Method for detecting opposite edge distance at crimping part of high-voltage wire based on computer vision

Abstract

The invention discloses a method for detecting the opposite edge distance of a high-voltage wire crimping position based on computer vision, and belongs to the technical field of wire crimping quality detection.The method comprises the steps that a geometric model is constructed according to the hexagonal cross section of the wire crimping position, and a preset target of an opposite edge distance detection task is determined according to the geometric model; obtaining the required contour distances in the geometric model through camera equipment and an edge detection algorithm; according to the obtained contour distances and the preset target, obtaining a detection result of the edge distance detection task; and obtaining the opposite edge distance according to a detection result. The contour distance of a geometric model at a wire crimping position is obtained by using an edge detection algorithm, whether the geometric model is a regular hexagon or not is judged according to the contour distance, and the non-regular hexagon degree of a measured joint is obtained according to edge distance measurement data. A data basis is provided for product quality judgment, the measurement error of manual measurement is reduced, the crimping quality of the high-voltage wire is guaranteed, and the production efficiency of the product is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of wire crimping quality detection, and in particular relates to a method for detecting edge distance at a high-voltage wire crimping point based on computer vision. Background Art

[0002] The crimping of conductor joints in high-voltage transmission lines is a very important step in the installation process of high-voltage transmission lines. The quality of crimping is directly related to the long-term stable operation of the power grid. With the improvement of power grid transmission standards, the cross-section of conductors continues to increase. When connecting large-section conductors, the resistance may increase due to insufficient edge distance at the crimping point, which may easily cause overheating and disconnection accidents during long-term operation. During daily operation and inspection or line reverse power transmission, defects at the crimping point are gradually exposed, but there is currently no quantitative evaluation and measurement of the edge distance of the conductor crimping.

[0003] The crimp between the wire clamp and the conductor is an extruded aluminum tube, which is hexagonal in shape after production and processing. Existing patents are mainly aimed at the measurement of the side length of a regular hexagon, but most of them use manual and mechanical contact measurement, which not only has low measurement efficiency, but people's subjective judgment will also cause errors in the measurement results. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision. By constructing a geometric model of the wire crimping point and obtaining the contour distance in the geometric model through an edge detection algorithm, high-precision margin distance detection of multiple parts can be achieved, thereby improving work efficiency and measurement accuracy and ensuring product quality.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] The present invention provides a method for detecting the margin distance of a high-voltage wire crimping point based on computer vision, comprising:

[0007] The geometric model is constructed based on the hexagonal cross section of the wire crimp;

[0008] Determining a preset target for a margin detection task according to the geometric model;

[0009] Acquire the distance of each contour in the geometric model by using a camera device and an edge detection algorithm;

[0010] Obtaining a detection result of the margin detection task according to the acquired contour distances and the preset target;

[0011] The margin distance is obtained based on the detection results.

[0012] Optionally, constructing a geometric model according to a hexagonal cross section at a wire crimping location includes:

[0013] Assume that the six vertices of the hexagonal cross section are A, B, C, D, E, and F, the six sides are AB, AF, BC, FE, CD, and ED, the three diagonals are AD, BE, and FC, the perpendicular line from vertex A to diagonal FC is hS1, the perpendicular line from vertex B to diagonal FC is hS2, the perpendicular line from vertex F to diagonal BE is hM1, the perpendicular line from vertex A to diagonal BE is hM2, the perpendicular line from vertex B to diagonal AD is hL1, the perpendicular line from vertex C to diagonal AD is hL2, and the projection of side FA on diagonal FC is S 1 , the projection of side BC on the diagonal FC is S 2 , the projection of edge FE on diagonal BE is M 1 , the projection of side AB on diagonal BE is M 2 , the projection of side AB on diagonal AD is L 1 , the projection of side CD on diagonal AD is L 2 , the distance between vertex A and vertex B is S 3 , the distance between vertex A and vertex F is M 3 , the distance between vertex B and vertex C is L 3 ;

[0014] Based on the above information: , , .

[0015] Optionally, determining a preset target for the margin detection task according to the geometric model includes: determining the preset target to satisfy both as well as , and / or, both as well as , and / or, both as well as .

[0016] The cross section of a regular hexagon is a regular geometric shape. When measuring the side distance, the contour distances in all directions of the hexagonal cross section are verified to see if they satisfy the geometric relationship of the regular hexagon. Using this geometric model, the regularity of the hexagon can be verified by simply using the relationship between the side lengths and the vertical distances, thereby simplifying the difficulty of measuring the side distance and improving the measurement accuracy of the side distance.

[0017] Optionally, the obtaining of each required contour distance in the geometric model by using a camera device and an edge detection algorithm includes:

[0018] Obtain contour images of the geometric model in the S direction, the M direction, and the L direction respectively by means of a camera device;

[0019] The contour distances in the S direction, M direction and L direction are obtained by using an edge detection algorithm according to the contour images in the S direction, M direction and L direction, wherein the contour distance in the S direction includes S 3 , S 1 , S 2 and the distance of the diagonal FC, the distance in the M direction includes M 3 、M 1 、M 2 and the distance to the diagonal BE, the distance to the L direction includes L 3 , L 1 , L 2 And the distance of diagonal AD.

[0020] By using edge detection algorithms to obtain contour distances in multiple directions, it is possible to mutually verify measurement methods and improve measurement accuracy.

[0021] Optionally, the acquiring, by means of a camera device, contour images of the geometric model in the S direction, the M direction, and the L direction, respectively, includes:

[0022] Installing camera devices in the S direction, M direction and L direction of the geometric model respectively, so that the field of view of each camera device is respectively oriented toward and perpendicular to the side AB, the side AF and the side BC;

[0023] The installation height of each camera device is adjusted so that the receptive field of each camera device exceeds the edges of the diagonal FC, the diagonal BE, and the diagonal AD, and the contour images in the S direction, the M direction, and the L direction are obtained respectively.

[0024] Optionally, obtaining a detection result of a margin detection task according to the acquired contour distances and the preset target includes:

[0025] According to the contour distances in the S direction, the M direction and the L direction, the distance values ​​of the vertical line distances hS1 and hS2 in the S direction are obtained, and / or the distance values ​​of the vertical line distances hM1 and hM2 in the M direction are obtained, and / or the distance values ​​of the vertical line distances hL1 and hL2 in the L direction are obtained;

[0026] Compare S 3 、M 3 , L 3 and hS1 and hS2, and / or, respectively, by comparing S 3 、M 3 , L 3 and the distance between hM1 and hM2, and / or, respectively, by comparing S 3 、M 3 , L 3 The distance value between hL1 and hL2 is used to obtain the comparison result;

[0027] It is determined whether the comparison result is consistent with the preset target. If so, the hexagonal cross section at the wire crimping location is a regular hexagon. If not, the hexagonal cross section at the wire crimping location is not a regular hexagon.

[0028] Optionally, the distance value between hS1 and hS2 is calculated as follows: , ;

[0029] The calculation formula of the distance value between hM1 and hM2 is: , ;

[0030] The calculation formula of the distance value of hL1 and hL2 is: , .

[0031] Optionally, obtaining the margin distance according to the detection result includes:

[0032] If the hexagonal cross section at the wire crimping location is a regular hexagon, three opposite side distances of the hexagonal cross section are obtained according to the distance values ​​of hS1, hS2, hM1, hM2, hL1 and hL2;

[0033] If the hexagonal cross-section at the wire crimping point is not a regular hexagon, each camera device is shifted to its corresponding opposite side, and the field of view direction of each camera device is rotated 180°. The contour distance and vertical distance in the other three directions except S, M and L in the geometric model are obtained by the same method, and the three opposite side distances of the hexagonal cross-section are obtained respectively by combining the distance values ​​of hS1, hS2, hM1, hM2, hL1 and hL2.

[0034] Optionally, when the hexagonal cross-section at the wire crimping location is a regular hexagon:

[0035] The average vertical distances in the S direction, the M direction, and the L direction are hS, hM, and hL, respectively. The calculation formulas for the average vertical distances in the S direction, the M direction, and the L direction are: , , ;

[0036] The distance between the sides AB and ED is 2hS. , the distance between the edges AF and CD is 2hM , the distance between the sides BC and FE is 2hL .

[0037] Optionally, when the hexagonal cross-section of the wire crimping portion is not a regular hexagon:

[0038] According to the perpendicular distances in the other three directions except the S direction, the M direction and the L direction in the geometric model, the corresponding average perpendicular distances are obtained respectively;

[0039] The average perpendicular distances in two relative directions in the geometric model are added together to obtain three opposite side distances of the hexagonal cross section.

[0040] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention uses the high accuracy and universality of image algorithms in computer vision, uses a camera to collect the contour of the hexagonal high-voltage wire crimping point with image information, and uses an edge detection algorithm to obtain the contour distance, and judges whether the hexagonal cross-section is a regular hexagon through multi-directional contour distances, so as to calculate the side distance of the high-voltage wire crimping point based on the judgment result and the contour distance. The method converts the measurement of the side distance of the crimping point into the detection and calculation of a regular hexagon through preset detection targets and comparative judgments. Compared with the previous need to design calibration objects of different sizes and materials for the crimping part, multiple calibrations are not required in subsequent complex environment detections. In the application process, it is only necessary to aim the laser at the crimping part and determine the position, that is, it can automatically perform high-precision and accurate measurements, and has a high degree of automation. It can be widely used in the crimping of transmission line tower conductors and has strong versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The figure is a flow chart of a method for detecting the margin distance at a crimping point of a high-voltage wire based on computer vision in one embodiment of the present invention;

[0042] Figure 2 The figure shows a schematic diagram of crimping and forming a connecting tube and a high-voltage conductor in an embodiment of the present invention;

[0043] Figure 3 The figure shows a schematic diagram of the geometric model of the S, M and L directions of the wire crimping in one embodiment of the present invention;

[0044] Figure 4 Shown is a schematic diagram of the geometric model of the O, P, and Q directions at the wire crimping point in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0046] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and the like are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", and the like may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0047] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0048] Example 1

[0049] like Figure 1 As shown, this embodiment provides a method for detecting the margin distance at a high-voltage wire crimping point based on computer vision, comprising:

[0050] The geometric model is constructed based on the hexagonal cross section of the wire crimp;

[0051] Determining a preset target for a margin detection task according to the geometric model;

[0052] Acquire the distance of each contour in the geometric model by using a camera device and an edge detection algorithm;

[0053] Obtaining a detection result of the margin detection task according to the acquired contour distances and the preset target;

[0054] The margin distance is obtained based on the detection results.

[0055] Example 2

[0056] Based on Example 1, this example also makes the following design.

[0057] First, if Figure 2As shown, in the production of high-voltage cable joint tubes and conductors, different crimping dies need to be selected according to the different high-voltage conductor diameters to ensure that both the joint tube and the high-voltage conductor are pressed into the shape of a regular hexagon. The edge detection algorithm in computer vision is combined to determine whether the hexagon at the crimping point is a regular hexagon, and the accurate calculation result of the opposite side distance is obtained based on the judgment result and the detected contour distance.

[0058] The above-mentioned method for detecting the margin distance at the crimping point of the high-voltage wire is divided into the following steps, as follows.

[0059] Step S1, constructing a geometric model according to the hexagonal cross section of the wire crimping location.

[0060] like Figure 3 As shown, in this embodiment, the six vertices A, B, C, D, E, and F of the hexagonal cross section are set, so as to determine that the six sides are AB, AF, BC, FE, CD, and ED, and the three diagonals are AD, BE, and FC, and then determine the relationship between the perpendicular line of each vertex and the two-way sides of the cross section, and obtain the following information: hS1 and hS2 are respectively the perpendicular lines from vertex A and vertex B to diagonal line FC, hM1 and hM2 are respectively the perpendicular lines from vertex F and vertex M to diagonal line BE, and hL1 and hL2 are respectively the perpendicular lines from vertex B and vertex C to diagonal line AD; S 3 It refers to the distance from vertex A to vertex B, that is, the length of side AB, L 3 It refers to the distance from vertex B to vertex C, that is, the length of edge BC, M 3 It refers to the distance from vertex A to vertex F, that is, the length of edge AF; S 1 is the length of the projection of side FA on the diagonal FC, S 2 is the length of the projection of side BC on diagonal FC, M 1 is the length of the projection of side FE on diagonal BE, M 2 is the length of the projection of side AB on diagonal BE, L 1 is the length of the projection of side AB on diagonal AD, L 2 is the length of the projection of side CD on diagonal AD. According to the geometric relationship in the geometric model, we can get , , .

[0061] Step S2: determining a preset target for the margin detection task according to the geometric model.

[0062] The calculation of the side distance of the hexagonal cross section is converted into detecting whether the hexagonal cross section is a regular hexagon. The ultimate goal of the side distance detection task is set to determine whether the geometric relationship in the geometric model satisfies the geometric relationship of a regular hexagon. That is, if the geometric model is measured and satisfies And satisfy / / It can be seen that the hexagonal cross section at the high-voltage wire crimping point represented by the geometric model is a regular hexagon, and the distance between the opposite sides of the hexagonal cross section can be solved based on the geometric characteristics of the regular hexagon.

[0063] Step S3: Acquire the distances of each contour in the geometric model by using a camera device and an edge detection algorithm.

[0064] A 1-megapixel industrial camera is installed in the S direction, M direction and L direction of the geometric model at the crimping point of the high-voltage wire; after the industrial camera is installed, the field of view of the camera is respectively oriented toward and perpendicular to the edge AB, the edge AF and the edge BC; then the installation height of each camera is adjusted so that the receptive field of each camera exceeds the edge of the diagonal FC, the diagonal BE and the diagonal AD respectively, and the contour images of the S direction, the M direction and the L direction are obtained respectively; finally, according to the contour images of the S direction, the M direction and the L direction, the edge detection algorithm is used to determine the best edge through image preprocessing and computer vision methods, and the contour distances of the S direction, the M direction and the L direction are obtained, wherein the contour distance of the S direction includes S 3 , S 1 , S 2 and the distance of the diagonal FC, the distance in the M direction includes M 3 、M 1 、M 2 and the distance to the diagonal BE, the distance to the L direction includes L 3 , L 1 , L 2 And the distance of diagonal AD.

[0065] Step S4: obtaining a detection result of the margin detection task according to the acquired contour distances and the preset target.

[0066] In step S4, first Figure 3 As shown, according to the contour distances in the S direction, the M direction and the L direction, the distance values ​​of the vertical line distances hS1 and hS2 in the S direction, the distance values ​​of the vertical line distances hM1 and hM2 in the M direction, and the distance values ​​of the vertical line distances hL1 and hL2 in the L direction are obtained:

[0067] , ;

[0068] , ;

[0069] , ;

[0070] Then, S 3 、M 3 , L 3 The distance values ​​of hS1 and hS2 are compared, or the distance values ​​of hM1 and hM2 are compared, or the distance values ​​of hL1 and hL2 are compared to obtain a comparison result, and it is determined whether the comparison result is consistent with a preset target. If they are consistent, the hexagonal cross-section at the high-voltage wire crimping location is a regular hexagon; if they are inconsistent, the hexagonal cross-section at the high-voltage wire crimping location is not a regular hexagon.

[0071] Step S5: derive the margin distance according to the detection result.

[0072] In step S5, the average vertical distance value corresponding to each direction is obtained according to the vertical distance values ​​in the S direction, the M direction and the L direction:

[0073] The average vertical distance in the S direction ;

[0074] The average vertical distance in the M direction ;

[0075] The average vertical distance in the L direction ;

[0076] If the hexagonal cross-section at the crimping point of the high-voltage wire is a regular hexagon, it can be seen that the opposite side distance between sides AB and ED is 2hS, which is the sum of hS1 and hS2, the opposite side distance between sides AF and CD is 2hM, which is the sum of hM1 and hM2, and the opposite side distance between sides BC and FE is 2hL, which is the sum of hL1 and hL2.

[0077] If the hexagonal cross section at the crimping point of the high-voltage wire is not a regular hexagon, the geometric model is also constructed for the other three directions of the hexagonal cross section, such as Figure 4 As shown, hO1 and hO2 are the perpendicular lines from vertex E and vertex D to diagonal FC, hQ1 and hQ2 are the perpendicular lines from vertex D and vertex C to diagonal BE, hP1 and hP2 are the perpendicular lines from vertex F and vertex E to diagonal AD; 3 It refers to the distance from vertex E to vertex D, that is, the length of edge ED, P 3 refers to the distance from vertex F to vertex E, that is, the length of edge FE, Q 3 It refers to the distance from vertex C to vertex D, that is, the length of side CD; 1 is the length of the projection of side FE on the diagonal FC, O 2 is the length of the projection of side CD on diagonal FC, Q 1 is the length of the projection of side ED on diagonal BE, Q 2is the length of the projection of side BC on diagonal BE, P 1 is the length of the projection of side AF on diagonal AD, P 2 It is the length of the projection of side ED onto the diagonal AD.

[0078] According to the geometric relationship in the geometric model, we can get , , .

[0079] Move each industrial camera to the following positions: Figure 4 The O direction, P direction and Q direction shown in the figure are respectively oriented to and perpendicular to the edge ED, the edge FE and the edge CD, and then the contour images and contour distances of the O direction, the P direction and the Q direction are obtained according to the operation process in step S3, and the vertical distances of the O direction, the P direction and the Q direction are obtained according to the operation process in S4, and the corresponding average vertical distances hO, hP and hQ are obtained according to the vertical distances of each direction:

[0080] The average perpendicular distance in the O direction ;

[0081] The average vertical distance in the P direction ;

[0082] The average vertical distance in the Q direction ;

[0083] The distance values ​​of the average perpendicular distances hO, hP and hQ in the O, P and Q directions are combined with the distance values ​​of the average perpendicular distances hS, hM and hL in the S, M and L directions to obtain three opposite side distances of the hexagonal cross-section, among which the opposite side distance between sides AB and ED is the sum of hS and hO, the opposite side distance between sides AF and CD is the sum of hM and hQ, and the opposite side distance between sides BC and FE is the sum of hL and hP.

[0084] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for detecting the margin distance of a high-voltage wire crimping point based on computer vision, characterized in that: include: The geometric model is constructed based on the hexagonal cross section of the wire crimp; Determining a preset target for a margin detection task according to the geometric model; Acquire the distance of each contour in the geometric model by using a camera device and an edge detection algorithm; Obtaining a detection result of the margin detection task according to the acquired contour distances and the preset target; The margin distance is obtained based on the detection results.

2. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 1 is characterized in that: The geometric model is constructed according to the hexagonal cross section at the wire crimping location, including: Assume that the six vertices of the hexagonal cross section are A, B, C, D, E, and F, the six sides are AB, AF, BC, FE, CD, and ED, the three diagonals are AD, BE, and FC, the perpendicular line from vertex A to diagonal FC is hS1, the perpendicular line from vertex B to diagonal FC is hS2, the perpendicular line from vertex F to diagonal BE is hM1, the perpendicular line from vertex A to diagonal BE is hM2, the perpendicular line from vertex B to diagonal AD is hL1, the perpendicular line from vertex C to diagonal The perpendicular line of line AD is hL2, the projection of side FA on diagonal line FC is S1, the projection of side BC on diagonal line FC is S2, the projection of side FE on diagonal line BE is M1, the projection of side AB on diagonal line BE is M2, the projection of side AB on diagonal line AD is L1, the projection of side CD on diagonal line AD is L2, the distance between vertex A and vertex B is S3, the distance between vertex A and vertex F is M3, and the distance between vertex B and vertex C is L3; Based on the above information: , , .

3. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 2 is characterized in that: The step of determining a preset target for the margin detection task according to the geometric model includes: determining the preset target to satisfy simultaneously the following conditions according to the geometric characteristics of the regular hexagon and the geometric model: as well as , and / or, both as well as , and / or, both as well as .

4. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 3 is characterized in that: The method of obtaining the required contour distances in the geometric model by using a camera device and an edge detection algorithm includes: Obtain contour images of the geometric model in the S direction, the M direction, and the L direction respectively by means of a camera device; According to the contour images of the S direction, M direction and L direction, the contour distances in the S direction, M direction and L direction are obtained using an edge detection algorithm, wherein the contour distance in the S direction includes S3, S1, S2 and the distance of the diagonal FC, the distance in the M direction includes M3, M1, M2 and the distance of the diagonal BE, and the distance in the L direction includes L3, L1, L2 and the distance of the diagonal AD.

5. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 4 is characterized in that: The step of obtaining contour images in the S direction, the M direction, and the L direction of the geometric model by using a camera device includes: Installing camera devices in the S direction, M direction and L direction of the geometric model respectively, so that the field of view of each camera device is respectively oriented toward and perpendicular to the side AB, the side AF and the side BC; The installation height of each camera device is adjusted so that the receptive field of each camera device exceeds the edges of the diagonal FC, the diagonal BE, and the diagonal AD, and the contour images in the S direction, the M direction, and the L direction are obtained respectively.

6. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 5 is characterized in that: The detection result of the margin detection task is obtained according to the acquired contour distances and the preset target, including: According to the contour distances in the S direction, the M direction and the L direction, the distance values ​​of the vertical line distances hS1 and hS2 in the S direction are obtained, and / or the distance values ​​of the vertical line distances hM1 and hM2 in the M direction are obtained, and / or the distance values ​​of the vertical line distances hL1 and hL2 in the L direction are obtained; Compare the distance values ​​between S3, M3, and L3 and the distance value between hS1 and hS1 respectively, and / or compare the distance values ​​between S3, M3, and L3 and the distance value between hM1 and hM2 respectively, and / or compare the distance values ​​between S3, M3, and L3 and the distance value between hL1 and hL2 respectively, to obtain a comparison result; It is determined whether the comparison result is consistent with the preset target. If so, the hexagonal cross section at the wire crimping location is a regular hexagon. If not, the hexagonal cross section at the wire crimping location is not a regular hexagon.

7. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 6 is characterized in that: The calculation formula of the distance value between hS1 and hS1 is: , ; The calculation formula of the distance value between hM1 and hM2 is: , ; The calculation formula of the distance value of hL1 and hL2 is: , .

8. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 7, characterized in that: The step of obtaining the margin distance according to the detection result includes: If the hexagonal cross section at the wire crimping location is a regular hexagon, three opposite side distances of the hexagonal cross section are obtained according to the distance values ​​of hS1, hS2, hM1, hM2, hL1 and hL2; If the hexagonal cross-section at the wire crimping point is not a regular hexagon, each camera device is shifted to its corresponding opposite side, and the field of view direction of each camera device is rotated 180°. The contour distance and vertical distance in the other three directions except S, M and L in the geometric model are obtained by the same method, and the three opposite side distances of the hexagonal cross-section are obtained respectively by combining the distance values ​​of hS1, hS2, hM1, hM2, hL1 and hL2.

9. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 8, characterized in that: When the hexagonal cross-section of the wire crimping is a regular hexagon: The average vertical distances in the S direction, the M direction, and the L direction are hS, hM, and hL, respectively. The calculation formulas for the average vertical distances in the S direction, the M direction, and the L direction are: , , ; The distance between the sides AB and ED is 2hS. , the distance between the edges AF and CD is 2hM , the distance between the sides BC and FE is 2hL .

10. The method for detecting the margin distance at the crimping point of a high-voltage wire based on computer vision according to claim 8 or 9, characterized in that: When the hexagonal cross-section of the wire crimping is not a regular hexagon: According to the perpendicular distances in the other three directions except the S direction, the M direction and the L direction in the geometric model, the corresponding average perpendicular distances are obtained respectively; The average perpendicular distances in two relative directions in the geometric model are added together to obtain three opposite side distances of the hexagonal cross section.

Images