] The welding point checking module 106 is used to constrain the two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set so that the spacing and angle of the X-shaped neighbors are approximately consistent, and obtain a constraint result.

[0127] In this embodiment, after all welding points in the region growing image are obtained, the two-dimensional coordinates of each welding point need to be constrained so that the spacing and angle of the X-shaped neighbors are approximately consistent to obtain a constrained result.

[0128] For example, reference Figure 4 , with the two-dimensional coordinates S of the welding point in the second row and second column 2，3 As an example of the consistency constraint, obtain the two-dimensional coordinates S of the welding point in the second row and third column 2，2 The two-dimensional coordinates S of the welding point in the first row and second column of the X-type neighbor 1，2 , the first row and third column two-dimensional coordinates of the welding point S 1，3 , the two-dimensional coordinates S of the welding point in the third row and second column 3，2And the two-dimensional coordinates S of the welding point in the third row and third column 3，3 . With S 1，2 and S 2，3 Get the first vector X1, with S 1，3 and S 2，3 Get the second vector X2, with S 3，2 and S 2，3 Get the third vector X3, with S 3，3 and S 2，3 Obtain the fourth vector X4. If the angle between the first vector X1 and the second vector X2 is 90°, the angle between the second vector X2 and the third vector X4 is 90°, the angle between the fourth vector X4 and the third vector X3 is 90°, the angle between the third vector X3 and the first vector X1 is 90°, the angle between the first vector X1 and the fourth vector X4 is 180°, and the angle between the second vector X2 and the third vector X3 is 180°, then it can be determined that the two-dimensional coordinates of the above five welding points meet the X-type nearest neighbor spacing and angle approximate consistency constraints and no adjustment is required. If the above conditions are not met, appropriate adjustments are required to meet the X-type nearest neighbor spacing and angle approximate consistency constraints.

[0129] When the two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set are constrained to be approximately consistent with the spacing and angle of X-shaped neighbors to obtain a constraint result, the two-dimensional coordinates of each welding point included in the constraint result are adjusted to form the welding point trajectory data corresponding to the screen to be detected.

[0130] It can be seen that the embodiment of the present application realizes the acquisition of the screen mesh set in the screen to be inspected based on the two-dimensional projection transformation of the three-dimensional point cloud, and determines the two-dimensional coordinates of the welding points based on the corner point set of each screen mesh in the screen mesh set. The whole process is automatically executed without the need for human intervention, thereby improving the efficiency of locating the welding points in the screen to be inspected.

[0131] This application also provides a processing device, see Figure 6 , Figure 6 A schematic diagram of the structure of the processing device of the present application is shown. Specifically, the processing device provided by the present application includes a processor, which is used to implement the following when executing the computer program stored in the memory. Figure 1 The steps in the corresponding embodiment; or, the processor is used to execute the computer program stored in the memory to implement the following Figure 5 The functions of each module in the corresponding embodiment.

[0132] For example, a computer program may be divided into one or more modules / units, one or more of which are stored in a memory and executed by a processor to implement the present application. One or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in a computer device.

[0133] A processing device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the illustrations are merely examples of processing devices and do not limit the processing device. The processing device may include more or fewer components than shown, or a combination of certain components, or different components. For example, the processing device may also include input / output devices, network access devices, and buses, and the processor, memory, input / output devices, and network access devices are connected via a bus.

[0134] The processor can be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the processing device and connects the various parts of the entire processing device using various interfaces and lines.

[0135] The memory can be used to store computer programs and / or modules. The processor implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store an operating system, at least one application required for a function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created based on the use of the processing device (such as audio data, video data, etc.). In addition, the memory can include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0136] The display screen is used to display characters of at least one character type output by the input and output unit.

[0137] Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices, processing equipment and corresponding modules can refer to the following. Figure 1 The description in the corresponding embodiments will not be repeated here in detail.

[0138] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be accomplished by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0139] To this end, an embodiment of the present application provides a computer-readable storage medium, which stores multiple instructions, which can be loaded by a processor to execute the present application as follows: Figure 1 For the steps in the corresponding embodiment, the specific operations can refer to Figure 1 The description in the corresponding embodiments will not be repeated here.

[0140] The computer-readable storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0141] Due to the instructions stored in the computer readable storage medium, the present application can be executed as follows: Figure 1 The steps in the corresponding embodiments can therefore be implemented as follows: Figure 1 The beneficial effects that can be achieved in the corresponding embodiments are detailed in the previous description and will not be repeated here.

[0142] The above is a detailed introduction to the automatic welding trajectory point detection method, device and storage medium provided by the present application. The principles and implementation methods of the present application are explained using specific examples in the embodiments of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for detecting automatic welding track points of a screen mesh, characterized in that: The method comprises: Obtaining a three-dimensional point cloud of the sieve corresponding to the sieve to be tested; Based on local area statistics, outliers and noise points in the mesh three-dimensional point cloud are screened out to obtain a screened three-dimensional point cloud; Projecting the filtered three-dimensional point cloud along a preset Z axis and performing gridding processing to obtain a two-dimensional image, and expanding the two-dimensional image based on a preset graphic mask to obtain a screen gap connection image; Processing the mesh gap connectivity image based on region growing to obtain a region growing image, a mesh set in the region growing image, and a mesh corner point set corresponding to each mesh in the mesh set; Determining a two-dimensional coordinate set of welding points according to the transverse corner point directions of the screen mesh set and the mesh spacing of the screen mesh includes: If there is a mesh corner point set in the mesh set with the smallest distance between the mesh center point coordinates and the origin, obtain the corresponding target mesh, and record the target mesh as the first row and first column mesh P 1，1 , and get the first row and first column of the screen mesh P 1，1 The corresponding mesh corner point set; among them, the first row and first column of the screen mesh P 1，1 Located at row 1 and column 1 in the region growing image; In the region growth image, search from left to right in the horizontal corner direction for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 ; The initial values ​​of i and j are both 1, and the value range of i is [1, M], and the value range of j is [1, N], where M is a positive integer and N is a positive integer; If it is determined that j+1 does not exceed N, based on the mesh size P of the i-th row and j-th column i，j The left corner coordinates and the mesh size P of the sieve in the i-th row and j+1th column i，j+1 The center point of the right corner coordinates obtains the two-dimensional coordinates S of the welding point in row i and column j i，j ; Increment j by 1; If it is determined that j does not exceed N, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 Steps; If it is determined that j exceeds N, increment i by 1 and reset j to 1; If it is determined that i does not exceed M, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 Steps; If it is determined that i exceeds M, obtain the two-dimensional coordinates S of the welding point in row 1 and column 1 1，1 To the two-dimensional coordinate S of the welding point in row M and column N M，N Composing a two-dimensional coordinate set of welding points; The two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set are constrained so that the spacing and angle of the X-shaped neighbors are approximately consistent, and a constraint result is obtained.

2. The method according to claim 1, characterized in that The projecting of the filtered three-dimensional point cloud along a preset Z axis and performing gridding processing to obtain a two-dimensional image includes: Projecting the filtered three-dimensional point cloud along the Z axis to obtain a Z-axis projection image; The Z-axis projection image is gridded with 1 / 8 of the diameter of the cross section of each steel bar in the screen to be detected as the length and width of a single pixel to obtain a two-dimensional image.

3. The method according to claim 1, characterized in that The process of processing the screen gap connectivity image based on region growing to obtain a region growing image, a screen mesh set in the region growing image, and a mesh corner point set corresponding to each screen mesh in the screen mesh set includes: Aggregate the pixels or sub-regions in the mesh gap connected image to satisfy the conditions corresponding to the region growing process to obtain a region growing image, a mesh set in the region growing image, and a mesh corner point set corresponding to each mesh in the mesh set.

4. The method according to claim 1, wherein The method of removing outliers and noise points from the mesh three-dimensional point cloud based on local area statistics to obtain a three-dimensional point cloud after removal includes: Obtaining a domain smoothing range corresponding to the local domain statistics; wherein the domain smoothing range is 3*3 in size; Each pixel point in the screen three-dimensional point cloud is locally smoothed based on the domain smoothing range to obtain a screened three-dimensional point cloud.

5. A screen mesh automatic welding trajectory point detection device, characterized in that: include: A three-dimensional point cloud acquisition module is used to obtain a three-dimensional point cloud of the screen corresponding to the screen to be tested; A three-dimensional point cloud screening module is used to screen out wild points and noise points in the three-dimensional point cloud of the screen based on local field statistics to obtain a screened three-dimensional point cloud; a two-dimensional projection module, configured to project the three-dimensional point cloud after screening along a preset Z axis and perform gridding processing to obtain a two-dimensional image, and to dilate the two-dimensional image based on a preset graphic mask to obtain a screen gap connection image; a screen mesh acquisition module, configured to process the screen gap connectivity image based on region growing to obtain a region growing image, a screen mesh set in the region growing image, and a mesh corner point set corresponding to each screen mesh in the screen mesh set; A welding point identification module is used to determine a two-dimensional coordinate set of welding points based on the transverse corner point directions of the screen mesh set and the mesh spacing of the screen mesh; comprising: If there is a mesh corner point set in the mesh set with the smallest distance between the mesh center point coordinates and the origin, obtain the corresponding target mesh, and record the target mesh as the first row and first column mesh P 1，1 , and get the first row and first column of the screen mesh P 1，1 The corresponding mesh corner point set; among them, the first row and first column of the screen mesh P 1，1 Located at row 1 and column 1 in the region growing image; In the region growth image, search from left to right in the horizontal corner direction for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 ; The initial values ​​of i and j are both 1, and the value range of i is [1, M], and the value range of j is [1, N], where M is a positive integer and N is a positive integer; If it is determined that j+1 does not exceed N, based on the mesh size P of the i-th row and j-th column i，j The left corner coordinates and the mesh size P of the sieve in the i-th row and j+1th column i，j+1 The center point of the right corner coordinates obtains the two-dimensional coordinates S of the welding point in row i and column j i，j ; Increment j by 1; If it is determined that j does not exceed N, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 Steps; If it is determined that j exceeds N, increment i by 1 and reset j to 1; If it is determined that i does not exceed M, return to the execution of searching the area growth image in the horizontal corner direction and from left to right for the mesh P of the i-th row and j-th column. i，j The mesh size P of the sieve in row i and column j+1 with the minimum spacing i，j+1 Steps; If it is determined that i exceeds M, obtain the two-dimensional coordinates S of the welding point in row 1 and column 1 1，1 To the two-dimensional coordinate S of the welding point in row M and column N M，N Composing a two-dimensional coordinate set of welding points; The welding point checking module is used to constrain the two-dimensional coordinates of each welding point in the welding point two-dimensional coordinate set to be approximately consistent with the spacing and angle of the X-shaped neighbors to obtain a constraint result.

6. The device according to claim 5, characterized in that The two-dimensional projection module is used for: Projecting the filtered three-dimensional point cloud along the Z axis to obtain a Z-axis projection image; The Z-axis projection image is gridded with 1 / 8 of the diameter of the cross section of each steel bar in the screen to be detected as the length and width of a single pixel to obtain a two-dimensional image.

7. The device according to claim 5, characterized in that The screen mesh acquisition module is used for: Aggregate the pixels or sub-regions in the mesh gap connected image to satisfy the conditions corresponding to the region growing process to obtain a region growing image, a mesh set in the region growing image, and a mesh corner point set corresponding to each mesh in the mesh set.

8. A processing device, characterized in that The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the method according to any one of claims 1 to 4 is executed.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the method according to any one of claims 1 to 4.

Images