Continuous casting slab size detection method and system based on line laser binocular camera

Abstract

The invention discloses a continuous casting slab size detection method and system based on a line laser binocular camera. Comprising the following steps: acquiring continuous casting slab laser point cloud data scanned by a laser binocular camera; carrying out clustering analysis on the laser point cloud data, and respectively calculating edge points and center points in a point cloud cluster; a plurality of line segments are extracted from the laser point cloud data, the length of each line segment is calculated, the length mean value and the standard deviation of all the line segments are calculated, the standard length range corresponding to each line segment is calculated, the length of each line segment is compared with the corresponding standard length range, and if the length of each line segment is not in the range, the line segments are removed; fitting an optimal straight line equation according to the central point, and calculating an inclination angle of an optimal straight line under the current coordinate system; and converting the width data and the length data of the point cloud data of the continuous casting plate to the positive direction of the current coordinate system by using the inclination angle of the optimal straight line under the current coordinate system to obtain the size of the continuous casting plate. According to the method, the size detection accuracy of the continuous casting slab is improved.

Description

Technical Field

[0001] This application relates to the technical field of continuous casting slab size detection, and particularly to a continuous casting slab size detection method and system. Background Art

[0002] When a steel mill produces continuous casting slabs, it is necessary to detect the size data of the high-temperature continuous casting slabs in real time to confirm whether the produced continuous casting slabs meet the standards. Usually, due to the influence of high temperature, this measurement method needs to use non-contact equipment for indirect measurement to obtain the length and width size data of the slab, thus avoiding direct contact between people and high-temperature continuous casting slabs.

[0003] In the prior art, an infrared thermal imager is usually arranged above or on the side of the continuous casting line to collect the surface temperature image of the slab in real time. Since the edge dissipates heat faster, the temperature of the slab edge is usually lower than that of the middle part. This temperature difference provides a basis for edge recognition: the system uses the temperature difference between the slab edge and the surrounding environment or inside the slab in the infrared image, and identifies the area with a significant change in temperature gradient through an algorithm, that is, the left and right edges of the slab. Usually, the edge temperature is lower and the inner temperature is higher. An image processing algorithm (such as Canny edge detection) is used to extract the contour of the temperature mutation area in the infrared image and mark the edge. The infrared imaging device is calibrated to be able to convert the pixel distance in the image into the actual physical distance. By calculating the number of pixels between the left and right edges and combining the field of view angle and height of the imager, the actual size of the slab can be obtained. In some cases, heat conduction or convection may cause a small temperature difference between the slab edge and the inside, resulting in an insufficiently significant temperature gradient, thereby affecting the accuracy of edge detection. The infrared imaging algorithm relies on the temperature difference to identify the slab edge, but the temperature distribution on the slab surface is not uniform. Especially in different cooling stages, the temperature difference between the edge and the middle part may not be obvious. This non-obvious temperature difference will cause the edge to be blurred, making it difficult for the algorithm to accurately distinguish the actual edge position of the slab, thereby causing an edge positioning deviation and resulting in inaccurate width measurement. Especially when the edge temperature is close to the internal temperature, significant size calculation deviations may occur. Summary of the Invention

[0004] Based on this, in view of the above technical problems, a continuous casting slab size detection method and system based on a line laser binocular camera are provided to solve the problem of incorrect calculation of the size of high-temperature continuous casting billets in the prior art.

[0005] In a first aspect, a continuous casting slab size detection method based on a line laser binocular camera, the method includes:

[0006] When it is determined that the continuous casting slab reaches the scanning area of the laser binocular camera, obtain the laser point cloud data of the continuous casting slab scanned by the laser binocular camera;

[0007] Perform clustering analysis on the laser point cloud data, and calculate the edge points and central points in the clustering clusters of the point cloud respectively;

[0008] Extract multiple line segments from the laser point cloud data, calculate the length of each line segment, calculate the mean value and standard deviation of the lengths of all line segments of the continuous casting slab, calculate the standard length range corresponding to each line segment according to the mean value and standard deviation of the length of each line segment, compare the length of each line segment with the corresponding standard length range, and if it is not within the range, eliminate it to obtain the laser point cloud data with abnormal point clouds eliminated;

[0009] Fit the optimal straight line equation according to the central points of the clustering clusters after eliminating abnormal point clouds, and calculate the inclination angle of the optimal straight line in the current coordinate system;

[0010] Calculate the width data and length data of the continuous casting slab point cloud data according to the coordinates of the edge points of the clustering clusters after eliminating abnormal point clouds, and convert the width data and length data of the continuous casting slab point cloud data to the positive direction of the current coordinate system by using the inclination angle of the optimal straight line in the current coordinate system to obtain the accurate dimensions of the continuous casting slab.

[0011] In the above solution, optionally, before performing clustering analysis on the laser point cloud data, it further includes:

[0012] Perform Gaussian filtering on the laser point cloud data to remove the point cloud data that does not meet the required height range.

[0013] In the above solution, optionally, when determining that the continuous casting slab reaches the scanning area of the laser binocular camera, it includes: receiving the data of the encoder set on the roller shaft, and judging that the continuous casting slab reaches the scanning area of the laser binocular camera according to the encoder data.

[0014] In the above solution, optionally, performing clustering analysis on the laser point cloud data includes:

[0015] Perform clustering analysis on the laser point cloud data, retain the points that meet the clustering threshold, and continuously perform iterative clustering until the clustering center no longer changes or reaches the maximum number of iterations to obtain the clustering clusters.

[0016] In the above solution, optionally, the standard length range corresponding to each line segment is: [μ - kσ, μ + kσ], where μ is the mean value, σ is the standard deviation, and k is a coefficient set according to the actual situation.

[0017] In the above solution, optionally, fitting the optimal straight line equation according to the central points of the clustering clusters after eliminating abnormal point clouds includes:

[0018] Fit the equation according to the coordinates of the central points of the clustering clusters after eliminating abnormal point clouds by using the RANSAC algorithm, and select the optimal straight line equation.

[0019] In the above solution, optionally, after obtaining the laser point cloud data of the continuous casting slab scanned by the laser binocular camera, the following steps are further included:

[0020] According to the ROI area set by the user, only the point cloud data under the current region of interest is retained.

[0021] In a second aspect, a continuous casting slab size detection system based on a line laser binocular camera, the system includes:

[0022] A continuous casting slab laser point cloud data acquisition module: used to obtain the laser point cloud data of the continuous casting slab scanned by the laser binocular camera when it is determined that the continuous casting slab reaches the scanning area of the laser binocular camera;

[0023] A clustering analysis module: used to perform clustering analysis on the laser point cloud data, and calculate the edge points and center points in the clustering clusters of the point cloud respectively;

[0024] An abnormal point cloud removal module: used to extract multiple line segments from the laser point cloud data, calculate the length of each line segment, calculate the average value and standard deviation of the lengths of all line segments of the continuous casting slab, calculate the standard length range corresponding to each line segment according to the average value and standard deviation of the length of each line segment, compare the length of each line segment with the corresponding standard length range, and if it is not within the range, remove it to obtain the laser point cloud data with abnormal point clouds removed;

[0025] An inclination angle calculation module: used to fit the optimal straight line equation according to the center point of the clustering cluster after removing abnormal point clouds, and calculate the inclination angle of the optimal straight line in the current coordinate system;

[0026] A continuous casting slab size calculation module: used to calculate the width data and length data of the continuous casting slab point cloud data according to the coordinates of the edge points of the clustering cluster after removing abnormal point clouds, and convert the width data and length data of the continuous casting slab point cloud data to the positive direction of the current coordinate system by using the inclination angle of the optimal straight line in the current coordinate system to obtain the accurate size of the continuous casting slab.

[0027] In a third aspect, a computer device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method for detecting the size of a continuous casting slab based on a line laser binocular camera described in the first aspect above are implemented.

[0028] In a fourth aspect, a computer program product includes computer programs / instructions, and when the computer programs / instructions are executed by a processor, the steps of the method for detecting the size of a continuous casting slab based on a line laser binocular camera described in the first aspect above are implemented.

[0029] This application has at least the following beneficial effects:

[0030] In this application, the Euclidean clustering algorithm is used to cluster each frame of point cloud data, so that the point cloud becomes a whole from parts. Then, combined with the minimum bounding rectangle algorithm, the edge points of each clustering cluster are extracted, and the width and length of a single clustering cluster are indirectly restored. Due to the inclination angle between the continuous casting slab and the laser line during scanning, there are abnormalities at the head and tail of the point cloud of the scanned continuous casting slab. Therefore, the point cloud is filtered by calculating the variance of the point cloud width, and the invalid point cloud is removed. In addition, due to factors such as the angular position inclination of the installation of the laser binocular camera and the shape of the scanned continuous casting slab not conforming to the actual situation, effective and accurate dimension data cannot be directly obtained. The ransac algorithm is combined to fit the straight line equation formed by the center points of each cluster of point cloud data, and the slope of the straight line equation is used as the inclination angle of the continuous casting slab. If the inclination angle of the detected continuous casting slab is greater than the set threshold, the actual dimension is the product of the current measured dimension and the sine value of the inclination angle, so as to obtain the accurate dimension of the continuous casting slab. Description of the Drawings

[0031] Figure 1 It is a schematic flow chart of a method for detecting the size of a continuous casting slab based on a line laser binocular camera provided by an embodiment of this application;

[0032] Figure 2 It is a specific schematic flow chart of a method for detecting the size of a continuous casting slab based on a line laser binocular camera provided by an embodiment of this application;

[0033] Figure 3 It is a schematic diagram of the setting of the laser binocular camera in an embodiment of this application;

[0034] Figure 4 It is a flow chart of point cloud data acquisition of the laser binocular camera provided by an embodiment of this application. Detailed Embodiments

[0035] In order to make the purpose, technical solutions and advantages of this application clearer, the following further details this application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0036] In one embodiment, as Figure 1 and Figure 2 shown, a method for detecting the size of a continuous casting slab based on a line laser binocular camera is provided, and the method includes:

[0037] Step S1: When it is determined that the continuous casting slab reaches the scanning area of the laser binocular camera, obtain the laser point cloud data of the continuous casting slab scanned by the laser binocular camera.

[0038] In step S1, the laser binocular camera is installed directly above the slab conveyor belt, as Figure 3The figure shows a schematic diagram of the setting of the laser binocular camera. The object transported by the scene conveyor belt is a high-temperature continuous casting slab. A line laser sensor is fixedly installed above the slab. After receiving the point position signal sent by the encoder, the binocular camera behind the line laser starts to collect image data and passes it to the algorithm end for processing and calculation of the current slab size data. Before use, level the viewing angle range of the binocular camera and the slab so that the detection area can completely cover the continuous casting slab. When the binocular camera sensor is turned on, the camera sensor exposure, gain, detection ROI and trigger mode are set through its software program. The binocular camera extracts the laser point cloud data in the current field of view, and calculates the three-dimensional space coordinates in the camera coordinate system based on the binocular ranging, and then retains only the point cloud under the current area of interest according to the set ROI area.

[0039] The judgment of when the continuous casting slab reaches the scanning area of the laser binocular camera is based on the encoder data and signal status. According to the encoder data, when the continuous casting slab reaches the scanning area of the laser binocular camera, the current point cloud data at the corresponding moment is obtained and saved.

[0040] Step S2: performing cluster analysis on the laser point cloud data, and respectively calculating edge points and center points in the point cloud clusters.

[0041] In step S2, each frame of laser point cloud data is clustered in Euclidean order, and the edge points and center points in the point cloud clusters are calculated respectively. The Euclidean distance measurement method is defined as shown in the following figure, where x and y are two three-dimensional point coordinates:

[0042]

[0043] For real scene data, it is necessary to cluster the centers of the three-dimensional point cloud data for each frame, retain the points that meet the clustering threshold, and continuously iterate the clustering until the cluster center no longer changes or the maximum number of iterations is reached.

[0044] Step S3: extract multiple line segments from the laser point cloud data, calculate the length of each line segment, calculate the mean length and standard deviation of all line segments of the continuous casting slab, calculate the standard length range corresponding to each line segment according to the mean length and standard deviation of each line segment, compare the length of each line segment with the corresponding standard length range, and remove it if it is not within the range to obtain laser point cloud data with abnormal point cloud removed.

[0045] In step S3, for the abnormal detection of laser line segments, the variance distribution between the current line segment and the laser line segments in the adjacent front and rear regions is evaluated to eliminate invalid or abnormal point cloud data. Calculate its length based on the endpoint coordinates of the straight line. Use statistical methods to identify line segments with abnormal lengths, and calculate the mean and standard deviation of the lengths of all scanned line segments of the current slab. By setting a threshold, line segments with lengths exceeding μ + kσ (k is a constant) or less than μ - kσ are regarded as abnormal.

[0046]

[0047] Calculate the mean length of all scanned line segments through the following formula:

[0048]

[0049] Calculate the mean and standard deviation of the lengths of all scanned line segments of the current slab through the following formula:

[0050]

[0051] Step S4: Fit the optimal straight-line equation based on the center point of the clustering cluster after eliminating abnormal point clouds, and calculate the inclination angle of the optimal straight line in the current coordinate system.

[0052] In step S4, based on the RANSAC algorithm, for the center coordinate points of the clustering clusters of each frame of laser point cloud data belonging to the same continuous casting slab, fit the optimal straight-line equation and calculate the straight-line inclination angle in the current coordinate system. Input the center point coordinates of each laser line clustering cluster, use the RANSAC algorithm, and iterate multiple times to find the best-fitting straight line. The straight-line equation is usually expressed as y = mx + b. Where m is the slope, which represents the inclination angle of the continuous casting slab.

[0053] Step S5: Calculate the width data and length data of the continuous casting slab point cloud data based on the edge point coordinates of the clustering cluster after eliminating abnormal point clouds, and convert the width data and length data of the continuous casting slab point cloud data to the positive direction of the current coordinate system using the inclination angle of the optimal straight line in the current coordinate system to obtain the accurate dimensions of the continuous casting slab.

[0054] Calculate the width and length of the point cloud belonging to the continuous casting slab, and convert the continuous casting slab to the positive direction of the current coordinate system according to the inclination angle calculated in step S4, and output the dimension data.

[0055] Characteristics of the line laser binocular scanning sensor: It can calculate the coordinate information of the laser hitting the target more precisely, and can obtain the two-dimensional coordinates of the point cloud more accurately through pixel coordinates and calibration parameters, and then combine the binocular ranging algorithm to obtain the three-dimensional point cloud.

[0056] Since the point cloud data output by the line laser consists of discrete three-dimensional space point coordinates and lacks point sequence information, and there are angular deviations in the sensor installation, the shape of the continuously cast slab obtained by scanning does not match the actual situation, and accurate dimension data cannot be directly obtained. Therefore, it is necessary to indirectly calculate the dimension data of the continuously cast slab through self-developed algorithms in combination with the original point cloud data.

[0057] Therefore, in the above method for detecting the dimensions of a continuously cast slab based on a line laser binocular camera, the point cloud data of each frame is clustered by the Euclidean clustering algorithm to make the point cloud from parts into a whole, and then the edge coordinates of each clustering cluster are extracted in combination with the minimum bounding rectangle algorithm to indirectly restore the width and length of a single clustering cluster. Since there is an inclination angle between the continuously cast slab and the laser line during scanning, abnormal points exist at the head and tail of the point cloud of the continuously cast slab after scanning. Therefore, the point cloud is filtered by calculating the variance of the point cloud width to eliminate invalid point clouds. In addition, due to factors such as angular position inclination in the sensor installation and the shape of the scanned continuously cast slab not conforming to the actual situation, effective and accurate dimension data cannot be directly obtained. The RANSAC algorithm is combined to fit the straight line equation composed of the center point coordinates of each cluster of point cloud data, and the slope of the straight line equation is used as the inclination angle of the continuously cast slab. If the inclination angle of the detected continuously cast slab is greater than the set threshold, the actual dimension is the product of the current measured dimension and the sine value of the inclination angle, so as to obtain the accurate dimension of the continuously cast slab.

[0058] Compared with the prior art, in the line laser binocular vision measurement technical solution proposed by the present invention, a non-contact scanning method for continuously cast slabs in a high-temperature environment is established, and the final dimension data of the continuously cast slab is obtained through algorithm processing based on the point cloud data acquired by the sensor. In terms of measurement accuracy, the width error can be guaranteed to be within the range of (-2, +2) mm, and the length error range is within the range of (-20, +50) mm. The method of using laser scanning plus a binocular camera effectively avoids problems such as measurement limitation or inaccuracy caused by too high temperature, and also avoids problems such as poor robustness in using a single vision sensor for recognition. In addition, compared with image data, laser point cloud data can use less storage capacity of computer equipment. The storage capacity required for each continuously cast slab data is only about dozens of megabytes on average, and it also indicates that less computing resources are occupied.

[0059] In one embodiment, a system for detecting the dimensions of a continuously cast slab based on a line laser binocular camera is provided. The system includes:

[0060] A continuously cast slab laser point cloud data acquisition module: used to obtain the laser point cloud data of the continuously cast slab scanned by the laser binocular camera when it is determined that the continuously cast slab reaches the scanning area of the laser binocular camera;

[0061] A clustering analysis module: used to perform clustering analysis on the laser point cloud data and calculate the edge points and center points in the clustering clusters of the point cloud respectively;

[0062] Abnormal point cloud elimination module: It is used to extract multiple line segments from the laser point cloud data, calculate the length of each line segment, calculate the average value and standard deviation of the lengths of all line segments of the continuous casting slab, calculate the corresponding standard length range for each line segment according to the average value and standard deviation of the length of each line segment, compare the length of each line segment with the corresponding standard length range, and if it is not within the range, eliminate it to obtain the laser point cloud data with abnormal point clouds eliminated;

[0063] Tilt angle calculation module: It is used to fit the optimal straight line equation according to the center point of the clustering cluster after eliminating abnormal point clouds, and calculate the tilt angle of the optimal straight line in the current coordinate system;

[0064] Size calculation module of the continuous casting slab: It is used to calculate the width data and length data of the continuous casting slab point cloud data according to the edge point coordinates of the clustering cluster after eliminating abnormal point clouds, and convert the width data and length data of the continuous casting slab point cloud data to the positive direction of the current coordinate system by using the tilt angle of the optimal straight line in the current coordinate system to obtain the accurate size of the continuous casting slab.

[0065] For the specific limitations of a continuous casting slab size detection system based on a line laser binocular camera, reference can be made to the limitations of a continuous casting slab size detection method based on a line laser binocular camera in the above text, which will not be elaborated here. Each module in the above continuous casting slab size detection system based on a line laser binocular camera can be implemented in whole or in part by software, hardware and their combination. The above modules can be embedded in the processor of the computer device in hardware form or be independent of it, or be stored in the memory of the computer device in software form so that the processor can call and execute the operations corresponding to the above modules.

[0066] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be shown in Figure Y. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements the above method for detecting the size of a continuous casting slab based on a line laser binocular camera.

[0067] In one embodiment, a computer program product is further provided, including a computer program / instructions, which, when executed by the processor, involves all or part of the processes in the method of the above embodiment.

[0068] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0069] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0070] The above-described embodiments merely represent several implementation manners of this application. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of the patent of this application should be subject to the appended claims.

Claims

1. A method for detecting the size of a continuous casting slab based on a line laser binocular camera, characterized in that, The method includes: When it is determined that the continuous casting slab reaches the scanning area of the laser binocular camera, obtaining the laser point cloud data of the continuous casting slab scanned by the laser binocular camera; Performing clustering analysis on the laser point cloud data, and respectively calculating the edge points and central points in the clustering clusters of the point cloud; Extracting multiple line segments from the laser point cloud data, calculating the length of each line segment, calculating the average value and standard deviation of the lengths of all line segments of the continuous casting slab, calculating the standard length range corresponding to each line segment according to the average value and standard deviation of the length of each line segment, comparing the length of each line segment with the corresponding standard length range, and if it is not within the range, removing it to obtain the laser point cloud data with abnormal point cloud removed; Fitting the optimal straight line equation according to the central points of the clustering clusters after removing the abnormal point cloud, and calculating the inclination angle of the optimal straight line in the current coordinate system; Calculating the width data and length data of the continuous casting slab point cloud data according to the edge point coordinates of the clustering clusters after removing the abnormal point cloud, and converting the width data and length data of the continuous casting slab point cloud data to the positive direction of the current coordinate system by using the inclination angle of the optimal straight line in the current coordinate system to obtain the accurate dimensions of the continuous casting slab.

2. The method for detecting the size of a continuous casting slab based on a line laser binocular camera according to claim 1, wherein Before performing clustering analysis on the laser point cloud data, it further includes: Performing Gaussian filtering on the laser point cloud data to remove the point cloud data that does not meet the required height range.

3. The continuous casting slab size detection method based on a line laser binocular camera according to claim 1, characterized in that When it is determined that the continuous casting slab reaches the scanning area of the laser binocular camera, it includes: receiving the data of the encoder arranged on the roller shaft, and determining that the continuous casting slab reaches the scanning area of the laser binocular camera according to the encoder data.

4. The method for detecting the size of a continuous casting slab based on a line laser binocular camera according to claim 1, wherein Performing clustering analysis on the laser point cloud data includes: Performing clustering analysis on the laser point cloud data, retaining the points that meet the clustering threshold, and continuously performing iterative clustering until the clustering center no longer changes or reaches the maximum number of iterations to obtain the clustering clusters.

5. The method for detecting the size of a continuous casting slab based on a line laser binocular camera according to claim 1, wherein The standard length range corresponding to each line segment is: [μ - kσ, μ + kσ], where μ is the average value, σ is the standard deviation, and k is a coefficient set according to the actual situation.

6. The method for detecting the size of continuous casting slabs based on a line laser binocular camera according to claim 1, wherein Fitting the optimal straight line equation according to the central points of the clustering clusters after removing the abnormal point cloud includes: Fitting the equation according to the central point coordinates of the clustering clusters after removing the abnormal point cloud by using the RANSAC algorithm, and selecting the optimal straight line equation.

7. The method for detecting the size of a continuous casting slab based on a line laser binocular camera according to claim 1, characterized in that, After obtaining the laser point cloud data of the continuous casting slab scanned by the laser binocular camera, it further includes: According to the ROI area set by the user, only retaining the point cloud data in the current region of interest.

8. A continuous casting slab size detection system based on a line laser binocular camera, characterized in that, The system includes: Continuous casting slab laser point cloud data acquisition module: used to obtain the laser point cloud data of the continuous casting slab scanned by the laser binocular camera when it is determined that the continuous casting slab reaches the scanning area of the laser binocular camera; Clustering analysis module: used to perform clustering analysis on the laser point cloud data, and respectively calculate the edge points and central points in the clustering clusters of the point cloud; Abnormal point cloud elimination module: used to extract multiple line segments from the laser point cloud data, calculate the length of each line segment, calculate the average value and standard deviation of the lengths of all line segments of the continuous casting slab, calculate the corresponding standard length range for each line segment according to the average value and standard deviation of the length of each line segment, compare the length of each line segment with the corresponding standard length range, and eliminate it if it is not within the range to obtain the laser point cloud data with abnormal point cloud eliminated; Tilt angle calculation module: used to fit the optimal straight line equation according to the center point of the clustering cluster after eliminating the abnormal point cloud, and calculate the tilt angle of the optimal straight line in the current coordinate system; Size calculation module of the continuous casting slab: used to calculate the width data and length data of the continuous casting slab point cloud data according to the edge point coordinates of the clustering cluster after eliminating the abnormal point cloud, and convert the width data and length data of the continuous casting slab point cloud data to the positive direction of the current coordinate system by using the tilt angle of the optimal straight line in the current coordinate system to obtain the accurate size of the continuous casting slab.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

10. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, it implements the steps of the method according to any one of claims 1 to 8.

Images