Intelligent cloth cutting method for arc protection clothes production

Abstract

The invention relates to the technical field of image processing, in particular to an intelligent cloth cutting method for arc protection clothes production, and the method comprises the steps: obtaining all point clouds of to-be-cut cloth of arc protection clothes, preliminarily determining the overall deformation quantity of the to-be-cut cloth, and obtaining the overall deformation quantity of the to-be-cut cloth; dynamically setting a local range of each point cloud of the to-be-cut cloth through the overall deformation quantity, determining a local curved surface based on the local range, calculating the local deformation possibility of each point cloud, and executing different cutting strategies in response to a judgment result of whether the to-be-cut cloth is deformed or not according to the local deformation possibility of all the point clouds. Therefore, intelligent cutting of the to-be-cut cloth of the arc protection clothes is realized. According to the method, the defects that traditional image processing is easily influenced by visual angles and illumination and is interfered by complex textures of the cloth are overcome, the accuracy of judging the deformation of the cloth is improved, and the cutting precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an intelligent cutting method for fabrics used in the production of arc protection clothing. Background Art

[0002] In various power operation scenarios, arc protection clothing is a key equipment to ensure the safety of workers, and the cutting process in its production process is crucial. At present, the cutting of arc protection clothing mostly adopts traditional processes. Workers set the tool cutting path in professional cutting software according to the design drawings, convert the shape and size of the protective clothing parts into digital instructions, and then place the fabric on the cutting equipment workbench, and the tool cuts according to the preset path, speed and pressure.

[0003] However, the fabric of arc-proof clothing needs to be heat-resistant and arc-resistant, and its texture is relatively thick. It is very easy to deform, stretch or shrink during the cutting process. If the deformation problem is not detected and cutting is carried out directly, the fabric with deformation problems will easily be used for subsequent sewing, which will affect the quality of the arc-proof clothing.

[0004] During the cutting process, the traditional method of detecting whether the fabric has deformation problems mainly relies on image processing algorithms. The image of the fabric before and after cutting is obtained through image acquisition equipment, and the edge detection algorithm is used to identify the edge contour of the fabric, which is compared with the preset standard contour to determine whether deformation occurs.

[0005] However, the traditional image processing methods are easily affected by factors such as viewing angle and lighting, which affect the accuracy of deformation detection. For example, when detecting cloth wrinkles, the shadows of wrinkles under different lighting will interfere with the judgment of the depth and shape of wrinkles. In addition, since the fabric fiber structure of arc protection clothing is complex and the surface texture is rich and irregular, these complex textures are prone to interference when identifying edges and contours using traditional image processing methods. Textures are easily misjudged as deformation edges, resulting in inaccurate judgment of fabric deformation and affecting cutting accuracy. Summary of the invention

[0006] In order to solve the problem that in the cutting process of arc protection clothing fabric, the traditional image processing method is easily affected by the environment, and the fiber structure of the arc protection clothing fabric is complex, resulting in inaccurate judgment of the fabric deformation, which in turn affects the cutting accuracy. The present invention proposes an intelligent cutting method for arc protection clothing production, comprising: All point clouds of the fabric to be cut of the arc protection suit are obtained by laser scanning, and the overall deformation of the fabric to be cut is determined according to the size difference between the minimum bounding box of all point clouds of the fabric to be cut and the minimum bounding box of all point clouds of the standard fabric; The local range of each point cloud of the cloth to be cut is dynamically set by the overall deformation amount, the local surface is determined based on the local range, and the local deformation possibility of each point cloud is calculated: ; In the formula, For the The local deformation possibility of a point cloud, For the The total number of point clouds within the local range of a point cloud, For the The curvature of the local surface of the point cloud, It is The average curvature in the local range of the point cloud, For the The normal vector of the point cloud, For the In the local area of ​​the point cloud The normal vector of the point cloud, for and The cosine of the angle between For the The point cloud density of the local area of ​​the point cloud, For the The average point cloud density in the local area of ​​the point cloud, For the The roughness of the local surface of a point cloud, is the preset curvature difference threshold, is the absolute value symbol; In response to the determination result of whether the cloth to be cut generates deformation according to the local deformation possibility of all point clouds, different cutting strategies are executed to realize intelligent cutting of the cloth to be cut of the arc protection clothing.

[0007] The above technical solution obtains point cloud data through laser scanning, which can fully and accurately reflect the three-dimensional shape information of the fabric. Compared with traditional two-dimensional image processing, it can avoid interference from factors such as viewing angle and lighting, and more accurately obtain the actual shape of the fabric. The overall deformation is determined by comparing the size difference between the minimum bounding box of the point cloud of the fabric to be cut and the standard fabric, which provides a macro-level quantitative basis for the subsequent analysis of the deformation of the fabric. And further, the local range of each point cloud is dynamically set according to the overall deformation, which can more flexibly adapt to the deformation of different areas of the fabric. And further, based on the local range of each point cloud, its local detail features are extracted from the micro level. These local detail features are based on the three-dimensional geometric features of the fabric, rather than relying on image grayscale and edges like traditional image processing, which can more accurately reflect the real deformation of the fabric, effectively avoid the interference of complex textures of the fabric, thereby improving the accuracy of fabric deformation judgment. And further, the deformation of the fabric is determined according to the possibility of local deformation, and different cutting strategies are implemented to achieve intelligent cutting, which improves the accuracy and efficiency of fabric cutting in the production process of arc protection clothing.

[0008] Further, the method of dynamically setting the local range of each point cloud of the cloth to be cut is: Calculate the size of the local range of each point cloud; , where is the size of the local range of each point cloud, is the size of the minimum bounding box of all point clouds of the cloth to be collected, is an adjustment factor greater than 0, is the magnification factor, is the round-up symbol, is the overall deformation of the cloth to be cut, for The threshold value, is the natural exponential function; Taking each point cloud as the center, the local range of each point cloud is determined according to the size of the local range of each point cloud.

[0009] The above technical solution starts from the perspective of material mechanics and geometric deformation. Taking into account that the arc protection clothing fabric is a flexible material, the overall deformation is closely related to the local deformation. The local range size of each point cloud is dynamically set by the size of the overall deformation, which can not only flexibly control the influence of the overall deformation on the local range, but also ensure the rationality of the local range setting.

[0010] Further, the result of determining whether the cloth to be cut is deformed according to the local deformation possibility of all point clouds includes: determining whether there is a suspected deformed point cloud in all point clouds in response to the comparison result of the local deformation possibility of each point cloud and a preset local deformation possibility threshold; if there is no suspected deformed point cloud, determining that the cloth to be cut is not deformed; if there is a suspected deformed point cloud, clustering all suspected deformed point clouds to obtain multiple clusters, and taking the area where each cluster is located as a suspected deformation area; If the area ratio of all suspected deformation areas in the cloth to be cut is greater than the preset area ratio threshold, it is determined that the cloth to be cut is deformed, and all suspected deformation areas are determined as deformation areas; if the area ratio of all suspected deformation areas in the cloth to be cut is not greater than the preset area ratio threshold, it is determined that the cloth to be cut is not deformed, and there is no deformation area in the cloth to be cut.

[0011] The above technical solution uses a progressive deformation judgment strategy to first locate potential problem areas to clarify the direction for subsequent analysis, then cluster the suspected deformation point clouds and determine the overall deformation based on the area ratio, taking into account both local details and overall status, and comprehensively and accurately grasping the deformation of the fabric. Finally, it provides a direct basis for formulating cutting strategies, allowing cutting to be targetedly adjusted according to the actual status of the fabric, thereby ensuring cutting quality.

[0012] Furthermore, the average curvature in the local range of the point cloud and the average point cloud density in the local range of the point cloud are determined as follows: Get the The local range of the point cloud The curvature of each local surface of the point cloud is calculated The mean curvature of all local surfaces of the point cloud is taken as The average curvature in the local range of a point cloud; Get the The local range of the point cloud The point cloud density of each local area of ​​the point cloud is calculated The mean of the point cloud density of all local ranges of the point cloud is taken as the The average point cloud density within the local range of a point cloud.

[0013] Furthermore, the method for determining the local surface based on the local range is: The surface fitting technology is used to perform surface fitting on all point clouds within the local range of each point cloud to obtain the local surface of the point cloud.

[0014] Furthermore, the method for determining the overall deformation of the cloth to be cut is: According to the size difference between the minimum bounding box of all point clouds of the cloth to be cut and the minimum bounding box of all point clouds of the standard cloth, the deformation rate of the cloth to be cut in the length direction is calculated. , Deformation rate in width direction and the deformation rate in the height direction ; Calculate the overall deformation of the cloth to be cut:

[0015] In the formula, is the overall deformation of the cloth to be cut, , , They are the deformation weights of the cloth to be cut in the length, width and height directions respectively.

[0016] The above technical solution comprehensively considers the deformation of the cloth in three-dimensional space, and introduces the deformation weights of the cloth in different directions based on the force characteristics analysis of the cloth material, which can more accurately reflect the real deformation state of the cloth and avoid inaccurate judgment of the overall deformation due to ignoring the deformation in a certain direction.

[0017] Furthermore, the method for determining whether there are suspected deformed point clouds in all point clouds is as follows: Preset local deformation possibility threshold; If the local deformation possibility of a point cloud is greater than the local deformation possibility threshold, the point cloud is regarded as a suspected deformed point cloud. If the local deformation possibility of a point cloud is not greater than the local deformation possibility threshold, the point cloud is regarded as a normal point cloud.

[0018] Furthermore, the methods for implementing different cropping strategies are: If the fabric to be cut is deformed, the cutting path is adjusted to avoid the deformed area; if the fabric to be cut is not deformed, the pre-set cutting path is executed.

[0019] Furthermore, the method for determining the roughness of the local surface is: The variance of the Euclidean distances from all point clouds within the local range of each point cloud to the local surface of the point cloud is calculated, and the variance is used as the roughness of the local surface of the point cloud.

[0020] Furthermore, the method for obtaining the deformation weights of the cloth to be cut in the length, width and height directions is as follows: Prepare in advance a fabric sample of the same material as the fabric to be cut, apply different tensions to the fabric sample in the length, width, and height directions, and record the deformation generated under each tension; The relationship curve between the tension and deformation of the fabric sample in each direction is obtained by using the curve fitting technology. Based on the relationship curve, the ratio of the average deformation change rate and the average tension change rate of the fabric sample in each direction is obtained as the stretch sensitivity of the fabric sample in each direction. The sum of the stretch sensitivities in each direction is taken as the total stretch sensitivity. The ratio of the stretch sensitivity of the cloth sample in the length, width and height directions to the total stretch sensitivity is used as the deformation weight of the cloth to be cut in the length, width and height directions, respectively.

[0021] The above technical scheme takes into account that arc-proof clothing fabrics made of different materials have different fiber structures and weaving methods, and their tensile properties in various directions are also different. By conducting tensile tests on fabric samples of the same material, we can have an in-depth understanding of the tensile properties of the fabric to be cut in different directions, and essentially determine the deformation weight based on the physical properties of the fabric, so that the subsequent calculation of the overall deformation amount can better reflect the actual deformation of the fabric.

[0022] The present invention has the following effects: The present invention uses laser scanning to obtain point cloud data, avoiding the disadvantages of traditional image processing that is easily affected by viewing angle, lighting and interference from complex texture of fabric. By means of multi-dimensional point cloud data analysis, a whole-first-part-later method is adopted to accurately analyze the real deformation of the fabric based on the three-dimensional geometric features of the fabric, avoiding interference from complex texture of the fabric, thereby improving the accuracy of fabric deformation judgment, and adjusting the cutting path according to the accurate deformation judgment result, thereby improving cutting accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0025] Reference Figure 1 The present invention provides a method for intelligently cutting fabric for arc protection clothing production, comprising steps S1 to S5: S1: Acquisition and preprocessing of 3D point cloud of fabric to be cut for arc protection clothing.

[0026] A high-precision laser line scanner is used to perform a full-scale scan of the fabric to be cut for arc protection clothing. As the fabric of arc protection clothing is thick, partial occlusion may occur during the scanning process. To ensure the acquisition of complete and accurate point cloud data, the fabric needs to be scanned from multiple angles.

[0027] During the scanning process, due to various factors, the point cloud data may fluctuate abnormally, including outliers or noise data. Therefore, in order to improve the stability and reliability of the data, after the scanning is completed, all point cloud data of the cloth to be cut need to be preprocessed: including filtering and denoising algorithms to denoise it, which can effectively avoid the interference of individual outliers and ensure the stability of the point cloud data, thereby ensuring the accuracy and reliability of the entire detection solution, and providing high-quality data support for subsequent cloth deformation detection and intelligent cutting.

[0028] S2: Perform a preliminary overall deformation analysis on the cloth to be cut to determine the overall deformation amount of the cloth to be cut.

[0029] The minimum bounding box is the smallest cuboid that can completely contain all point cloud data in the three-dimensional point cloud data. It is used to describe the spatial range and shape characteristics of the point cloud data as a whole. Therefore, the overall deformation of the cloth to be cut can be obtained by analyzing the minimum bounding box of all point cloud data of the cloth to be cut.

[0030] Specifically, the steps include: S21: Obtain the minimum bounding box of all point cloud data of the to-be-cut cloth and the standard cloth respectively.

[0031] A cloth of the same size as the cloth to be cut and conforming to the standard (without surface deformation) is obtained in advance as the standard cloth, and all point cloud data of the standard cloth are obtained according to the operation in step S1.

[0032] Get the minimum bounding box of all point cloud data of the cloth to be cut, and record its length, width and height as , , , and obtain the minimum bounding box of all point cloud data of the standard cloth, and its length, width and height are , , .

[0033] S22: determining the overall deformation amount of the cloth to be cut according to the difference between the size of the minimum bounding box of all point cloud data of the cloth to be cut and the size of the minimum bounding box of all point cloud data of the standard cloth.

[0034] Specifically, they include: S221: Calculate the deformation rate of the cloth to be cut in each direction.

[0035] Calculate the deformation rate of the cloth to be cut in the length direction:

[0036] In this formula, is the deformation rate of the fabric to be cut in the length direction, is the absolute value symbol.

[0037] Calculate the deformation rate of the cloth to be cut in the width direction:

[0038] In this formula, is the deformation rate of the cloth to be cut in the width direction, is the absolute value symbol.

[0039] Calculate the deformation rate of the cloth to be cut in the height direction:

[0040] In this formula, is the deformation rate of the cloth to be cut in the height direction, is the absolute value symbol.

[0041] S222: Determine the deformation weights of the cloth to be cut in each direction.

[0042] The deformation weights of the cloth to be cut in each direction are determined according to the stretch sensitivity of the cloth to be cut in each direction. Stretch sensitivity is an important reference when analyzing the possibility of local deformation of cloth. The higher the stretch sensitivity of the cloth to be cut in a certain direction, the easier it is for the cloth to be cut to deform when subjected to force in that direction. In the case of complex forces in a local area, the possible deformation trend of the local area can be more accurately judged based on the stretch sensitivity in each direction, thereby improving the accuracy of local deformation analysis.

[0043] Prepare in advance a fabric sample of the same material as the fabric to be cut, apply different tensions (gradually increasing tensions) to the fabric sample in the length, width, and height directions, and record the deformation of the fabric sample under each tension in each direction.

[0044] Firstly, the curve fitting technology is used to obtain the relationship curve between the tension and deformation of the fabric sample in each direction (hereinafter referred to as the relationship curve in each direction). Since the relationship curves in three directions are included, the horizontal axis of the relationship curve in each direction is the gradually increasing tension, and the vertical axis is the deformation of the fabric sample under each tension.

[0045] Then, on the relationship curve in each direction, a deformation change rate is calculated for every two deformations, and a tension change rate is calculated for every two tensions. According to this logic, all deformation change rates and all tension change rates in the relationship curve in each direction can be obtained. Further, the average deformation change rate and the average tension change rate are obtained, and the ratio of the average deformation change rate to the average tension change rate is used as the stretch sensitivity of the fabric sample in each direction. Since the fabric to be cut and the fabric sample are made of the same material, the stretch sensitivity of the fabric sample in each direction is also the stretch sensitivity of the fabric to be cut in each direction.

[0046] Taking into account the special fiber structure and weaving method of arc protection clothing fabrics, the tensile properties in different directions are different. By conducting tensile tests on fabric samples of the same material in different directions, the tensile sensitivity of the fabric to be cut in length, width and height can be accurately obtained, and then the deformation weight can be determined, which effectively reflects the anisotropy of the fabric and makes up for the overall deformation judgment error that may be caused by not considering the directional differences.

[0047] Finally, the stretch sensitivity of the fabric sample in the length direction, the stretch sensitivity of the fabric sample in the width direction, and the stretch sensitivity of the fabric sample in the height direction are recorded as , and . Calculate the deformation weights of the cloth to be cut in the length, width, and height directions respectively:

[0048] In this formula, It is the deformation weight of the cloth to be cut in the length direction.

[0049]

[0050] In this formula, It is the deformation weight of the cloth to be cut in the width direction.

[0051]

[0052] In this formula, It is the deformation weight of the cloth to be cut in the height direction.

[0053] Here, the stretch sensitivity of the cloth to be cut in each direction is divided by the total stretch sensitivity to obtain the deformation weight in each direction, which can reflect the relative proportion of the stretch sensitivity in each direction in the whole. The contribution of cloth to the overall deformation in different directions is different. This calculation method can reasonably reflect the relative importance of the stretch sensitivity in each direction in the overall deformation. For example, if the stretch sensitivity in the length direction accounts for a larger proportion of the total stretch sensitivity, it means that the length direction has a greater impact on the overall deformation. When calculating the overall deformation, it is given a higher weight to make the result more in line with the actual situation.

[0054] S223: The overall deformation amount of the cloth to be cut is obtained by comprehensively analyzing the deformation rates and deformation weights of the cloth to be cut in various directions.

[0055] Specifically, the overall deformation of the cloth to be cut is determined based on the following formula:

[0056] In this formula, is the overall deformation of the cloth to be cut, , , is the deformation weight of the cloth to be cut in the length, width and height directions. In the range, the value of deformation weight is Within the interval, , then by performing a weighted average operation on the deformation rate of the cut cloth in each direction, it can be made The value of Within the range.

[0057] Since the fabric to be cut for arc protection clothing may be deformed in length, width and height during the cutting process, it is necessary to comprehensively consider the deformation rate of the fabric to be cut in each direction to avoid focusing on deformation in a single direction while ignoring the influence of other directions. This formula comprehensively presents the deformation of the fabric in three-dimensional space, providing a basis for accurately grasping the overall morphological changes of the fabric.

[0058] S3: Perform local deformation analysis on the cloth to be cut to obtain the local deformation possibility of each point cloud.

[0059] Arc-proof clothing fabrics have the characteristics of diverse fiber composition, special weaving method, and complex surface texture to meet special performance requirements such as arc protection and high temperature resistance. This complex structure makes it easy for the overall deformation of the fabric to be cut to be within an acceptable range during the cutting process, but local areas of the fabric to be cut are still deformed. For example, some parts of the fabric to be cut may have slight misalignment, wrinkles or stretching between fibers due to uneven local force. If these local deformations are not detected, quality problems may occur in the subsequent production of arc-proof clothing.

[0060] The overall deformation and local deformation of the fabric to be cut for arc protection clothing are closely related: From the perspective of material mechanics, when the fabric to be cut for arc protection clothing has a large overall deformation, it is largely because some local areas of the fabric have undergone significant deformation. Since the fabric of arc protection clothing is composed of multiple fibers through a special weaving method, its internal structure is not completely uniform. When subjected to external forces, the fibers in local areas are more likely to be dislocated, slipped or broken. These local changes accumulate and lead to an increase in the overall deformation. For example, at some weaving nodes of the fabric, the fibers are interwoven to different degrees of tightness. When subjected to force, the fibers near these nodes may first undergo small displacements. As the external force continues to act, these local small displacements gradually accumulate and eventually manifest as a large overall deformation. The significant deformation of these local areas will become a potential hidden danger that affects the protective performance of arc protection clothing. The protection principle of arc protection clothing depends on the integrity of its fabric structure. Abnormal changes in fibers in local areas will destroy the original tight structure of the fabric, making it easier for arcs and high temperatures to penetrate the fabric, thereby reducing the arc protection clothing's ability to block arcs and high temperatures.

[0061] From the perspective of geometric deformation, when the overall deformation of the fabric to be cut for arc protection clothing is large, it usually means that it is subjected to a large external force. Due to the diverse fiber composition and special weaving method of arc protection clothing fabrics, it has obvious unevenness when subjected to force. In the process of being subjected to external forces such as stretching, bending or torsion, the stress and strain borne by different parts of the fabric are not the same. For example, during the stretching process, some fiber bundles may be subjected to greater tension, while the fiber bundles in adjacent areas are relatively less stressed, which leads to large differences in tensile and compressive deformations in different parts. In this case, local areas are more likely to have concentrated and obvious deformations. Such concentrated and obvious deformations in local areas will have an adverse effect on the durability and protective effect of arc protection clothing.

[0062] It can be seen that the overall deformation of the fabric to be cut can reflect the possibility of local deformation to a certain extent. There is a close relationship between the overall deformation and the local deformation. In the process of deformation analysis, full consideration of this relationship is of great significance for accurately detecting fabric deformation, optimizing the cutting process, and ensuring the quality and performance of arc protection clothing.

[0063] Therefore, this step performs a detailed local deformation analysis on each point cloud to determine whether there is a small local deformation in the cloth to be cut, so as to improve the accuracy of deformation analysis and subsequent cutting.

[0064] Specifically, the steps include: S31: Dynamically set the local range of each point cloud according to the size of the overall deformation of the cloth to be cut.

[0065] The local range of the point cloud provides a reasonable spatial boundary for extracting local detail features. When analyzing the local detail features of each point cloud, a reasonable local range can reflect the actual deformation situation near the point cloud. If the local range is too large, interference from distant point clouds will cause the calculated features to deviate from reality; if the local range is too small, the data information contained is insufficient and the feature representativeness is poor. Therefore, it is necessary to set an appropriate local range to provide a reliable reference range for accurate local deformation analysis.

[0066] Therefore, when performing deformation analysis on the cloth to be cut, the threshold of the overall deformation of the cloth to be cut is first set. for (Experience points).

[0067] when When , it indicates that the overall deformation of the cloth to be cut is in a relatively small state. In this case, the local deformation of the cloth to be cut is relatively uniform and small, and the deformation complexity of the local area is low. At this time, it is a more reasonable strategy to set a larger local range for each point cloud. The reason is that under the condition of small overall deformation, the point cloud data changes relatively smoothly within a larger local range for each point cloud, and the analysis results will not be disturbed by local abnormal changes. By setting a larger local range, more point cloud data can be covered in one analysis process. This means that there is no need to analyze each small neighborhood of each point cloud one by one, thereby reducing the total number of analyses and improving computational efficiency. For example, there are local areas with uniform deformation in the cloth to be cut. The deformation characteristics reflected by each point cloud in this local area in a larger neighborhood are similar. Therefore, it is suitable for analysis in a larger local range, thereby saving computing resources and time costs.

[0068] when When , it means that the overall deformation of the cloth to be cut has reached a large degree. In this case, it often means that the cloth to be cut has more complex local deformation. In order to improve the analysis accuracy of local deformation, a smaller local range should be set for each point cloud. Since the overall deformation is large, it means that the deformation difference between different areas inside the cloth to be cut may be very significant, and complex conditions such as stress concentration and fiber dislocation are very likely to occur in local areas. At this time, a smaller local range can focus more accurately on the local features around each point cloud, effectively avoid interference from other larger deformation differences, and then analyze the local deformation more accurately. For example, in concentrated areas of deformation such as wrinkles, stretching or compression of the cloth to be cut, a smaller local range can capture the detailed information of these local deformations more finely.

[0069] Therefore, based on the above, for each point cloud of the cloth to be cut, the size of the local range of each point cloud is determined according to the following formula:

[0070] In the formula, is the size of the local extent of each point cloud (length or width or height), is the size of the minimum bounding box (length, width or height) of all point clouds of the cloth to be collected, and and Both represent the size of the same dimension, that is, when calculating the length of the local range of each point cloud, represents the length of the local range of each point cloud, is the length of the minimum bounding box of all point clouds of the cloth to be collected; when calculating the width of the local range of each point cloud, Indicates the width of the local range of each point cloud, is the width of the minimum bounding box of all point clouds to be collected; when calculating the height of the local range of each point cloud, Represents the height of the local range of each point cloud, It is the height of the minimum bounding box of all point clouds of the cloth to be collected. It is an adjustment coefficient greater than 0, which is used to control the influence of the overall deformation on the local size of each point cloud. (Experience points), is the magnification factor, which is used to ensure that the local range of each point cloud is much smaller than the size of the minimum bounding box of all point clouds to be cut. The value range of is, here (Experience points), is the round-up symbol, is the overall deformation of the cloth to be cut, for The threshold value, is a natural exponential function.

[0071] In this formula, It reflects the difference between the overall deformation of the cloth to be cut and the threshold value, and dynamically adjusts the size of the local range of each point cloud according to the size of the difference. In the setting process, the size of the minimum bounding box is used as a reference point. Under the condition that the size of the local range of each point cloud is much smaller than the size of the minimum bounding box, when the overall deformation is large, a smaller local range is set for each point cloud; when the overall deformation is small, a larger local range is set for each point cloud. Specifically: when When , it means that the overall deformation of the cloth to be cut is large, and the cloth to be cut may have complex local deformation. , Will follow increases with the increase of The amplification effect makes The value of will increase rapidly, making The value will change with The increase of Smaller (the size of the local range of each point cloud is much smaller than the size of the smallest bounding box), and smaller than ( ) is calculated This is consistent with the logic of setting a smaller overall deformation when the overall deformation of the cloth to be cut is larger.

[0072] For example, when hour; ; ; ; at this time, , set the size of the local range of each point cloud to be the size of the minimum bounding box of all point clouds of the cloth to be cut Times and round up. When the overall deformation of the fabric to be cut is small Computational In terms of size, it is still relatively small.

[0073] when When , it means that the overall deformation of the cloth to be cut is small, the local deformation of the cloth is relatively uniform and small, and the deformation of the local area is not complicated. ,along with The decrease of The absolute value of increases, due to the natural exponential function When the independent variable is less than 0, the function value is between 0 and 1, and the smaller the independent variable is, the closer the function value is to 0. .so, Will follow decreases and approaches 0, then The amplification effect is relatively weak. The value of will be close to 1 and greater than 1, that is Will follow decreases and increases (approaches 1 and is greater than 1), then The value of The decrease of Also slightly reduced, the calculated Relative to time Calculated , which is still relatively large, which is consistent with the logic of setting a larger local range when the overall deformation of the cloth to be cut is small.

[0074] For example, when hour; ; ; ; at this time, , set the size of the local range of each point cloud to 0.15 times the size of the minimum bounding box of all point clouds of the cloth to be cut and round up. When the overall deformation of the fabric to be cut is large Computational In terms of size, it is still relatively large.

[0075] Through the above scheme, when the overall deformation of the cloth to be cut is small, a relatively large local range is set, which can not only take advantage of the stable change characteristics of point cloud data, cover more point clouds, reduce the number of analyses to improve calculation efficiency, but also ensure that the local range is much smaller than the minimum bounding box size, avoid introducing too much interference from distant point clouds, and ensure the accuracy of the analysis. When the overall deformation is large, a smaller local range is set to accurately focus on the local complex deformation area, effectively avoid interference from large deformation differences outside the neighborhood, improve analysis accuracy, and also meet the requirement that the local range is much smaller than the minimum bounding box size, making the analysis more targeted.

[0076] On the one hand, it avoids analysis errors caused by excessive local range, and on the other hand, it ensures that under different overall deformation conditions, the local characteristics of the fabric can be effectively focused on for analysis, providing a reliable basis for accurately judging fabric deformation and reasonably adjusting the cutting path.

[0077] According to this formula, the size of the local range of each point cloud is calculated, that is, the length, width and height of the local range. Finally, with each point cloud as the center, the local range of each point cloud is obtained according to the size of the local range. The local range finally obtained is a rectangular range centered on each point cloud. If the size (length, width, height) of the local range of a point cloud is 5, 4, and 3 respectively, then With the point cloud as the center, a rectangular range with a length of 5, a width of 4, and a height of 3 is obtained, and the rectangular range is used as the local range of the point cloud.

[0078] S32: Extracting local detail features of each point cloud according to the local range of each point cloud.

[0079] First, for each point cloud, a surface fitting technique is used to fit all point clouds within a local range of the point cloud, and the fitted surface is used as the local surface of the point cloud. Specifically, the least squares method is used to fit all point clouds within a local range of the point cloud, and a radial basis function can also be used to fit all point clouds within a local range of the point cloud.

[0080] The surface of arc protection clothing fabric has a complex geometric shape. The local surface of each point cloud is obtained by using surface fitting technology, which can effectively extract the geometric features of the fabric to be cut in the area near the point cloud. Through surface fitting, the discrete point cloud is converted into a continuous surface model, which intuitively presents the curvature and shape changes of the fabric surface. For example, at the wrinkles of the fabric, the local surface can clearly show the curvature and trend of the wrinkles, providing an intuitive basis for analyzing the deformation of the area.

[0081] Next, the normal vector of each point cloud is obtained and its normal vector is calculated. The normal vector is defined as a vector perpendicular to the tangent plane of the surface at a specific point, which contains the directional characteristics of the surface at that point. The change in the direction of the normal vector can intuitively reflect the change in the normal direction of the surface of the cloth to be cut. Since the cloth to be cut is composed of fibers, the arrangement direction of the fibers inside the cloth to be cut can be further inferred by analyzing the change in the direction of the normal vector. When the direction of the normal vectors of all point clouds within the local range of a point cloud changes greatly, it is very likely that the fiber arrangement in the area corresponding to the local range of the cloth to be cut is relatively disordered, because the irregular arrangement of the fibers will cause the difference in the direction of the tangent plane of the surface at different points to increase, thereby causing the direction of the normal vector to show a large change. In addition, if the direction of the normal vectors of all point clouds within the local range of a point cloud changes significantly, in addition to the possibility of disordered fiber arrangement, it may also indicate that the cloth to be cut is subjected to external forces in the area corresponding to the local range, because the external force will change the shape and internal structure of the cloth, thereby affecting the tangent plane of the surface, which is ultimately reflected in the change of the normal vector. For example, when the cloth to be cut is twisted or stretched unevenly, the direction of the normal vector in the corresponding area will change significantly.

[0082] Then, for each point cloud, the curvature of the local surface is calculated based on the geometric properties of its local surface. Specifically, based on the geometric properties of the surface, the curvature is approximately calculated by analyzing the normal vector and bending characteristics of the surface. The normal vectors at each point cloud corresponding to the local surface are obtained, and the principal curvature direction is determined by analyzing the degree of curvature of the surface in the direction of different normal vectors. Finally, the value of the principal curvature is calculated based on the degree of curvature of the surface in the direction of the principal curvature, which is used as the curvature of the local surface.

[0083] Curvature, as a quantity used to describe the degree of curvature of a curve or surface, has a clear physical meaning in different dimensions. In the context of three-dimensional surfaces, curvature reflects the degree of curvature of the surface at a specific point. By calculating the curvature of the local surface of each point cloud, we can gain an in-depth understanding of the deformation of the cloth to be cut near that point. Generally speaking, an area with a large curvature often means that the cloth to be cut has a more obvious deformation in that area, for example, wrinkles or more severe stretching may occur. Wrinkles will cause the surface of the cloth to be cut to form a complex curved shape, resulting in an increase in the curvature of the local surface; similarly, severe stretching will change the original shape of the cloth, causing a significant change in the degree of curvature of the local area, which is reflected in the increase of the curvature value. This method of judging the degree of deformation of the cloth by curvature provides an important quantitative basis for accurately identifying the local deformation of the cloth.

[0084] Next, the point cloud density in the local range of each point cloud is obtained. For each point cloud, the average distance between the point cloud and all point clouds in its local range is calculated to reflect the point cloud density around the point. The smaller the average distance, the higher the point cloud density. In order to obtain a more intuitive point cloud density representation, the inverse of the average distance is taken as the point cloud density.

[0085] Point cloud density is an important indicator that reflects the density of point cloud distribution in space. If the point cloud density in a local range of a point cloud is large, it means that the point cloud distribution in this local range is relatively dense. From the actual situation of the cloth, the dense distribution of point clouds often implies that the cloth in this area may have been squeezed. Because in the process of squeezing, the local space of the cloth is compressed, and the point clouds that were originally distributed more evenly will gather together, resulting in an increase in the point cloud density. On the contrary, when the point cloud density is small, it means that the point cloud in this area is relatively sparse. This sparse distribution usually indicates that the cloth may have undergone a stretching process in this area, because stretching will expand the local range of the cloth, and the point cloud will be distributed in a larger space, thereby reducing the point cloud density. Therefore, a local range where the point cloud density increases or decreases sharply is more likely to produce local deformation than a local range where the point cloud density changes slowly.

[0086] Finally, the roughness of the local surface of each point cloud is also an important indicator reflecting the deformation of the cloth to be cut. Even if the curvature of the local surfaces of different point clouds is the same, the difference in the roughness of the surface can reveal the difference in the microscopic state of the cloth surface. When the local surface is relatively rough, it often means that there are more microscopic unevenness on the surface of the cloth, and this microscopic unevenness is a specific manifestation of local deformation. It may be due to the action of external force, resulting in local displacement or arrangement changes of the cloth fibers, and then forming microscopic unevenness on the surface. By analyzing the roughness of the local surface of the point cloud, we can have a more comprehensive understanding of the local deformation of the cloth surface and provide richer information for accurately judging the overall deformation of the cloth.

[0087] Specifically, for each point cloud, the variance of the Euclidean distance from all point clouds within the local range of the point cloud to the local surface of the point cloud is first calculated, and the variance is used as the roughness of the local surface of the point cloud. The variance of the distance can effectively quantify the discreteness of the distance from the point cloud to the local surface. The unevenness of the surface of the cloth to be cut is manifested as follows: there is a difference in the distance between all point clouds within the local range of each point cloud and the local surface of the point cloud. The more the difference, the greater the discreteness of these distances, that is, the greater the variance of the distance, which means that the microscopic unevenness of the cloth surface is more serious and the roughness is higher; conversely, the smaller the variance, the more concentrated the distance, the smoother the surface, and the lower the roughness.

[0088] S33: Determine the local deformation possibility of each point cloud based on the local detail features of each point cloud.

[0089] When analyzing the local surface of each point cloud, a key point is that even if the curvature of the local surfaces of different point clouds is the same, there may still be significant differences in the roughness of the surfaces. Therefore, when comprehensively analyzing the local deformation possibility of each point cloud, the difference in curvature of the local surfaces of different point clouds can be used to determine whether it is necessary to exclude the influence of the curvature difference on the local deformation possibility. When the curvature difference is very small, it means that the curvature of the local surface of the point cloud is closer to the average level of the surrounding area. At this time, the influence of the curvature difference on the local deformation possibility is relatively small, and other factors (such as point cloud density, normal vector direction, etc.) have a more significant influence on the local deformation possibility, so the dimension of curvature difference can be ignored when calculating the local deformation possibility. On the contrary, when the curvature difference is large, the contribution of the curvature difference to the local deformation possibility needs to be retained. However, regardless of whether the curvature difference is large or small, when calculating the local deformation possibility, the dimension of the roughness of the surface must always be retained, and the synergistic effect of the roughness on other factors is used to highlight the deformation possibility.

[0090] First, analyze the relative difference between the curvature of the local surface of each point cloud and the average curvature in the local range of the point cloud. When the relative difference is large, it means that the curvature of the local surface of the point cloud is significantly different from the average level, indicating that there may be serious wrinkles or stretching in the local range of the point cloud, so the contribution to the local deformation is large; conversely, when the relative difference is small, it means that the curvature of the local surface of the point cloud is close to the average level, indicating that the local range of the point cloud is relatively flat, so the contribution to the local deformation is small.

[0091] Specifically, The calculation formula for the relative difference between the curvature of the local surface of a point cloud and the average curvature within the local range of the point cloud is:

[0092] In this formula, For the The curvature of the local surface of the point cloud, It is The average curvature in the local range of the point cloud, is the absolute value symbol.

[0093] Then, according to The relative difference between the curvature of the local surface of the point cloud and the average curvature within the local range of the point cloud is There are two ways to calculate the local deformation possibility of a point cloud.

[0094] The first one, when , explain The relative difference between the curvature of the local surface of a point cloud and the average curvature within the local range of the point cloud is small. The curvature of the local surface of a point cloud is close to the average curvature of the local range. The relative difference between the curvature of the local surface of a point cloud and the average curvature in the local range of the point cloud has a relatively small impact on the local deformation possibility. When calculating the local deformation possibility, ignore This is because in this case, other factors such as point cloud density, normal vector direction, etc. have a more prominent effect on local deformation. For example, in a relatively flat area of ​​the cloth, although the curvature difference is small, the change in point cloud density may reflect potential stretching or squeezing. At this time, paying attention to other factors can more accurately judge the possibility of local deformation.

[0095] Therefore, the The calculation formula for the local deformation possibility of a point cloud is:

[0096] In this formula, in this formula, For the The local deformation possibility of a point cloud, and its numerical value reflects the possibility of deformation in the local area where the point cloud is located. For the The total number of point clouds within the local range of a point cloud, For the The curvature of the local surface of a point cloud intuitively reflects the degree of curvature of the local surface of the point cloud. For example, in the wrinkled area of ​​the cloth, The value of is usually larger, indicating that the bending here is more severe. It is The average curvature in the local range of a point cloud, which represents the overall curvature trend of the local range. For the The normal vector of the point cloud, For the In the local area of ​​the point cloud The normal vector of the point cloud, for and The cosine value of the angle between use The value in the interval is used to represent the point cloud. Because the surface of the cloth to be cut is continuous and smoothly transitioned (even if there are wrinkles, it changes continuously), the adjacent point clouds represent the adjacent positions on the surface of the cloth. Due to the continuity of the cloth, the normal vector directions of adjacent point clouds will not have sudden and extreme turns, so the normal vector angle usually does not exceed 180. For the The point cloud density of the local area of ​​the point cloud, For the The average point cloud density in the local area of ​​the point cloud, For the The roughness of the local surface of a point cloud, is the preset curvature difference threshold, set to 0.3 (empirical value), is the absolute value symbol.

[0097] In this formula, the method for obtaining the mean curvature is: The local range of the point cloud The curvature of each local surface of the point cloud is calculated The mean curvature of all local surfaces of the point cloud is taken as The average curvature within a local range of a point cloud.

[0098] In this formula, the method for obtaining the average point cloud density is: The local range of the point cloud The point cloud density of each local area of ​​the point cloud is calculated The mean of the point cloud density of all local ranges of the point cloud is taken as the The average point cloud density within the local range of a point cloud.

[0099] In this formula, The calculation is The average value of the cosine value of the angle between the normal vector of a point cloud and the normal vectors of all point clouds in its local range. The interval is monotonically decreasing. The larger the average value, the higher the The smaller the angle between the normal vector of a point cloud and the normal vectors of all point clouds in its local range, the more consistent the directions of the normal vectors of all point clouds in the local range are, and the area corresponding to the local range of the cloth to be cut is likely to be relatively flat, and the possibility of local deformation is small; conversely, when the average value is smaller, it means that the direction of the normal vectors of all point clouds in the local range changes more, suggesting that the area corresponding to the local range of the cloth to be cut may have disordered fiber arrangement or be affected by complex external forces, and the possibility of local deformation is greater. Therefore, through Build and negative correlation.

[0100] In this formula, , this part calculates the The relative difference between the point cloud density of the local range of a point cloud and the average point cloud density in the local range. When the value is greater than 1, it means that the point cloud in the local range is relatively dense, indicating that the cloth to be cut may be squeezed at the position corresponding to the local range. When the value is less than 1, it means that the point cloud in the local range is relatively sparse, and the cloth to be cut may be stretched at the position corresponding to the local range. This relative difference reflects the impact of the change in point cloud density on the local deformation.

[0101] The second type, when , explain The curvature of the local surface of a point cloud is relatively different from the average curvature in the local range of the point cloud. The curvature of the local surface of a point cloud is significantly different from the average curvature within the local range of the point cloud. For example, the curvature difference is often large at the folds of the cloth. The relative difference between the curvature of the local surface of a point cloud and the average curvature within the local range of the point cloud has a relatively large impact on the local deformation possibility. When calculating the local deformation possibility, the relative difference between the local surface curvature of a point cloud and the average curvature within the local range of the point cloud has a relatively large impact on the local deformation possibility. This part of the impact.

[0102] Therefore, the The calculation formula for the local deformation possibility of a point cloud is:

[0103] The meaning of each character in this formula is exactly the same as that of the character in the previous formula.

[0104] Through the coordinated analysis of multiple factors, the local deformation characteristics of the cloth under different external force combinations can be accurately captured. For example, when the cloth to be cut is squeezed and twisted at the same time, the point cloud density will increase due to squeezing, the normal vector direction will be disordered due to twisting, the curvature will also change, and the roughness may change due to factors such as surface friction. By comprehensively considering multiple factors related to the local deformation of the cloth, such as curvature differences, normal vector direction differences, point cloud density differences, and surface roughness, the local deformation of the cloth can be comprehensively described from multiple angles. Compared with the method that only considers a single factor, it can more accurately reflect the actual degree of local deformation.

[0105] In addition, in order to facilitate comparison, the maximum and minimum values ​​of the local deformation possibilities of all point clouds are obtained, and the local deformation possibility of each point cloud is normalized using the minimum-maximum normalization method so that the value of the local deformation possibility of each point cloud is within Within the range.

[0106] S4: Determine whether the cloth to be cut will be deformed according to the local deformation possibility of all point clouds.

[0107] First, determine whether there are suspected deformed point clouds in all point clouds: The preset local deformation possibility threshold is 0.6 (empirical value); if the local deformation possibility of a point cloud is greater than 0.6, the higher the possibility that the cloth to be cut will produce local deformation at the position corresponding to the point cloud, the point cloud will be regarded as a suspected deformed point cloud; if the local deformation possibility of a point cloud is less than or equal to 0.6, the lower the possibility that the cloth to be cut will produce local deformation at the position corresponding to the point cloud, the point cloud will be regarded as a normal point cloud.

[0108] Next, among all the point clouds, if there is no suspected deformation point cloud, it is determined that the cloth to be cut has not been deformed; if there is a suspected deformation point cloud, an iterative self-organizing clustering algorithm is used, and the appropriate neighborhood radius is set to 0.05 meters (because the scanning accuracy of the point cloud obtained by laser scanning is high, the point cloud distribution is relatively dense, and a smaller neighborhood radius can accurately identify adjacent data points), and the minimum number of points is set to 5 (because all point cloud data have been pre-processed, the data quality is good, the accuracy of the point cloud is high, and a smaller number of points can represent a valid cluster). All suspected deformation points are divided into different clusters through this clustering algorithm, and the area where each cluster is located is taken as a suspected deformation area.

[0109] Finally, the preset area ratio threshold is (Experience value), if the proportion of all suspected deformation areas in the area of ​​the cloth to be cut is greater than , determine that the fabric to be cut has deformed, and determine all suspected deformation areas as deformation areas; if the proportion of all suspected deformation areas in the area of ​​the fabric to be cut is less than or equal to , indicating that the overall shape of the cloth to be cut is relatively flat, and there is no obvious deformation. It is determined that the cloth to be cut has not been deformed. Therefore, there is no deformation area in the cloth to be cut.

[0110] S5: Execute different cutting strategies according to different judgment results to achieve intelligent cutting of the fabric to be cut for the arc protection clothing.

[0111] If the judgment result shows that the fabric to be cut has been deformed, the system will immediately issue a notification to inform the staff that the current fabric is deformed. The staff needs to readjust the cutting path according to the specific deformation of the fabric to avoid the deformed area. This is because if the deformed area is not avoided during the cutting process, it will affect the quality of the final protective clothing. For example, cutting in a deformed area may cause the physical properties of the fabric to change, reducing its ability to protect against electric arcs and high temperatures, or causing deviations in the pattern of the protective clothing, affecting the comfort and safety of wearing.

[0112] If the judgment result shows that the fabric to be cut has not been deformed, the pre-set cutting path is directly executed. The pre-set cutting path is based on the standard fabric and the design requirements of the protective clothing, which can ensure that the shape and size of the cut fabric accurately meet the standards for making arc protection clothing.

[0113] Through the above series of steps, namely, first accurately judging the overall deformation, deeply analyzing the local deformation based on the overall deformation analysis, and finally clustering the deformation position according to the judgment results of the local deformation, it provides comprehensive, accurate and targeted information for the adjustment of the cutting process of arc protection clothing fabrics.

[0114] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. An intelligent cutting method for fabrics used in arc protection clothing production, characterized in that: include: All point clouds of the fabric to be cut of the arc protection suit are obtained by laser scanning, and the overall deformation of the fabric to be cut is determined according to the size difference between the minimum bounding box of all point clouds of the fabric to be cut and the minimum bounding box of all point clouds of the standard fabric; The local range of each point cloud of the cloth to be cut is dynamically set by the overall deformation amount, the local surface is determined based on the local range, and the local deformation possibility of each point cloud is calculated: ； In the formula, For the The local deformation possibility of a point cloud, For the The total number of point clouds within the local range of a point cloud, For the The curvature of the local surface of the point cloud, It is The average curvature in the local range of the point cloud, For the The normal vector of the point cloud, For the In the local area of ​​the point cloud The normal vector of the point cloud, for and The cosine of the angle between For the The point cloud density of the local area of ​​the point cloud, For the The average point cloud density in the local area of ​​the point cloud, For the The roughness of the local surface of a point cloud, is the preset curvature difference threshold, is the absolute value symbol; In response to the determination result of whether the cloth to be cut generates deformation according to the local deformation possibility of all point clouds, different cutting strategies are executed to realize intelligent cutting of the cloth to be cut of the arc protection clothing.

2. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 1 is characterized in that: The method of dynamically setting the local range of each point cloud of the cloth to be cut is; Calculate the size of the local range of each point cloud; , where is the size of the local range of each point cloud, is the size of the minimum bounding box of all point clouds of the cloth to be collected, is an adjustment factor greater than 0, is the magnification factor, is the round-up symbol, is the overall deformation of the cloth to be cut, for The threshold value, is the natural exponential function; Taking each point cloud as the center, the local range of each point cloud is determined according to the size of the local range of each point cloud.

3. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 1, characterized in that: The determination result of whether the cloth to be cut is deformed according to the local deformation possibility of all point clouds includes: determining whether there is a suspected deformed point cloud in all point clouds in response to the comparison result of the local deformation possibility of each point cloud and a preset local deformation possibility threshold; if there is no suspected deformed point cloud, determining that the cloth to be cut is not deformed; if there is a suspected deformed point cloud, clustering all suspected deformed point clouds to obtain multiple clusters, and taking the area where each cluster is located as a suspected deformation area; If the area ratio of all suspected deformation areas in the cloth to be cut is greater than the preset area ratio threshold, it is determined that the cloth to be cut is deformed, and all suspected deformation areas are determined as deformation areas; if the area ratio of all suspected deformation areas in the cloth to be cut is not greater than the preset area ratio threshold, it is determined that the cloth to be cut is not deformed, and there is no deformation area in the cloth to be cut.

4. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 1 is characterized in that: The average curvature within the local range of the point cloud and the average point cloud density within the local range of the point cloud are determined as follows: Get the The local range of the point cloud The curvature of each local surface of the point cloud is calculated The mean curvature of all local surfaces of the point cloud is taken as The average curvature in the local range of a point cloud; Get the The local range of the point cloud The point cloud density of each local area of ​​the point cloud is calculated The mean of the point cloud density of all local ranges of the point cloud is taken as the The average point cloud density within the local range of a point cloud.

5. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 1, characterized in that: The method for determining the local surface based on the local range is: The surface fitting technology is used to perform surface fitting on all point clouds within the local range of each point cloud to obtain the local surface of the point cloud.

6. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 2, characterized in that: The method to determine the overall deformation of the cloth to be cut is: According to the size difference between the minimum bounding box of all point clouds of the cloth to be cut and the minimum bounding box of all point clouds of the standard cloth, the deformation rate of the cloth to be cut in the length direction is calculated. , Deformation rate in width direction and the deformation rate in the height direction ; Calculate the overall deformation of the cloth to be cut: ； In the formula, is the overall deformation of the cloth to be cut, , , They are the deformation weights of the cloth to be cut in the length, width and height directions respectively.

7. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 3 is characterized in that: The method for determining whether there are suspected deformed point clouds in all point clouds is: Preset local deformation possibility threshold; If the local deformation possibility of a point cloud is greater than the local deformation possibility threshold, the point cloud is regarded as a suspected deformed point cloud. If the local deformation possibility of a point cloud is not greater than the local deformation possibility threshold, the point cloud is regarded as a normal point cloud.

8. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 3 is characterized in that: The methods to implement different cropping strategies are: If the fabric to be cut is deformed, adjust the cutting path to avoid the deformed area; If the fabric to be cut does not deform, the preset cutting path is executed.

9. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 5, characterized in that: The method for determining the roughness of the local surface is: The variance of the Euclidean distances from all point clouds within the local range of each point cloud to the local surface of the point cloud is calculated, and the variance is used as the roughness of the local surface of the point cloud.

10. The intelligent cutting method for fabrics used in arc protection clothing production according to claim 6, characterized in that: The method for obtaining the deformation weights of the cloth to be cut in the length, width, and height directions is: Prepare in advance a fabric sample of the same material as the fabric to be cut, apply different tensions to the fabric sample in the length, width, and height directions, and record the deformation generated under each tension; The relationship curve between the tension and deformation of the fabric sample in each direction is obtained by using the curve fitting technology. Based on the relationship curve, the ratio of the average deformation change rate and the average tension change rate of the fabric sample in each direction is obtained as the stretch sensitivity of the fabric sample in each direction. The sum of the stretch sensitivities in each direction is taken as the total stretch sensitivity. The ratio of the stretch sensitivity of the cloth sample in the length, width and height directions to the total stretch sensitivity is used as the deformation weight of the cloth to be cut in the length, width and height directions, respectively.

Images