Methods, systems, and storage media for detecting unilateral overflow defects in steel coils.

Abstract

This invention provides a method, system, and storage medium for detecting unilateral overflow defects in steel coils. The method includes: Step S1, obtaining a point cloud model of the steel coil; Step S2, obtaining planar regions in the point cloud model using a 3D cloud extraction algorithm; Step S3, segmenting a first point cloud set and a second point cloud set on the inner and outer cylindrical surfaces of the steel coil in the point cloud model; Step S4, calculating the overflow distance values ​​from the first and second point clouds to the planar regions; Step S5, applying a first threshold to the overflow distance values ​​to extract target point clouds from the first point cloud set; Step S6, performing principal component analysis on the target point clouds to determine whether a unilateral overflow defect exists in the steel coil. By using the steel coil point cloud model, 3D information of the unilateral overflow portion can be obtained in real time, and the overflow distance value of the point cloud can be calculated, allowing for accurate determination of the overflow distance of the current unilateral overflow portion of the steel coil, resulting in more reliable detection results.

Description

Technical Field

[0001] This invention relates to the application of point cloud data processing in steel coil defect detection, and in particular to a method, system and storage medium for detecting unilateral overflow defects in steel coils. Background Technology

[0002] In traditional steel mills, steel coil production is often affected by manufacturing issues during the coiling process, resulting in a defect where one side of the coil overflows. This overflow refers to a portion of the coil protruding from the normal end face, leading to a decrease in the yield rate.

[0003] Traditional methods for detecting one-sided overflow defects involve capturing images of the steel coil with a camera, comparing the images with preset raw data (or reference images), identifying differences, segmenting these differences, and then using a threshold to determine if overflow exceeds the acceptable range. This approach has inherent flaws:

[0004] 1. Two-dimensional images only provide color information and do not provide depth information. The limited feedback of single color information makes it impossible to completely and accurately detect overflow defects.

[0005] 2. The processing of two-dimensional images requires complex algorithms and high precision in contour extraction. However, the color of steel coils is very uniform, and the metallic color is easily affected by external light, making it impossible to detect some small or irregular overflows.

[0006] Therefore, there is an urgent need for a method that can intelligently detect one-sided overflow of steel coils. Summary of the Invention

[0007] To address the technical problems identified in the prior art, this invention provides a method for detecting unilateral overflow defects in steel coils. The method includes: Step S1, obtaining a point cloud model of the steel coil; Step S2, obtaining a planar region in the point cloud model of the steel coil using a 3D cloud extraction algorithm; Step S3, segmenting and removing a first point cloud set and a second point cloud set from the inner and outer cylindrical surfaces of the steel coil on the point cloud model; Step S4, calculating the overflow distance value from the remaining point cloud to the planar region; Step S5, applying a first threshold judgment to the overflow distance value to extract a target point cloud from the first point cloud set; and Step S6, performing principal component analysis on the target point cloud to determine whether a unilateral overflow defect exists in the steel coil.

[0008] In an optional scheme of this application, step S3 includes: step S31, extracting the center point coordinates of the steel coil point cloud model; step S32, based on the inner diameter information and outer diameter information of the steel coil and the center point coordinates, segmenting and removing the first point cloud on the outer cylindrical surface and the second point cloud on the inner cylindrical surface of the steel coil.

[0009] In an optional scheme of this application, S1 includes: S11, scanning the steel coil with a laser radar along a preset route to obtain raw point cloud data; S12, extracting point cloud features from the raw point cloud data based on the features of the steel coil; S13, using a clustering segmentation algorithm on the steel coil data to obtain a steel coil point cloud model.

[0010] In an optional embodiment of this application, step S4 includes: step S41, obtaining the first vector equation of the planar region; step S42, calculating the projection coordinates of each point of the remaining point cloud to the first vector equation; and step S43, calculating the overflow distance value of each point to the projection coordinates.

[0011] In an optional embodiment of this application, step S5 includes: step S51, determining the difference between the absolute value of the overflow distance value and a preset first threshold to obtain a difference value; step S52, extracting point clouds with a difference value greater than zero as target point clouds.

[0012] In an optional embodiment of this application, step S5 includes: step S61, projecting the target point cloud onto the direction corresponding to the axial direction of the steel coil to obtain a dimensionality-reduced point cloud set; step S62, reconstructing the point cloud set to form a three-dimensional reconstructed point cloud.

[0013] Construct the image; Step S63: Calculate the similarity between the 3D reconstructed image and the training image; Step S64: Determine the similarity based on the similarity threshold.

[0014] Determine whether the steel coil has a one-sided overflow defect.

[0015] In an optional embodiment of this application, the detection method further includes: step S65, in the case that the steel coil has a one-sided overflow defect, acquiring an image of the current steel coil through an image acquisition unit and uploading it to a host computer.

[0016] In an optional scheme of this application, step S5 includes: step S61', projecting the target point cloud onto the direction of the corresponding steel coil axis to form a reduced-dimensional point cloud set; step S62', performing a second threshold judgment on the side length connected by the reduced-dimensional point cloud set to determine whether the steel coil has a single-sided overflow defect.

[0017] A first aspect of this application provides a detection system for unilateral overflow defects in steel coils, comprising: a point cloud acquisition module for receiving raw point cloud data and obtaining a point cloud model of the steel coil; a first extraction module for obtaining planar regions in the point cloud model of the steel coil using a 3D cloud extraction algorithm; a segmentation module for segmenting a first point cloud set and a second point cloud set on the inner cylindrical surface and outer cylindrical surface of the steel coil on the point cloud model and discarding them; a calculation module for calculating the overflow distance value from the remaining point cloud to the planar region; an extraction module for performing a first threshold judgment on the overflow distance value to extract a target point cloud from the remaining point cloud; and a judgment module for performing principal component analysis on the target point cloud to determine whether the steel coil has a unilateral overflow defect.

[0018] Finally, this application also provides a computer-readable storage medium storing one or more programs, one or more of which can be executed by one or more processors to implement the detection method described above. Beneficial effects

[0019] This invention provides a method for detecting unilateral overflow defects in steel coils. By acquiring a point cloud model of the steel coil, the overflow distance from the point cloud to its end face is calculated to obtain a target point cloud. PCA analysis is then performed on the target point cloud to determine if a unilateral overflow defect exists in the steel coil. By combining the features of the steel coil structure and the overflow defect to design an algorithm, this detection method has the following technical advantages over existing image recognition methods: 3D information of the unilateral overflow portion can be acquired in real time using the steel coil point cloud model, and the overflow distance value of the point cloud can be calculated, allowing for accurate acquisition of the overflow distance of the current unilateral overflow portion of the steel coil, resulting in more reliable detection results.

[0020] Other features and advantages of the embodiments of the present invention will be described in the following detailed description section. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a schematic flowchart of the method for detecting unilateral overflow defects in steel coils provided in an embodiment of the present invention.

[0023] Figure 2 An exemplary demonstration of a steel coil point cloud model provided by an embodiment of the present invention is provided;

[0024] Figure 3 This is a schematic diagram of the specific process of step S1 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention.

[0025] Figure 4 This is a schematic diagram of the specific process of step S3 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention.

[0026] Figure 5 This is a schematic diagram of the specific process of step S4 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention;

[0027] Figure 6 The specific process of step S5 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention.

[0028] Schematic diagram;

[0029] Figure 7 This is a schematic diagram of the specific process of step S6 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention.

[0030] Figure 8 An exemplary demonstration is provided by the dimensionality reduction point cloud set in the embodiments of the present invention;

[0031] Figure 9 This is a schematic diagram of the specific process of step S6 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention.

[0032] Figure 10 This is a schematic diagram of the module of the steel coil single-sided overflow defect detection system provided in an embodiment of the present invention.

[0033] Attached Figure

[0034] 100. Detection system; 101. Point cloud acquisition module;

[0035] 101. First extraction module; 103. Segmentation module;

[0036] 104. Calculation module; 105. Second extraction module;

[0037] 106. Judgment Module. Detailed Implementation

[0038] To make the above and other features and advantages of the present invention clearer, the invention will be further described below with reference to the accompanying drawings. It should be understood that the specific embodiments given herein are for the purpose of explanation to those skilled in the art and are exemplary only, not restrictive.

[0039] As mentioned above, existing detection methods rely on manual methods or image recognition. However, due to the lack of depth information in image recognition and the influence of the steel coil's material and color, it is impossible to accurately detect overflow defects in the steel coil. Therefore, this invention provides a method for detecting unilateral overflow defects in steel coils, aiming to solve the above-mentioned technical problems.

[0040] Please see Figure 1 , Figure 1 This is a schematic flowchart of the method for detecting unilateral overflow defects in steel coils provided in an embodiment of the present invention.

[0041] The present invention provides a method for detecting single-sided overflow defects in steel coils, comprising the following steps:

[0042] Step S1: Obtain the point cloud model of the steel coil;

[0043] Step S2: Obtain the planar region in the point cloud model of the steel coil using a 3D cloud extraction algorithm;

[0044] Step S3: On the point cloud model of the steel coil, segment out the first point cloud and the second point cloud of the inner cylindrical surface and the outer cylindrical surface of the steel coil and remove them.

[0045] Step S4: Calculate the overflow distance from the remaining point cloud to the planar region;

[0046] Step S5: Perform a first threshold judgment on the overflow distance value to extract the target point cloud from the remaining point cloud;

[0047] Step S6: Perform principal component analysis on the target point cloud to determine whether the steel coil has a one-sided overflow defect.

[0048] Specifically, step S1 describes obtaining a steel coil point cloud model from point cloud data. In this embodiment of the invention, a point cloud is a set of points. Compared to an image point cloud, it includes depth parameters. In other words, a 3D point cloud directly provides 3D spatial data. In contrast, image recognition in the prior art requires inferring 3D data through perspective geometry, and the data derived from perspective geometry may not meet the requirements and may contain some deviations. Therefore, using a steel coil point cloud model can significantly improve the accuracy of detection.

[0049] It can be understood that the point cloud in the steel coil point cloud model is actually a point cloud dataset A in a preset coordinate system, such as... Figure 2 As shown.

[0050] Please see Figure 3 , Figure 3 Step S1 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention

[0051] A detailed flowchart;

[0052] In a specific scheme, S1 includes:

[0053] S11. Scan the steel coil using a lidar along a preset route to obtain raw point cloud data;

[0054] S12. Extract point cloud features from the original point cloud data based on the characteristics of the steel coil;

[0055] S13. A clustering segmentation algorithm is used to obtain a point cloud model of the steel coil data.

[0056] It is understandable that the steel coil point cloud model in step S1 needs to extract targets from the collected raw point cloud data. The two key steps of target extraction are: feature extraction and selection, and classification. The steel coil is scanned sequentially by the LiDAR along the scanning route to obtain the overall raw point cloud data. Since the steel coil to be identified is usually in an environment where various point clouds are mixed together, when identifying targets in the steel coil point cloud model, there needs to be an index or value to maximize the difference between the steel coil point cloud model and other point clouds, i.e., the target features mentioned above.

[0057] In this application, cylindrical fitting is preferably used for segmentation. The SACMODEL_CYLINDER model (a cylinder defined in the point cloud library) provided by PCL (Point Cloud Library) is used for cylindrical segmentation via a stochastic parameter estimation method, thereby extracting the steel coil point cloud model. For example, the original point cloud data is loaded, and the normal is iteratively calculated using the stochastic parameter estimation method to perform cylindrical segmentation.

[0058] In another feasible example, each point cloud may include RGB information in addition to its three-dimensional coordinates (x, y, z). This information can be extracted based on the point cloud color. The original point cloud data is first preprocessed by filtering and downsampling, and then clustered and segmented by the color of the steel coil to obtain the steel coil point cloud model.

[0059] It is understandable that the method of target extraction is not restricted. Since the steel coil structure is a simple column, other methods such as edge extraction can be used to mainly remove background noise from the original point cloud data to obtain the steel coil point cloud model.

[0060] In step S2, the planar region in the steel coil point cloud model is obtained by extracting features from the point cloud, which corresponds to the end face in the steel coil structure.

[0061] The plane can be extracted using random parameter estimation methods or region growing methods, and segmentation can be achieved through plane fitting.

[0062] In a specific case, the depth algorithm in 3D cloud extraction can also be used to extract specific parts of a planar region. For example, input all point clouds in the steel coil point cloud model and configure the relevant parameters of the point clouds. Arbitrarily set several outliers, such as (*cloud)[0].x=x1;(*cloud)[1].z=x2;(*cloud)[2].z=x3; By calculating the distance from all point clouds in the steel coil point cloud model to the outliers, the distances are divided by setting a threshold. Since the distance from point clouds located on the same plane to the fixed point is the same, the planar region can be filtered and extracted.

[0063] The function implemented in step S3 is to segment the first point cloud set a1 on the outer cylindrical surface of the steel coil and the second point cloud set a2 on the inner cylindrical surface of the steel coil on the point cloud model to reduce the amount of calculation. The overflow distance is obtained by calculating the distance between the remaining point cloud and the planar region. The overflow distance value is judged by the first threshold to extract the target point cloud, that is, the part that exceeds the planar region, from the first point cloud set.

[0064] Understandably, according to the definition of "overflow" in steel coils, it specifically occurs during the coiling process due to factors such as wedge-shaped gaps in the pinch rollers, poor alignment during the coiler feeding process, camber in the strip, and excessive gaps in the auxiliary coiling rollers. These factors cause some layers of the steel coil to protrude relative to the planar area of ​​the end face, resulting in unilateral overflow. A typical point cloud model of a steel coil consists of the end face, an outer cylindrical surface, and an inner cylindrical surface. By removing the outer and inner cylindrical surfaces, the point cloud of the end face can be retained. This significantly reduces computational effort and improves response speed.

[0065] Therefore, in this application, it is necessary to segment the first point set a1 and the second point set a2 corresponding to the inner and outer cylindrical surfaces and calculate their distances to the planar region. This corresponds to the protruding parts of the outer and inner rings of the steel coil.

[0066] Please see Figure 4 , Figure 4 Step S3 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention

[0067] A detailed flowchart;

[0068] In one specific embodiment, step S3 includes:

[0069] Step S31: Extract the coordinates of the center point of the steel coil point cloud model;

[0070] Step S32: Based on the inner diameter information, outer diameter information, and center point coordinates of the steel coil, segment the first point cluster on the outer cylindrical surface and the second point cluster on the inner cylindrical surface of the steel coil and discard them.

[0071] Specifically, step S31 can use cylindrical surface fitting of three-dimensional point cloud. First, input the point cloud dataset A of the steel coil point cloud model, obtain the center point coordinates, and then extract the point cloud of the inner cylindrical surface and the outer cylindrical surface through the segmentation algorithm. That is, extract the segmented point set located on the inner cylindrical surface and the outer cylindrical surface from the point cloud dataset A and remove them.

[0072] For example, the center point coordinates of the steel coil point cloud model are fitted using a random parameter estimation method. After obtaining the center point coordinates, the point cloud within the range of the outer and inner diameters is filtered using the center point coordinates and known outer and inner diameter information to obtain the remaining point cloud of the end face.

[0073] Please see Figure 5 , Figure 5 This is a schematic flowchart of step S4 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention; step S4 includes:

[0074] Step S41: Obtain the first vector equation of the planar region;

[0075] Step S42: Calculate the coordinates of the projection points of each point in the remaining point cloud onto the first vector equation;

[0076] Step S43: Calculate the overflow distance value from each point to the coordinates of the projection point.

[0077] It is understandable that by arbitrarily selecting three point clouds in the planar region to obtain the vector equation of the plane formed, and iterating multiple times, the first vector equation of the plane is determined. If the first vector equation is not parallel to the axial direction of the steel coil, it proves that the steel coil is misaligned. Steps S42 to S43 can also directly calculate the overflow distance value from each point to the coordinates of the projection point using the distance calculation module encapsulated in the software. If the steel coil is not misaligned, the overflow distance value is the straight-line distance from the point cloud to the plane.

[0078] Please see Figure 6 , Figure 6 This is a schematic diagram of the specific flow of step S5 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention; step S5 includes:

[0079] Step S51: Determine the difference between the absolute value of the overflow distance and the preset first threshold to obtain the difference value;

[0080] Step S52: Extract point clouds with differences greater than zero as target point clouds.

[0081] Optionally, the first threshold is set to 10cm. Since the calculated overflow distance value has positive and negative parts, corresponding to the protrusions and depressions on the end face respectively, the absolute value is aligned and the difference is made with 10cm. The target point cloud with the difference greater than 10cm is extracted. This target point cloud reflects the protrusions or depressions on the end face of the steel coil.

[0082] Please see Figure 7 , Figure 7 This is a schematic diagram of the specific flow of step S6 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention; step S6 includes:

[0083] Step S61: Project the target point cloud onto the direction corresponding to the axial direction of the steel coil to obtain the dimensionality-reduced point cloud set;

[0084] Step S62: Reconstruct the reduced point cloud to form a three-dimensional reconstructed image;

[0085] Step S63: Calculate the similarity between the 3D reconstructed image and the training image;

[0086] Step S64: Determine whether the steel coil has a one-sided overflow defect based on the similarity threshold;

[0087] Step S65: In the case of a single-sided overflow defect in the steel coil, the image of the current steel coil is acquired by the image acquisition unit and uploaded to the host computer.

[0088] Steps S61 to S64 above describe a method based on the fusion of image information and depth information. Firstly...

[0089] By reducing the dimensionality of the target point cloud, we obtain, as follows: Figure 8 The reduced point cloud is shown, and then the steel coil overflow is reconstructed based on the reduced point cloud.

[0090] The 3D reconstructed image of the defect scene is compared with the training images of previous steel coil overflow defects, and the similarity is ranked to determine whether the steel coil has a unilateral overflow defect. Step S65 involves reading the current defect image through the image acquisition unit and uploading it to the host computer for storage when an overflow defect occurs in the steel coil, so as to facilitate remote visualization and subsequent traceability analysis by the user.

[0091] Please see Figure 9 , Figure 9 This is another specific flowchart of step S6 in the method for detecting single-sided overflow defects in steel coils provided in this embodiment of the invention;

[0092] In another embodiment of the present invention, step S6 includes:

[0093] Step S61': Project the target point cloud onto the direction corresponding to the axial direction of the steel coil to form a reduced point cloud set;

[0094] Step S62': Perform a second threshold judgment on the side length connected by the reduced point cloud to determine whether the steel coil has a one-sided overflow defect.

[0095] It can be understood that the reduced point cloud reflects the projection of the overflow part of the steel coil onto the axial direction of the steel coil. By connecting and integrating the reduced point cloud to form a continuous line and calculating its side length, when the side length exceeds the preset side length threshold, it is determined that the steel coil has a single-sided overflow defect and needs to be reworked.

[0096] In summary, this invention provides a method for detecting unilateral overflow defects in steel coils. By acquiring a point cloud model of the steel coil, the overflow distance from the point cloud to its end face is calculated to obtain a target point cloud. PCA analysis is then performed on the target point cloud to determine whether a unilateral overflow defect exists in the steel coil. An algorithm is designed by combining the characteristics of the steel coil structure and the overflow defect. Compared to existing image recognition methods, this detection method has the following technical advantages: 3D information of the unilateral overflow portion can be acquired in real time using the steel coil point cloud model; the overflow distance value of the point cloud can be calculated, allowing for accurate determination of the overflow distance of the current unilateral overflow portion of the steel coil, resulting in more reliable detection results.

[0097] Please see Figure 10 , Figure 10 This is a schematic diagram of the modules of the steel coil single-sided overflow defect detection system provided in an embodiment of the present invention;

[0098] A second aspect of this application also provides a detection system 100 for unilateral overflow defects in steel coils, the detection system 100 comprising:

[0099] Point cloud acquisition module 101 is used to receive raw point cloud data and obtain a steel coil point cloud model;

[0100] The first extraction module 102 is used to obtain the planar region in the point cloud model of the steel coil through a 3D cloud extraction algorithm;

[0101] The segmentation module 103 is used to segment the inner cylindrical surface and the outer cylindrical surface of the steel coil into a first point cloud and a second point cloud on the point cloud model of the steel coil.

[0102] Calculation module 104 is used to calculate the overflow distance values ​​from the first point cloud and the second point cloud to the planar region;

[0103] The second extraction module 105 is used to perform a first threshold judgment on the overflow distance value in order to extract the target point cloud;

[0104] The judgment module 106 is used to perform principal component analysis on the target point cloud to determine whether the steel coil has a one-sided overflow defect.

[0105] It is understandable that the above system detection 100 implements its own functions by encapsulating each sub-module. When applied to defect detection of steel coils, it can obtain more accurate 3D information and determine whether the steel coil needs to be reworked by judging the size of the overflow distance.

[0106] This invention also provides an automatic detection device (not shown), which includes:

[0107] The point cloud acquisition device, optionally a lidar, is used to acquire raw point cloud data of the steel coil;

[0108] The host computer includes the above-mentioned detection system 100 for single-sided overflow defects in steel coils.

[0109] Finally, this embodiment of the invention also provides a computer-readable storage medium storing...

[0110] One or more programs, which can be executed by one or more processors, to implement the detection method described above.

[0111] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0112] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0113] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0114] Furthermore, those skilled in the art should understand that if all or part of the sub-modules involved in the automatic feeding equipment provided in the embodiments of the present invention are combined or replaced by means of merging, simple changes, mutual transformation, etc., such as moving the position of each component; or setting the product they constitute as a whole; or having a detachable design; any combined components can form a device / apparatus / system with a specific function, and using such a device / apparatus / system to replace the corresponding components of the present invention also falls within the protection scope of the present invention.

[0115] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0116] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for detecting unilateral overflow defects in steel coils, characterized in that, include: Step S1: Obtain the point cloud model of the steel coil; Step S2: Obtain the planar region in the point cloud model of the steel coil using a 3D cloud extraction algorithm; wherein, the planar region is the end face of the corresponding steel coil structure, and the extracted plane is segmented by plane fitting; Step S3: On the point cloud model of the steel coil, the first point cloud and the second point cloud are segmented from the inner cylindrical surface and the outer cylindrical surface of the steel coil and then discarded. Step S4: Calculate the overflow distance value from the remaining point cloud to the planar region; Step S5: Perform a first threshold judgment on the overflow distance value to extract the target point cloud from the remaining point cloud, wherein the overflow distance value of the target point cloud is greater than the first threshold. Step S6: Perform principal component analysis on the target point cloud to determine whether the steel coil has a one-sided overflow defect; Step S6 includes: Step S61': Project the target point cloud onto the direction corresponding to the axial direction of the steel coil to form a reduced point cloud set; Step S62': Perform a second threshold judgment on the side length connected by the dimensionality reduction point cloud to determine whether the steel coil has a single-sided overflow defect. If the side length exceeds the second threshold, it is determined that the steel coil has an overflow defect.

2. The detection method according to claim 1, characterized in that, Step S3 includes: Step S31: Extract the coordinates of the center point of the steel coil point cloud model; Step S32: Based on the inner diameter information, outer diameter information, and center point coordinates of the steel coil, segment the first point cluster on the outer cylindrical surface and the second point cluster on the inner cylindrical surface of the steel coil and discard them.

3. The detection method according to claim 1, characterized in that, S1 includes: Step S11: Scan the steel coil with a lidar along a preset route to obtain raw point cloud data; Step S12: Extract point cloud features from the original point cloud data based on the features of the steel coil; Step S13: Use a clustering segmentation algorithm to obtain a point cloud model of the steel coil data.

4. The detection method according to claim 1, characterized in that, Step S4 includes: Step S41: Obtain the first vector equation of the planar region; Step S42: Calculate the coordinates of the projection points of each point in the remaining point cloud onto the first vector equation; Step S43: Calculate the overflow distance value from each point to the coordinates of the projection point.

5. The detection method according to claim 4, characterized in that, Step S5 includes: Step S51: Determine the difference between the absolute value of the overflow distance and the preset first threshold to obtain the difference value; Step S52: Extract point clouds with differences greater than zero as target point clouds.

6. A detection system for unilateral overflow defects in steel coils, characterized in that, include: The point cloud acquisition module is used to receive raw point cloud data and obtain a point cloud model of the steel coil; The first extraction module is used to obtain the planar region in the point cloud model of the steel coil through a 3D cloud extraction algorithm; The segmentation module segments the inner cylindrical surface and the outer cylindrical surface of the steel coil into a first point cloud and a second point cloud, and then discards them. The calculation module is used to calculate the overflow distance value from the remaining point cloud to the planar region; The extraction module is used to perform a first threshold judgment on the overflow distance value in order to extract the target point cloud from the remaining point cloud; The judgment module is used to perform principal component analysis on the target point cloud to determine whether the steel coil has a one-sided overflow defect; Principal component analysis of the target point cloud is performed to determine whether the steel coil has a one-sided overflow defect, including: The input projects the target point cloud onto the direction of the corresponding steel coil axis to form a dimensionality-reduced point cloud set; A second threshold is used to determine whether the steel coil has a single-sided overflow defect by judging the length of the side connected by the reduced point cloud.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, and one or more of the programs can be executed by one or more processors to implement the detection method as described in any one of claims 1-5.

Images