A method, system and device for testing hardness of a work seat in a closed cabin

By using grayscale maps and feature analysis in the seat hardness test in the sealed cabin, the optional weights of the test are constructed, solving the problem of identification of screw holes and connections, and improving the accuracy and safety of the test.

CN118710642BActive Publication Date: 2024-11-12NORTHWESTERN POLYTECHNICAL UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411187707.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-12
Estimated Expiration
2044-08-28

AI Technical Summary

Technical Problem

In the hardness test of seats in sealed cabins, it is difficult for the prior art to accurately identify and avoid screw holes and connections, resulting in deviations in hardness test results and may cause irreversible damage to the seat structure.

Method used

By obtaining the grayscale map of the seat components, extracting edges and clustering, the distribution rule index, morphological difference coefficient, local chaos and characteristic deviation degree are constructed, combined with these characteristics to build test optional weights, and the hardness test points are selected to exclude Screw holes and connections.

Benefits of technology

It improves the accuracy of hardness test, avoids damage to the seat members, and ensures the reliability of the test results and the safety of the seat.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118710642B_ABST
    Figure CN118710642B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of hardness testing, and specifically to a hardness testing method, system, and equipment for a work seat in a closed cabin, the method comprising: obtaining a grayscale image of a seat component and extracting edges; constructing a distribution rule index for each component edge; constructing a morphological difference coefficient for each component edge, combined with a distribution rule index for each component edge; constructing a first optional weight for each component edge; constructing a local disorder degree for each component edge; constructing a second optional weight for each component edge; constructing a test optional weight for each component edge, so as to obtain a seat component hardness test point and conduct a seat hardness test. The present application can screen out and exclude test points that may be important parts such as screw holes or connections when testing the hardness of a seat, thereby ensuring test accuracy and avoiding damage to seat components during the test.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of hardness testing, and in particular to a method, system, and equipment for testing the hardness of a work seat in a closed cabin. Background Art

[0002] With the development of economy, the safety requirements for enclosed cabins such as aerospace are gradually increasing. In addition to the safety performance of the enclosed cabin itself, the safety of the seats in the enclosed cabin has also been paid attention to. For the work seats in the enclosed cabin, their safety is mainly reflected in the hardness of their structure. Therefore, the importance of seat hardness testing in enclosed cabins has gradually attracted people's attention.

[0003] Usually, the hardness of sealed cabin seats is measured by Rockwell hardness measuring equipment, which is an automated Rockwell hardness measurement of the components of sealed cabin seats. The main process is to place the sealed cabin seat components on the test bench of the Rockwell hardness measuring equipment through a robotic arm. The Rockwell hardness measuring equipment automatically starts and selects multiple test points for pressure, and then measures the hardness of each point of the component.

[0004] However, in the process of automated testing, since the structure of the seat components themselves is relatively complex, the component bodies have different shapes, and the components have many parts such as screw holes and connection structures, when conducting hardness tests, if the selected test points are special parts of these components, it may not only cause irreversible damage to the seat connection structure, but also may cause deviations in the results of the hardness test. Therefore, image processing technology is usually used to exclude special parts of the components. However, since the screw holes and other parts are similar to the structural texture of the seat structure itself, the special parts of the components may be misjudged during the image processing process, which in turn leads to deviations in the subsequent test point selection. Summary of the invention

[0005] In a first aspect, an embodiment of the present application provides a method for testing the hardness of a work seat in a closed cabin, the method comprising the following steps:

[0006] Obtain the grayscale image of the seat component and extract the edge, which is recorded as the component edge;

[0007] All component edges are clustered to obtain clusters; the difference between the length of each component edge and the length of all component edges in the cluster and the discrete degree of all component edges are analyzed to construct the distribution regularity index of each component edge;

[0008] Constructing a morphological difference coefficient of each component edge according to the grayscale value difference and morphological difference of the pixel points of each component edge and the other component edges in the seat component grayscale image, and constructing a first optional weight of each component edge in combination with the distribution rule index;

[0009] According to the position distribution of each component edge and the edges of other components in the grayscale image of the seat component, the adjacent component edge and edge span of each component edge are obtained, and the local disorder degree of each component edge is constructed by combining the difference and distribution of the edge spans of each component edge and its adjacent component edges;

[0010] According to the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the seat component grayscale image, the characteristic deviation degree of each component edge is constructed, and the second optional weight of each component edge is constructed in combination with the local disorder degree of each component edge;

[0011] A test optional weight for each component edge is constructed based on the first optional weight and the second optional weight for each component edge to obtain a seat component hardness test point and perform a seat hardness test.

[0012] Preferably, the construction process of the distribution regularity index of each component edge is:

[0013] Analyze the variance of the distance between the midpoint of each component edge and the midpoints of all component edges in its cluster, and record it as the distribution rule variance of each component edge;

[0014] Analyze the mean difference between the length of each component edge and the length of all component edges in its cluster, and record it as the distribution length difference of each component edge;

[0015] The distribution rule index of each component edge is determined based on the distribution rule variance and the distribution length difference. The distribution rule variance and the distribution length difference of each component edge are both positively correlated with the distribution rule index of each component edge.

[0016] Preferably, the process of constructing the morphological difference coefficient of each component edge is:

[0017] When the edge of the component is a closed contour, a first preset value is used as the morphological difference weight of each component edge; when the edge of the component is not a closed contour, a second preset value is used as the morphological difference weight of each component edge, and the first preset value is greater than the second preset value;

[0018] Analyze the average grayscale value of all pixels within the minimum circumscribed rectangle of each component edge in the grayscale image of the seat component, and record it as the first form reflection value of each component edge;

[0019] Analyze the average of the differences between the first form reflectance value of each component edge in the grayscale image of the seat component and the first form reflectance values ​​of all other component edges, and record it as the second form reflectance value of each component edge;

[0020] Analyze the average of the second morphological reflectance values ​​of each component edge and all component edges in the seat component grayscale image, and record the morphological reflectance difference of each component edge;

[0021] The morphological difference coefficient of each component edge is determined based on the morphological difference weight and the morphological reflection difference. The morphological difference weight of each component edge is the result of forward fusion of the morphological difference weight of each component edge and the morphological reflection difference of each component edge.

[0022] Preferably, the process of constructing the first optional weight of each component edge is:

[0023] The first optional weight of each component edge is determined based on the distribution rule index and the morphological difference coefficient of each component edge, and the first optional weight of each component edge is the result of forward fusion of the distribution rule index and the morphological difference coefficient of each component edge.

[0024] Preferably, the method for obtaining the adjacent component edge and edge span of each component edge is:

[0025] The range of the ordinate and the range of the abscissa of each component edge are calculated and compared respectively, and the largest range is taken as the component edge span of each component edge;

[0026] The distance between the midpoint of each component edge and the midpoints of the remaining component edges in the seat component grayscale image is analyzed, and a preset number of component edges with the smallest distance are used as the adjacent component edges of each component edge.

[0027] Preferably, the construction process of the local disorder degree of each component edge is:

[0028] Analyze the difference between the distances between the midpoint of each component edge and the midpoints of all its adjacent component edges and take the average value, which is recorded as the first chaotic difference of each component edge;

[0029] Analyze the mean difference between the edge span of each component edge and the edge span of its adjacent component edge, which is recorded as the second chaotic difference;

[0030] The local disorder degree of each component edge is determined based on the first disorder difference and the second disorder difference, and the first disorder difference and the second disorder difference of each component edge are both positively correlated with the local disorder degree of each component edge.

[0031] Preferably, the step of constructing the characteristic deviation degree of each component edge and constructing the second optional weight of each component edge in combination with the local disorder degree of each component edge includes:

[0032] Analyze the distance between the midpoint of each component edge and the midpoint of each boundary of the seat component grayscale image, and take the boundary where the midpoint with the shortest distance to the midpoint of each component edge is located as the adjacent image boundary of each component edge;

[0033] The difference between the distance from the midpoint of each component edge to the midpoint of its adjacent image boundary and the distance from the midpoint of the remaining component edges to the midpoint of the adjacent image boundary is averaged, and recorded as the first deviation degree of each component edge;

[0034] Analyze the mean of the similarity between each component edge and all its adjacent component edges, and record it as the second deviation of each component edge;

[0035] Determine a characteristic deviation degree of each component edge based on the first deviation degree and the second deviation degree, wherein the characteristic deviation degree of each component edge is positively correlated with the first deviation degree of each component edge and negatively correlated with the second deviation degree of each component edge;

[0036] A second optional weight of each component edge is determined based on the local disorder degree and the characteristic deviation degree, and the second optional weight of each component edge is positively correlated with the local disorder degree and negatively correlated with the characteristic deviation degree.

[0037] Preferably, the step of constructing a test optional weight for each component edge to obtain a seat component hardness test point and perform a seat hardness test comprises:

[0038] Determine a test selectable weight for each component edge based on the first selectable weight and the second selectable weight, wherein the first selectable weight and the second selectable weight are both negatively correlated with the test selectable weight for each component edge;

[0039] Set the initial test optional weights of all pixels in the grayscale image of the seat component; use the test optional weight of each component edge as the weight of the pixel at the corresponding component edge;

[0040] The initial test optional weights of all pixel points on the edge of the component are added to the test optional weights to obtain a test point selection weight map;

[0041] The test point selection weight map is used as the input of the sampling method, and the preset number of sampling points is set to obtain the seat hardness test points and input them into the hardness testing equipment.

[0042] In a second aspect, an embodiment of the present application provides a hardness testing system for a work seat in a closed cabin, the system comprising:

[0043] The seat component recognition and preprocessing module obtains the grayscale image of the seat component and extracts the edge, which is recorded as the component edge;

[0044] The feature extraction seat analysis module clusters all component edges to obtain clusters; analyzes the difference between the length of each component edge and the length of all component edges in its cluster, as well as the discreteness of all component edges to construct a distribution rule index for each component edge;

[0045] Constructing a morphological difference coefficient of each component edge according to the grayscale value difference and morphological difference of the pixel points of each component edge and the other component edges in the seat component grayscale image, and constructing a first optional weight of each component edge in combination with the distribution rule index;

[0046] According to the position distribution of each component edge and the edges of other components in the grayscale image of the seat component, the adjacent component edge and edge span of each component edge are obtained, and the local disorder degree of each component edge is constructed by combining the difference and distribution of the edge spans of each component edge and its adjacent component edges;

[0047] According to the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the seat component grayscale image, the characteristic deviation degree of each component edge is constructed, and the second optional weight of each component edge is constructed in combination with the local disorder degree of each component edge;

[0048] The seat hardness test module constructs a test optional weight for each component edge based on the first optional weight and the second optional weight of each component edge to obtain a seat component hardness test point and perform a seat hardness test.

[0049] In the third aspect, an embodiment of the present application also provides a hardness testing device for a work seat in an enclosed cabin, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0050] It can be seen from the above embodiments that the hardness testing method of a work seat in a closed cabin provided in the embodiments of the present application has at least the following beneficial effects:

[0051] This application constructs the morphological difference coefficient of each component edge based on the gray value difference and morphological difference of all pixel points between each component edge and the rest of the component edges in the grayscale image of the seat component through the edge features of the screw holes and joints in the seat components, and constructs the first optional weight of each component edge in combination with the distribution rule index of each component edge; constructs the characteristic deviation of each component edge based on the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the grayscale image of the seat component, and constructs the second optional weight of each component edge in combination with the local chaos of each component edge; constructs the test optional weight of each component edge based on the first optional weight and the second optional weight of each component edge, so as to obtain the hardness test points of the seat components and conduct the seat hardness test. This application can screen out and exclude the test points that may be important parts such as screw holes or joints when testing the hardness of the seat, which can ensure the test accuracy and avoid damage to the seat components during the test. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0053] Figure 1 The present invention is a flowchart of a method for testing the hardness of a work seat in a closed cabin;

[0054] Figure 2 Extract flow chart for testing optional weights;

[0055] Figure 3 This is a schematic diagram of the seat hardness test process;

[0056] Figure 4 The block diagram of a hardness test system for a work seat in a closed cabin. DETAILED DESCRIPTION

[0057] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the hardness testing method, system, and equipment for a work seat in a closed cabin proposed in accordance with the present application, its specific implementation method, structure, characteristics, and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0058] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0059] The following is a detailed description of the specific scheme of the hardness testing method, system and equipment for a work seat in a closed cabin provided by the present application in conjunction with the accompanying drawings.

[0060] See also Figure 1 , which shows a flowchart of a hardness testing method for a work seat in a closed cabin provided by an embodiment of the present application, the method comprising the following steps:

[0061] Step S1: Obtain a grayscale image of a seat component and extract edges.

[0062] A CMOS high-definition camera and an industrial light source are set above the test bench of the Rockwell hardness tester to illuminate the seat components and collect image data of the seat components to obtain a high-definition RGB image of the seat components. The RGB image of the seat components is converted into a grayscale image, and the grayscale image of the seat components is denoised and sharpened using a filtering algorithm and an edge enhancement algorithm to finally obtain a component grayscale image.

[0063] It should be noted that there are many commonly used filtering algorithms and edge enhancement algorithms. In this embodiment, mean filtering and Laplace operator are used to denoise and sharpen the grayscale image of the seat component. Under the premise of ensuring the purpose of denoising and sharpening, the implementer may also use other filtering and edge enhancement algorithms, and this embodiment does not impose any special restrictions.

[0064] Among them, mean filtering and Laplace operator are well-known technologies, and the specific processes of denoising and sharpening are not described in detail.

[0065] Step S2: cluster all component edges to obtain clusters; analyze the difference between the length of each component edge and the length of all component edges in the cluster to which it belongs, and the discrete degree of all component edges to construct a distribution rule index for each component edge; construct a morphological difference coefficient for each component edge based on the grayscale value difference and morphological difference of the pixel points between each component edge and the remaining component edges in the grayscale image of the seat component, and construct a first optional weight for each component edge in combination with the distribution rule index.

[0066] There are usually special structures on the surface of the seat components of the closed cabin, such as hollowing formed to save costs, different convex and concave positions formed for aesthetic design needs, etc. These structures are mainly shown as unevenly distributed edges in the image. There are also usually screw holes, connection holes and connection interfaces on the seat components, and these pores are usually shown as relatively regularly distributed closed edge contours in the image. Because there is usually a threaded metal surface inside the screw hole, the metal surface will have a more obvious reflective effect under the illumination of the light source.

[0067] S201: Analyze the length of the edge of the component, and obtain each cluster by combining the preset cutoff distance and the clustering algorithm.

[0068] In the grayscale image of the seat components, since the shapes and sizes of the edge contours of the seat components such as screw holes and seat connections are different, in order to better analyze and obtain the hardness test points in the seat components, the component edges in the grayscale image of the seat components need to be clustered and divided, specifically:

[0069] Edges in the grayscale image of seat components are extracted using edge detection and morphological algorithms, recorded as component edges, and then the closed contour is obtained using contour tracking. The total number of pixels on each component edge is counted and used as the length of each component edge. The lengths of all component edges and the preset cutoff distance are used as the input of the clustering algorithm, and the component edges within the preset cutoff distance are obtained to form each cluster. For each component edge, its minimum circumscribed rectangle is drawn.

[0070] The value of the preset cutoff distance is artificially set. In this embodiment, the value of the preset cutoff distance is 3. The implementer can set it according to the specific situation. This embodiment does not impose any special restrictions.

[0071] It should be noted that there are many commonly used edge detection algorithms, morphological algorithms and clustering algorithms. In this embodiment, the Canny edge detection algorithm and the expansion corrosion algorithm are used to obtain the edges in the grayscale image of the seat components, and the DPC clustering algorithm is used to cluster the edges of the components. In addition, the implementer can also use other methods to extract the edges of the components in the image and cluster the edges according to the length. This embodiment does not impose any special restrictions.

[0072] Among them, the Canny edge detection algorithm, the dilation-erosion algorithm, the contour tracking method and the DPC clustering algorithm are all well-known technologies, and the specific processes of edge extraction, contour tracking and clustering are not described in detail.

[0073] S202: Analyze the difference, distribution status and discrete degree between the length of each component edge and the length of all component edges in the cluster to which it belongs, and construct a distribution rule index for each component edge.

[0074] In the process of automated testing, since the structure of the seat components themselves is relatively complex, the component bodies have different shapes, and there are many parts such as screw holes in the components, when conducting hardness tests, if the selected test points are special parts of these components, it may not only cause irreversible damage to the seat connection structure, but also may cause deviations in the results of the hardness test. Therefore, in order to accurately identify the screw hole structure in the grayscale image of the seat components, it is necessary to analyze the distribution status, discreteness and length of the component edges in the clusters in the grayscale image of the seat components.

[0075] Specifically, the distribution regularity index of each component edge is constructed according to the difference between the length of each component edge and the length of all component edges in the cluster to which it belongs, the distribution status and the degree of dispersion. The specific process is as follows:

[0076] Analyze the variance of the distance between the midpoint of each component edge and the midpoints of all component edges in its cluster, and record it as the distribution rule variance of each component edge;

[0077] It should be noted that there are many methods for measuring the distance between two points. As an embodiment of the present application, this embodiment measures the Euclidean distance between each component edge and the midpoints of all edges in its cluster to measure the distance between the midpoint of each component edge and the center of all component edges in its cluster. The implementer may also use Manhattan distance, Mahalanobis distance and other methods to measure the distance between the midpoints of component edges. This embodiment does not make special requirements.

[0078] Among them, the specific calculation process of Euclidean distance, Manhattan distance and Mahalanobis distance is a well-known technology, and the specific process will not be repeated here.

[0079] Furthermore, the mean difference between the length of each component edge and the length of each component edge in its cluster is analyzed and recorded as the distribution length difference of each component edge.

[0080] It should be noted that, as an embodiment of the present application, in this embodiment, the difference between the lengths is measured by taking the absolute value of the difference between the length of each component edge and the length of each edge in its cluster. In addition, the implementer may also measure the difference between the lengths of component edges by Euclidean distance, Manhattan distance, etc., which is not particularly limited in this embodiment.

[0081] Among them, the specific calculation process of Euclidean distance, Manhattan distance and Mahalanobis distance is a well-known technology, and the specific process will not be repeated here.

[0082] Furthermore, the distribution regularity index of each component edge is determined based on the distance variance and the comprehensive length, and the distribution regularity variance and distribution length difference of each component edge are positively correlated with the distribution regularity index of each component edge.

[0083] It should be understood that a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship can be an additive relationship or a multiplicative relationship, etc., which is determined by the actual application and is not specifically limited in this application.

[0084] Preferably, as an embodiment of the present application, the distribution rule index of each component edge is the inverse of the product of the distribution rule variance of each component edge and the distribution length difference.

[0085] Preferably, as another embodiment of the present application, the distribution rule index of each component edge is an exponential function value with a natural constant as the base and the sum of the opposite of the distribution rule variance of each component edge and the opposite of the distribution length difference as the independent variable.

[0086] When the distribution regularity index of each component edge is larger, it means that the distribution intervals of all component edges in the cluster where the component edge is located are relatively similar, and the length of each component edge is similar, and the component edge is more likely to be an important part of the component such as a screw hole; conversely, when the distribution regularity index of each component edge is smaller, it means that the distribution intervals of all component edges in the cluster where the component edge is located are not similar, and the component edge is unlikely to be an important part of the component such as a screw hole.

[0087] S203: constructing a morphological difference coefficient of each component edge according to the grayscale value difference and morphological difference of all pixel points of each component edge and the other component edges in the seat component grayscale image.

[0088] There is a big difference in contrast between the screw holes and their surroundings. Screw holes are generally made of reflective materials. Therefore, in order to more accurately identify the edges of important parts such as screw holes, the interference of screw holes in the point selection process is eliminated based on the brightness difference of all pixel points between the edge of each component and the edges of other components in the grayscale image of the seat component.

[0089] Specifically, the morphological difference coefficient of each component edge is constructed according to the gray value difference and morphological difference of all pixels of each component edge and the other component edges in the seat component grayscale image, which is:

[0090] When the component edge is a closed contour, a first preset value is assigned to each component edge as the morphological difference weight of each component edge; when the component edge is not a closed contour, a second preset value is assigned to each component edge as the morphological difference weight of each component edge.

[0091] It should be noted that the first preset value is greater than the second preset value, and the values ​​of the first preset value and the second preset value are set manually. In this embodiment, the first preset value is 2, and the second preset value is 1. This embodiment does not impose any special restrictions. Under the premise that the first preset value is greater than the second preset value, the implementer can set it according to the specific situation.

[0092] Further, the average grayscale value of all pixels within the minimum circumscribed rectangle of each component edge in the grayscale image of the seat component is analyzed and recorded as the first form reflective value of each component edge;

[0093] Analyze the average of the differences between the first form reflectance value of each component edge in the grayscale image of the seat component and the first form reflectance values ​​of all other component edges, and record it as the second form reflectance value of each component edge;

[0094] Analyze the average of the second morphological reflectance values ​​of each component edge and all component edges in the seat component grayscale image, and record the morphological reflectance difference of each component edge;

[0095] It should be noted that, as an embodiment of the present application, the grayscale difference between the edges is measured by taking the absolute value of the difference between the grayscale values ​​of all similar points within the minimum circumscribed rectangle of each component edge and all other component edges in the seat component grayscale image.

[0096] Further, the morphological difference coefficient of each component edge is determined based on the morphological difference weight and the morphological reflection difference, and the morphological difference weight of each component edge is the result of forward fusion of the morphological difference weight of each component edge and the morphological reflection difference of each component edge.

[0097] It should be understood that forward fusion refers to combining two or more indicators by adding or multiplying them together in order to obtain a comprehensive indicator, thereby more comprehensively and accurately evaluating a phenomenon or problem. This fusion method is not limited to simple arithmetic operations, but can also include more complex statistical models and analysis methods. The implementer can choose according to the specific situation, and this embodiment does not impose any special restrictions.

[0098] Preferably, as an embodiment of the present application, the morphological difference weight of each component edge is the product of the morphological difference weight of each component edge and each component edge.

[0099] Preferably, as another embodiment of the present application, the morphological difference weight of each component edge is an exponential function value with a natural constant as the base and the sum of the morphological difference weight of each component edge and each component edge as the independent variable.

[0100] When the morphological difference coefficient of each component edge is larger, it means that the component edge is a closed contour and the reflective effect is more obvious than the reflective effects of other component edges, and it is more likely to be a screw hole on the component; conversely, when the morphological difference coefficient of each component edge is smaller, it means that the component edge is a closed contour and the reflective effect is less obvious than the reflective effects of other component edges, and it is impossible to be a screw hole on the component.

[0101] S204: Determine a first optional weight of each component edge according to the distribution rule index and the morphological difference coefficient of each component edge.

[0102] In order to avoid selecting hardness test points at screw holes and seat connections, the distribution regularity index and morphological difference coefficient of each component edge are combined to accurately select points.

[0103] Specifically, the first optional weight of each component edge is determined based on the distribution rule index and the morphological difference coefficient of each component edge, and the first optional weight of each component edge is the result of forward fusion of the distribution rule index and the morphological difference coefficient of each component edge.

[0104] It should be understood that forward fusion refers to combining two or more indicators by adding or multiplying them together in order to obtain a comprehensive indicator, thereby more comprehensively and accurately evaluating a phenomenon or problem. This fusion method is not limited to simple arithmetic operations, but can also include more complex statistical models and analysis methods. The implementer can choose according to the specific situation, and this embodiment does not impose any special restrictions.

[0105] Preferably, as an embodiment of the present application, the first optional weight of each component edge is the product of the distribution regularity index of each component edge and the morphological difference coefficient.

[0106] Preferably, as another embodiment of the present application, the component process of the first optional weight of each component edge is: calculate the logarithmic function value with the natural constant as the base and the distribution rule index of each component edge as the independent variable, recorded as the first important function value; calculate the logarithmic function value with the natural constant as the base and the morphological difference coefficient of each component edge as the independent variable, recorded as the second important function value; the first optional weight of each component edge is the sum of the first important function value and the second important function value of each component edge.

[0107] Preferably, as another embodiment of the present application, the first optional weight of each component edge is an exponential function value with a natural constant as the base and the sum of the distribution rule index and the morphological difference coefficient of each component edge as the independent variable.

[0108] When the distribution regularity index of each component edge is larger and the morphological difference coefficient is larger, the first optional weight of the corresponding component edge is larger, which means that the component edge is more likely to be the edge of important parts such as screw holes on the seat component, and these positions cannot be selected for hardness testing; conversely, when the distribution regularity index of each component edge is smaller and the morphological difference coefficient is smaller, the first optional weight of the corresponding component edge is smaller, which means that the edge is less likely to be the edge of important parts such as screw holes on the seat component, and these positions can be selected for hardness testing.

[0109] Step S3: construct the local chaos degree of each component edge according to the difference and distribution of the edge span between each component edge and its adjacent component edge; construct the characteristic deviation degree of each component edge according to the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the seat component grayscale image; and construct the second optional weight of each component edge in combination with the local chaos degree of each component edge.

[0110] However, in addition to screw holes, the joints of seat components are also very important. If pressure is applied to the joints of components when selecting hardness test points, it is likely to cause deformation and damage to the joints, which will lead to the inability to guarantee the strength of the assembled seats during subsequent assembly, resulting in hidden dangers. Usually, the connection parts on seat components will have relatively complex patterns due to different connection methods, and there will also be certain screw holes for connection fixation. Therefore, the distribution of component edges at the connection parts is more complex and closer to the boundary of the entire component. Compared with the component edges at the connection parts, the component edges formed in other situations have a relatively large span and are unevenly distributed.

[0111] Calculate the range of the ordinate and the abscissa of each component edge respectively, compare the range of the ordinate and the abscissa, and select the maximum value as the component edge span of each component edge. Analyze the distance between the midpoint of each component edge and the midpoints of the other component edges in the grayscale image of the seat component, and select the preset number of component edges with the smallest distance as the adjacent component edges of each component edge.

[0112] It should be noted that the value of the preset number is artificially set. In this embodiment, the value of the preset number is 10. The implementer can set it according to the specific situation. This embodiment does not impose any special restrictions.

[0113] S301: Constructing the local disorder degree of each component edge according to the difference and distribution of the edge span of each component edge and its adjacent component edges.

[0114] The connection structure of the seat is a key part to ensure the overall stability and safety of the seat. Hardness testing in these areas may cause damage to the connection structure, affecting the integrity and riding safety of the seat. Moreover, the material and structure of the connection are different from those of other parts of the seat, which may lead to deviations in the hardness test results and fail to accurately reflect the overall hardness characteristics of the seat. Therefore, it is necessary to identify the connection structure in the seat components by analyzing the difference and distribution of the edge span between the edge of each component and its adjacent component edge, and avoid selecting points at this point.

[0115] Specifically, the local disorder degree of each component edge is constructed according to the difference and distribution of the edge span of each component edge and its adjacent component edges, which is:

[0116] Analyze the mean of the difference between the midpoint of each component edge and the midpoints of all its adjacent component edges, and record it as the first chaotic difference of each component edge;

[0117] Furthermore, the mean difference between the edge span of each component edge and the edge span of its adjacent component edge is analyzed and recorded as the second chaotic difference.

[0118] It should be noted that, in this embodiment, the Euclidean distance is used to analyze the distance between the midpoint of each component edge and the midpoints of all its adjacent component edges. The implementer may also use the Mahalanobis distance or the Manhattan distance to measure the distance between the component edge and the adjacent component edges.

[0119] At the same time, this embodiment characterizes the difference between the distances between the midpoint of each component edge and the midpoints of all its adjacent component edges by taking the absolute value of the difference between the distances between the midpoint of each component edge and the midpoints of all its adjacent component edges;

[0120] The absolute value of the difference between the edge span of each component edge and its adjacent component edge is used to measure the difference between the component edge and the adjacent component edge. The implementer may also use other methods to measure the difference, which is not particularly limited in this embodiment.

[0121] Among them, Euclidean distance, Mahalanobis distance and Manhattan distance are all well-known technologies, and the specific process will not be repeated here.

[0122] Further, based on the first disorder difference and the second disorder difference, the first disorder difference and the second disorder difference of each component edge are positively correlated with the local disorder degree of each component edge.

[0123] It should be understood that a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship can be an additive relationship or a multiplicative relationship, etc., which is determined by the actual application and is not specifically limited in this application.

[0124] Preferably, as an embodiment of the present application, the local disorder degree of each component edge is the product of the first disorder difference and the second disorder difference of each component edge.

[0125] Preferably, as another embodiment of the present application, the local disorder degree of each component edge is a logarithmic function value with a natural constant as the base and the sum of the first disorder difference and the second disorder difference of each component edge as the independent variable.

[0126] The greater the local disorder of each component edge, the higher the degree of distribution disorder of the edges of the adjacent components at the location of the component edge, the higher the degree of morphological disorder, and the greater the difference in the characteristics of the component edge, the more likely it is to be the edge of the connection; conversely, the smaller the local disorder of each component edge, the lower the degree of distribution disorder of the edges of the adjacent components at the location of the component edge, the lower the degree of morphological disorder, and the smaller the difference in the characteristics of the component edge, the less likely it is to be the edge of the connection.

[0127] S302: constructing a characteristic deviation degree of each component edge according to the similarity between each component edge and the adjacent component edge and the position distribution difference between each component edge and the boundary of the seat component grayscale image.

[0128] The component edge distribution at the connection is more complex and closer to the boundary of the entire component. The component edges formed in other situations have a relatively large span and uneven distribution compared to the component edges at the connection. Therefore, the edges at the connection are identified and excluded by analyzing the similarity between each component edge and the edges of the adjacent components, as well as the position distribution difference between each component edge and the boundary of the grayscale image of the seat component.

[0129] Specifically, the characteristic deviation of each component edge is constructed according to the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the seat component grayscale image, which is specifically:

[0130] Analyze the distance between the midpoint of each component edge and the midpoint of each boundary of the seat component grayscale image, and take the boundary where the midpoint with the shortest distance to the midpoint of each component edge is located as the adjacent image boundary of each component edge;

[0131] It should be noted that in this embodiment, the Euclidean distance is used to measure the distance between the midpoint of the component edge and the midpoint of each boundary of the seat component grayscale image. The implementer can also use Mahalanobis distance, Manhattan distance, etc. to measure the difference between the component edge and each boundary of the seat component grayscale image. This embodiment does not impose any special restrictions.

[0132] Among them, the specific calculation process of Euclidean distance, Manhattan distance and Mahalanobis distance is a well-known technology, and the specific process will not be repeated here.

[0133] Further, the average of the differences between the distance between the midpoint of each component edge and the midpoint of its adjacent image boundary and the distance between the midpoint of all other component edges in the image and the midpoint of their adjacent image boundaries is recorded as the first deviation degree of each component edge;

[0134] It should be noted that there are many methods for measuring differences. This embodiment uses the method of taking the absolute value of the difference to measure the difference. The implementer may also use a ratio or other methods for measuring the difference, and this embodiment does not impose any special restrictions.

[0135] Further, the mean of the similarities between each component edge and all its adjacent component edges is analyzed and recorded as the second deviation of each component edge;

[0136] It should be noted that there are many methods to measure the similarity between each component edge and the edges of all its neighboring components. In this embodiment, ESSIM edge structure similarity is used to measure the similarity between the component edge and the edges of neighboring components. Implementers can also use Euclidean distance, cosine similarity or histogram methods to analyze the similarity between edges.

[0137] Among them, ESSIM edge structure similarity, Euclidean distance, cosine similarity and histogram are all well-known technologies, and the specific calculation process of the similarity will not be repeated here.

[0138] Further, the characteristic deviation of each component edge is determined based on the first deviation and the second deviation, and the characteristic deviation of each component edge is positively correlated with the first deviation of each component edge and negatively correlated with the second deviation of each component edge.

[0139] It should be understood that a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship may be an additive relationship or a multiplicative relationship, etc., which is determined by the actual application and this application does not impose any special restrictions. A negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It may be a subtractive relationship or a division relationship, etc., which is determined by the actual application.

[0140] Preferably, as an embodiment of the present application, the characteristic deviation degree of each component edge is the ratio of the first deviation degree to the second deviation degree of each component edge.

[0141] Preferably, as another embodiment of the present application, the characteristic deviation of each component edge is an exponential function value with a natural constant as the base and a difference between the first deviation and the second deviation of each component edge as the independent variable.

[0142] The larger the characteristic deviation of each component edge, the larger the difference in positional distribution between the component edge and the edges of other adjacent components, and the lower the degree of similarity, the more likely it is that it is the edge of the connection of the seat components; conversely, the smaller the characteristic deviation of each component edge, the smaller the difference in positional distribution between the component edge and the edges of other adjacent components, and the higher the degree of similarity, the less likely it is that it is the edge of the connection of the seat components.

[0143] S303: Calculate a second optional weight for each component edge according to the local disorder degree of each component edge and the characteristic deviation degree of the edge.

[0144] According to the above analysis, the second optional weight of each component edge is determined by combining the local disorder degree and the characteristic deviation degree, so as to screen out the influence of the seat connection on the selection of hardness test points. The construction process of the second optional weight of each component edge is:

[0145] The second optional weight of each component edge is determined based on the local chaos and the characteristic deviation. The second optional weight of each component edge is positively correlated with the local chaos of each component edge and negatively correlated with the characteristic deviation of each component edge.

[0146] It should be understood that a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship may be an additive relationship or a multiplicative relationship, etc., which is determined by the actual application and this application does not impose any special restrictions. A negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It may be a subtractive relationship or a division relationship, etc., which is determined by the actual application.

[0147] Preferably, as an embodiment of the present application, the second optional weight of each component edge is the ratio of the local disorder degree to the characteristic deviation degree of each component edge.

[0148] When the second optional weight of a single edge is larger, it means that the edge distribution of the adjacent component where the edge is located is more chaotic, the morphological disorder is higher, and the feature difference of the edge is larger, and the edge is more likely to be the edge at the component connection; conversely, when the second optional weight of a single edge is smaller, it means that the edge distribution of the adjacent component where the edge is located is less chaotic, the morphological disorder is lower, and the feature difference of the edge is smaller, and the edge is less likely to be the edge at the component connection.

[0149] Step S4: constructing a test optional weight of each component edge based on the first optional weight and the second optional weight of each component edge.

[0150] When conducting a hardness test, if the selected test points are special locations such as screw holes or seat connections, it may not only cause irreversible damage to the seat connection structure, but may also cause deviations in the hardness test results. Therefore, the first optional weight and the second optional weight are combined to exclude structures such as screw holes and seat connections in seat components.

[0151] Specifically, according to the first optional weight and the second optional weight of each component edge, the test optional weight of each component edge is constructed, specifically:

[0152] The test selectable weight of each component edge is determined based on the first selectable weight and the second selectable weight, and the first selectable weight and the second selectable weight of each component edge are negatively correlated with the test selectable weight of each component edge.

[0153] It should be understood that a negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtraction relationship or a division relationship, etc., which is determined by the actual application.

[0154] Preferably, as an embodiment of the present application, the test optional weight of each component edge is the reciprocal of the sum of the first optional weight and the second optional weight of each component edge.

[0155] Preferably, as another embodiment of the present application, the test optional weight of each component edge is the inverse of the product of the first optional weight and the second optional weight of each component edge.

[0156] Preferably, as another embodiment of the present application, the test optional weight of each component edge is an exponential function value with a natural constant as the base and the sum of the opposite of the first optional weight and the opposite of the second optional weight of each component edge as the independent variable.

[0157] The larger the optional test weight of each component edge is, the less likely it is that the component edge is a special location such as a screw hole or connection on the seat component, and is therefore more suitable as a selection point for hardness testing.

[0158] Preferably, as an embodiment of the present application, the test optional weight extraction flow chart is as follows Figure 2 shown.

[0159] Step S5: Obtaining seat component hardness test points based on the test optional weight of each component edge and performing a seat hardness test.

[0160] First, the initial test optional weights of all pixels in the grayscale image of the seat component are set to a third preset value, and then the test optional weights calculated in step S4 are assigned to the pixels contained in the corresponding edge and added to the initial test optional weights of the corresponding pixels, thereby obtaining a test point selection weight map.

[0161] It should be noted that the value of the third preset numerical value is set manually. As an embodiment of the present application, in this embodiment, the third preset numerical value is 1. The implementer can also set it according to the specific situation. This embodiment does not impose any special restrictions.

[0162] The smaller the weight, the more likely the pixel point is a special location such as a screw hole or connection on the seat, and the less likely it is to be selected. The test point selection weight map is used as the input of the sampling method, and the preset number of sampling points is set to obtain the seat hardness test points.

[0163] It should be noted that there are many commonly used sampling methods. As an embodiment of the present application, this embodiment adopts weighted random sampling to select test sampling points. The implementer may also adopt stratified sampling or priority sampling and other methods for sampling. This embodiment does not impose any special restrictions. Among them, the weighted random sampling method is a well-known technology, and the specific process of its sampling is not repeated here.

[0164] The value of the preset number of sampling points is artificially set. In this embodiment, the value of the preset number of sampling points is 5. The implementer can set it according to the specific situation. This embodiment does not impose any special restrictions.

[0165] The hardness automation test system inputs the output seat hardness test points into the Rockwell hardness test equipment, and the Rockwell hardness test equipment applies pressure, releases pressure, and calculates the hardness according to the seat hardness test points.

[0166] Preferably, as an embodiment of the present application, the seat hardness test process diagram is as follows: Figure 3 shown.

[0167] At this point, based on the test optional weights, the sampling and testing of the seat hardness test points are completed.

[0168] Based on the same inventive concept as the above method, the embodiment of the present application also provides a hardness testing system for a work seat in a closed cabin, comprising:

[0169] The seat component recognition and preprocessing module obtains the grayscale image of the seat component and extracts the edge, which is recorded as the component edge;

[0170] The feature extraction seat analysis module clusters all component edges to obtain clusters; analyzes the difference between the length of each component edge and the length of all component edges in the cluster to which it belongs and the discrete degree of all component edges to construct the distribution rule index of each component edge;

[0171] Constructing a morphological difference coefficient of each component edge according to the grayscale value difference and morphological difference of the pixel points of each component edge and the other component edges in the seat component grayscale image, and constructing a first optional weight of each component edge in combination with the distribution rule index;

[0172] According to the position distribution of each component edge and the edges of other components in the grayscale image of the seat component, the adjacent component edge and edge span of each component edge are obtained, and the local disorder degree of each component edge is constructed by combining the difference and distribution of the edge spans of each component edge and its adjacent component edges;

[0173] According to the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the seat component grayscale image, the characteristic deviation degree of each component edge is constructed, and the second optional weight of each component edge is constructed in combination with the local disorder degree of each component edge;

[0174] The seat hardness test module constructs a test optional weight for each component edge based on the first optional weight and the second optional weight of each component edge to obtain a seat component hardness test point and perform a seat hardness test.

[0175] The block diagram of a hardness testing system for a work seat in a closed cabin provided in an embodiment of the present application is as follows: Figure 4 shown.

[0176] Based on the same inventive concept as the above method, an embodiment of the present application also provides a hardness testing device for a work seat in an enclosed cabin, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods for hardness testing a work seat in an enclosed cabin when executing the computer program.

[0177] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0178] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0179] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for testing the hardness of a work seat in a closed cabin, characterized in that: The method comprises the following steps: Obtain the grayscale image of the seat component and extract the edge, which is recorded as the component edge; All component edges are clustered to obtain clusters; the difference between the length of each component edge and the length of all component edges in the cluster where it is located, as well as the discrete degree of the distance between the midpoints of all component edges are analyzed to construct the distribution regularity index of each component edge; Constructing a morphological difference coefficient of each component edge according to the grayscale value difference and morphological difference of the pixel points of each component edge and the other component edges in the seat component grayscale image, and constructing a first optional weight of each component edge in combination with the distribution rule index; According to the position distribution of each component edge and the edges of other components in the grayscale image of the seat component, the adjacent component edge and edge span of each component edge are obtained, and the local disorder degree of each component edge is constructed by combining the difference and distribution of the edge spans of each component edge and its adjacent component edges; According to the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the seat component grayscale image, the characteristic deviation degree of each component edge is constructed, and the second optional weight of each component edge is constructed in combination with the local disorder degree of each component edge; constructing a test optional weight for each component edge based on the first optional weight and the second optional weight for each component edge to obtain a seat component hardness test point and perform a seat hardness test; The construction process of the distribution rule index of each component edge is: Analyze the variance of the distance between the midpoint of each component edge and the midpoints of all component edges in its cluster, and record it as the distribution rule variance of each component edge; Analyze the mean difference between the length of each component edge and the length of all component edges in its cluster, and record it as the distribution length difference of each component edge; Determining a distribution rule index of each component edge based on the distribution rule variance and the distribution length difference, wherein the distribution rule variance and the distribution length difference of each component edge are both positively correlated with the distribution rule index of each component edge; The construction process of the morphological difference coefficient of each component edge is: When the edge of the component is a closed contour, a first preset value is used as the morphological difference weight of each component edge; when the edge of the component is not a closed contour, a second preset value is used as the morphological difference weight of each component edge, and the first preset value is greater than the second preset value; Analyze the average grayscale value of all pixels within the minimum circumscribed rectangle of each component edge in the grayscale image of the seat component, and record it as the first form reflection value of each component edge; Analyze the average of the differences between the first form reflectance value of each component edge in the grayscale image of the seat component and the first form reflectance values ​​of all other component edges, and record it as the second form reflectance value of each component edge; Analyze the average of the second morphological reflectance values ​​of each component edge and all component edges in the seat component grayscale image, and record the morphological reflectance difference of each component edge; Determine a morphological difference coefficient of each component edge based on the morphological difference weight and the morphological reflection difference, wherein the morphological difference coefficient of each component edge is a result of forward fusion of the morphological difference weight of each component edge and the morphological reflection difference of each component edge; The construction process of the first optional weight of each component edge is: Determine a first optional weight of each component edge based on a distribution rule index and a morphological difference coefficient of each component edge, wherein the first optional weight of each component edge is a result of forward fusion of the distribution rule index and the morphological difference coefficient of each component edge; The construction process of the local disorder degree of each component edge is: Analyze the difference between the distances between the midpoint of each component edge and the midpoints of all its adjacent component edges and take the average value, which is recorded as the first chaotic difference of each component edge; Analyze the mean difference between the edge span of each component edge and the edge span of its adjacent component edge, which is recorded as the second chaotic difference; Determining the local disorder degree of each component edge based on the first disorder difference and the second disorder difference, wherein the first disorder difference and the second disorder difference of each component edge are positively correlated with the local disorder degree of each component edge; The step of constructing the characteristic deviation degree of each component edge and constructing the second optional weight of each component edge in combination with the local disorder degree of each component edge includes: Analyze the distance between the midpoint of each component edge and the midpoint of each boundary of the seat component grayscale image, and take the boundary where the midpoint with the shortest distance to the midpoint of each component edge is located as the adjacent image boundary of each component edge; The difference between the distance from the midpoint of each component edge to the midpoint of its adjacent image boundary and the distance from the midpoint of the remaining component edges to the midpoint of the adjacent image boundary is averaged, and recorded as the first deviation degree of each component edge; Analyze the mean of the similarities between each component edge and all its adjacent component edges, and record it as the second deviation of each component edge; Determine a characteristic deviation degree of each component edge based on the first deviation degree and the second deviation degree, wherein the characteristic deviation degree of each component edge is positively correlated with the first deviation degree of each component edge and negatively correlated with the second deviation degree of each component edge; A second optional weight of each component edge is determined based on the local disorder degree and the characteristic deviation degree, and the second optional weight of each component edge is positively correlated with the local disorder degree and negatively correlated with the characteristic deviation degree.

2. A method for testing the hardness of a work seat in a closed cabin as claimed in claim 1, characterized in that: The method for obtaining the adjacent component edge and edge span of each component edge is: The range of the ordinate and the range of the abscissa of each component edge are calculated and compared respectively, and the largest range is taken as the component edge span of each component edge; The distance between the midpoint of each component edge and the midpoints of the remaining component edges in the seat component grayscale image is analyzed, and a preset number of component edges with the smallest distance are used as the adjacent component edges of each component edge.

3. A method for testing the hardness of a work seat in a closed cabin as claimed in claim 1, characterized in that: The method of constructing the optional test weight of each component edge to obtain the seat component hardness test point and perform the seat hardness test includes: Determine a test selectable weight for each component edge based on the first selectable weight and the second selectable weight, wherein the first selectable weight and the second selectable weight are both negatively correlated with the test selectable weight for each component edge; Set the initial test optional weights of all pixels in the grayscale image of the seat component; use the test optional weight of each component edge as the weight of the pixel at the corresponding component edge; The initial test optional weights of all pixel points on the edge of the component are added to the test optional weights to obtain a test point selection weight map; The test point selection weight map is used as the input of the sampling method, and the preset number of sampling points is set to obtain the seat hardness test points and input them into the hardness testing equipment.

4. A hardness testing system for a work seat in a closed cabin, implementing the method as claimed in claim 1, characterized in that: The system comprises: The seat component recognition and preprocessing module obtains the grayscale image of the seat component and extracts the edge, which is recorded as the component edge; The feature extraction seat analysis module clusters all component edges to obtain clusters; analyzes the difference between the length of each component edge and the length of all component edges in its cluster, as well as the discrete degree of the distance between the midpoints of all component edges to construct the distribution rule index of each component edge; Constructing a morphological difference coefficient of each component edge according to the grayscale value difference and morphological difference of the pixel points of each component edge and the other component edges in the seat component grayscale image, and constructing a first optional weight of each component edge in combination with the distribution rule index; According to the position distribution of each component edge and the edges of other components in the grayscale image of the seat component, the adjacent component edge and edge span of each component edge are obtained, and the local disorder degree of each component edge is constructed by combining the difference and distribution of the edge spans of each component edge and its adjacent component edges; According to the similarity between each component edge and the adjacent component edge, and the position distribution difference between each component edge and the boundary of the seat component grayscale image, the characteristic deviation degree of each component edge is constructed, and the second optional weight of each component edge is constructed in combination with the local disorder degree of each component edge; The seat hardness test module constructs a test optional weight for each component edge based on the first optional weight and the second optional weight of each component edge to obtain a seat component hardness test point and perform a seat hardness test.

5. A hardness testing device for a work seat in a closed cabin, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 3 are implemented.