Cable waterproof tape uniformity detection method and device based on image detection and medium

Abstract

The invention relates to the technical field of cable detection, in particular to a cable wrapping waterproof tape uniformity detection method and device based on image measurement and a medium, and the method comprises the steps: obtaining a positive visual angle image of a cable wrapping waterproof tape, and segmenting a waterproof tape region from the positive visual angle image through a trained segmentation model; identifying upper and lower boundaries of the waterproof belt, and dividing to obtain upper and lower edge point sets; geometric calibration is carried out on the segmented waterproof belt area, and the thickness of the waterproof belt is calculated based on the identified upper and lower boundaries of the waterproof belt; and uniformity evaluation is carried out on the thickness of the waterproof belt based on the stability degree of the thickness of the waterproof belt changing along the transverse direction. Compared with the prior art, the method has the advantages that the uniformity of the waterproof tape is analyzed and evaluated, and the uniformity level of the cable wrapping waterproof tape is effectively reflected, so that the mounting quality of the waterproof tape is effectively controlled, and the waterproof performance and the long-term operation reliability of a cable joint are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cable detection, and in particular to a method, device and medium for detecting uniformity of a cable waterproof belt based on image detection. Background Art

[0002] With the development of the power industry and the widespread promotion of cable applications, the performance and reliability of cables have become one of the important factors affecting the stable operation of power systems. As a key waterproof protective layer in the outer sheath of the cable, the quality of the cable waterproof tape is directly related to the overall waterproof performance and service life of the cable. In order to ensure the uniformity and consistency of the waterproof tape during the cable wrapping process, the power-on operation inspection stipulates the technical requirements for the waterproof tape wrapping: during the wrapping process, the waterproof tape should be evenly wrapped in a half-overlapping manner from 60mm from the outer sheath opening on one side to 60mm from the outer sheath opening on the other side. This specification aims to ensure the uniformity of the thickness of the waterproof tape along the length of the cable to reduce the local performance failure of the cable and the hidden dangers of long-term operation caused by uneven thickness. However, most traditional waterproof tape uniformity detection methods rely on manual detection or simple contact measurement, which is not only labor-intensive and inefficient, but also easily affected by human factors when facing complex cable structures and surface features, resulting in inaccurate detection results. Summary of the invention

[0003] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a method, device and medium for detecting uniformity of a cable waterproof tape based on image detection.

[0004] The purpose of the present invention can be achieved by the following technical solutions:

[0005] As a first aspect of the present invention, a method for detecting uniformity of a cable wrapped waterproof tape based on image measurement is provided, the steps comprising:

[0006] Obtain a front-view image of the cable-wrapped waterproof tape, and use the trained segmentation model to segment the waterproof tape area from the front-view image;

[0007] Identify the upper and lower boundaries of the waterproof belt and divide them into upper and lower edge point sets;

[0008] The segmented waterproofing zone is geometrically calibrated, and the thickness of the waterproofing zone is calculated based on the identified upper and lower boundaries of the waterproofing zone:

[0009] The uniformity of the thickness of the waterproof tape is evaluated based on the smoothness of the thickness change of the waterproof tape along the lateral direction.

[0010] As an optimal technical solution, when obtaining the front-view image of the cable wrapped waterproof tape, the camera is shot parallel to the surface of the waterproof tape; the width of the waterproof tape in the image should occupy more than 70% of the image width, and a complete section of the waterproof tape should be displayed; the upper and lower edges of the waterproof tape are located in the center of the image.

[0011] As a preferred technical solution, the specific process of identifying the upper and lower boundaries of the waterproof belt is as follows:

[0012] Gaussian filtering is used to smooth the image and remove texture noise outside the waterproof zone area in the image;

[0013] For the image after noise removal, calculate the horizontal and vertical gradients of each pixel;

[0014] By comparing the gradient amplitude of the pixel point with the two pixels before and after it in the gradient direction of the pixel point, the pixels of non-local maximum values ​​among all the pixels on the edge are suppressed, and only the pixels of local maximum values ​​are retained as edge points. If the pixel point is not greater than the gradient amplitude of the two pixels before and after it, the pixel point is suppressed to 0;

[0015] Set the gradient value threshold of two pixels, the high threshold T h and low threshold T l ; When the gradient value of the pixel point is higher than the higher threshold T h When the gradient value of the pixel point is between the high threshold T h and low threshold T l When the gradient value of the pixel point is lower than the low threshold T l When , it is classified as non-edge; strong edge points will be directly retained, while weak edge points are further determined to belong to the edge based on edge connections;

[0016] Calculate whether all weak edge points are connected to strong edge points. If so, mark the weak edge points as edge points, otherwise suppress them to 0.

[0017] After obtaining a series of edge pixels, a binary image with only edge pixels is output, where the pixel values ​​of the edge pixels are 255 and the pixel values ​​of other non-edge pixels are 0.

[0018] As an optimal technical solution, the upper and lower edge point sets are divided specifically as follows: extracting the index value of each edge pixel in the binary image of edge pixels; sorting and clustering all edge points according to their vertical coordinates, traversing the pixel indexes of all edge points in the binary image, scanning them column by column, clustering edge points with adjacent and continuous vertical coordinate index values ​​into the same set, and dividing all edge points into upper and lower edge point sets.

[0019] As a preferred technical solution, the geometric calibration of the segmented waterproof belt area is specifically as follows:

[0020] Perform a straight line fit on the upper edge point set, and calculate the offset angle θ between the fitted line and the horizontal correction line based on the slope k of the fitted line offset ;

[0021] According to the offset angle θ offset The image of the waterproof belt area is rotated and corrected so that the upper edge of the waterproof belt is parallel to the horizontal correction line.

[0022] As a preferred technical solution, the thickness of the waterproof belt is specifically calculated as follows: the upper and lower edges of the waterproof belt in the upper and lower edge point sets obtained by the division are extracted as a set of pixel points, and the corresponding upper edge pixel points and lower edge pixel points are found at each column position, and their row difference is calculated as the thickness of the waterproof belt at that position.

[0023] As a preferred technical solution, the uniformity of the waterproof tape thickness is evaluated and the following waterproof tape thickness uniformity score U is calculated:

[0024]

[0025] Where U is the thickness uniformity score, G is the normalization function, W is the width of the image, and T i represents the thickness of the waterproof tape in the i-th column of the image, Indicates the mean thickness.

[0026] As a preferred technical solution, the method sets a threshold for the obtained average score U;

[0027] When the calculated uniformity score U is less than the threshold, the thickness of the waterproofing tape meets the specifications and quality requirements;

[0028] On the contrary, it indicates that the thickness of the waterproof tape is unevenly distributed, and the wrapping process should be further optimized or the installation process should be adjusted.

[0029] As a second aspect of the present invention, a device for detecting uniformity of a cable wrapping waterproof tape based on image measurement is provided, comprising a memory, a processor, and a program stored in the memory, wherein when the processor executes the program, the method for detecting uniformity of a cable wrapping waterproof tape based on image measurement as described above is implemented.

[0030] As a third aspect of the present invention, a storage medium is provided, on which a program is stored, and when the program is executed, the method for detecting uniformity of a cable wrapping waterproof tape based on image measurement as described above is implemented.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention uses an image measurement-based uniformity detection method for wrapped waterproof tapes, which takes advantage of computer vision and image processing technology, and can accurately measure and analyze the thickness and overall uniformity of waterproof tapes. The front view image of the waterproof tape is obtained by a high-resolution camera, and the upper and lower edges of the waterproof tape are extracted by combining an image processing algorithm. The image is corrected by presetting a horizontal correction line and calculating the angle between the straight line fitted by the edge point set and the horizontal correction line. The uniformity coefficient is then calculated by the thickness distribution to evaluate whether the cable waterproof tape layout process is qualified, which can achieve accurate detection of the thickness distribution and overlap rate of the waterproof tape within the specified range. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of a method for detecting uniformity of a cable waterproof tape based on image detection according to the present invention;

[0034] Figure 2 A schematic diagram of the angle between the edge line fitting straight line and the horizontal correction line in the image correction step of the present invention;

[0035] Figure 3 Thickness detection and uniformity detection results of the cable wrapped waterproof tape of the present invention, a) RGB image, b) edge detection result, c) upper and lower edge point sets, d) thickness distribution. DETAILED DESCRIPTION

[0036] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0037] Example 1

[0038] As one of the implementation methods of the present invention, this embodiment provides a cable waterproof tape uniformity detection method based on image detection, which is specifically used for quality control during the cable joint connection and installation operation. This method is based on the technical requirements for the wrapped waterproof tape in the operation and inspection documents. Through image acquisition and processing technology, the thickness distribution of the wrapped waterproof tape is accurately detected, and its uniformity is analyzed and evaluated to determine whether the wrapping of the waterproof tape meets the specification requirements, thereby achieving effective control of the installation quality of the waterproof tape and improving the waterproof performance and long-term operation reliability of the cable joint.

[0039] The specific method of this application is shown in the attached Figure 1 , comprising steps S1-S5:

[0040] S1 Image acquisition: Use a high-resolution industrial camera to obtain the front view image data of the cable wrapped waterproof tape. The acquisition of the front view should ensure that the camera is parallel to the surface of the waterproof tape. The width of the waterproof tape in the image should occupy more than 70% of the image width, and try to display a complete section of the waterproof tape to ensure the accuracy of the measurement. At the same time, the upper and lower edges of the waterproof tape should be near the center of the image to avoid errors in the thickness calculation caused by the edge effect of the image. The collected image is preprocessed, and the waterproof tape area in the image is segmented using the trained segmentation model, and an image with only the cable joint waterproof tape is output. Since the segmentation model used is an existing technology, it will not be introduced in detail.

[0041] S2 Waterproof tape edge line detection: Since the cable wrapped waterproof tape usually has obvious boundaries, the upper and lower boundaries of the waterproof tape can be effectively identified through the edge detection algorithm. The contour pixel points of the cable wrapped joint waterproof tape are extracted through the following steps and divided into upper and lower contour pixel point sets.

[0042] 2.1 Denoising: Use Gaussian filtering to smooth the image to remove the fine texture noise outside the waterproof zone in the image and obtain a clear waterproof zone edge. Gaussian filtering convolves each pixel in the image with a Gaussian kernel, where the Gaussian kernel is represented as follows:

[0043]

[0044] Where G(Px,Py) represents the Gaussian kernel, σ represents the standard deviation, and Px and Py represent the horizontal and vertical coordinates of the image, respectively.

[0045] 2.2 Gradient calculation: In the image after noise removal, the Sobel operator is used to calculate the horizontal and vertical gradients of each pixel to determine the potential waterproof edge area in the image.

[0046] 2.3 Non-maximum suppression: This step is used to refine the edge. By comparing the gradient amplitude of the pixel point with the two pixels before and after it in the gradient direction, the pixels with non-local maximum values ​​among all the pixels on the edge are suppressed, and only the pixels with local maximum values ​​are retained as edge points. If the gradient amplitude is not greater than the two points before and after it, the point is suppressed to 0.

[0047] 2.4 Dual threshold detection: Set two thresholds: High threshold T h and low threshold T l Through double threshold detection, strong edge points will be directly retained, while weak edge points need to be further connected through edge to determine whether they belong to the edge.

[0048] 2.5 Edge connection: This step is used to eliminate isolated noise edge points and maintain the continuity and integrity of the edge of the waterproof belt. By calculating whether all weak edge points are connected to strong edge points, if so, the weak edge points are also marked as edge points, otherwise they are suppressed.

[0049] 2.6 Divide the upper and lower edge point sets: After obtaining a series of edge pixels, output a binary image with only edge pixels, where the pixel value of the edge pixels is 255, and the pixel value of other non-edge pixels is 0. Next, extract the index value of each edge pixel in the binary image. In order to distinguish the upper and lower edges of the waterproof belt, all edge points need to be sorted and clustered according to their vertical coordinates Py. Specifically, traverse the pixel indexes of all edge points in the binary image, scan them one by one by column, and cluster the edge points with adjacent and continuous Py coordinate index values ​​into the same set. Since the display of the waterproof belt in the image usually presents two obvious upper and lower edges, namely the upper and lower edge lines, all edge points can be divided into two upper and lower edge point sets.

[0050] S3 area calibration: In order to ensure the accuracy of calculating the thickness of the cable joint waterproof tape, the detection object needs to be kept horizontal and stable during the image acquisition process. However, since offset and jitter will inevitably occur during the acquisition process, it is necessary to perform geometric calibration on the segmented waterproof tape area to eliminate the tilt error and slight deformation in the image so that it can be aligned with the horizontal line, thereby improving the accuracy of subsequent thickness calculations. The specific calibration method is to preset a horizontal correction line in the image, compare the upper edge of the waterproof tape with the correction line, calculate the angle between the upper edge curve and the horizontal correction line, and then perform geometric correction on the entire image. Because in the actual process, the waterproof tape area of ​​the cable joint is affected by gravity, its edge line usually presents a certain curvature or irregular shape. Therefore, before performing the angle calculation, it is necessary to perform a straight line fitting operation on the upper edge point set to make its shape closer to the ideal straight line state, which is convenient for subsequent angle calculations. The specific steps of the straight line fitting operation are as follows:

[0051] 3.1 Calculate the mean of the x- and y-coordinates of each edge point:

[0052]

[0053]

[0054] 3.2 Calculate the slope j of the fitted straight line:

[0055]

[0056] 3.3 Calculate the intercept c:

[0057]

[0058] Through the above steps, we get the fitted straight line y=kx+c, and calculate the offset angle between the straight line and the horizontal correction line. The calculation formula is as follows:

[0059]

[0060] In the formula, θ offset represents the offset angle, and k is the slope of the straight line fitted by the upper edge point set. According to the angle θ, the image is rotated and corrected to make the upper edge of the waterproof belt parallel to the horizontal correction line. The accuracy of the correction angle must reach 0.1° to eliminate the tilt error caused by offset or jitter during the shooting process, thereby ensuring that the waterproof belt area is in an ideal horizontal state. After the image correction is completed, it can be used for subsequent thickness calculation and uniformity evaluation.

[0061] S4 Waterproof tape thickness calculation: In the image after edge detection and binarization, the upper and lower edges of the waterproof tape are extracted as a set of pixel points. Then it is necessary to find the corresponding upper edge pixel points and lower edge pixel points at each column position, and calculate their row difference as the thickness of the waterproof tape at that position. Assume that the image size is H×W (height is H, width is W), the thickness T(x) of the waterproof tape at a column x (where x ranges from 1 to W) can be calculated by the row difference of the upper and lower edge pixels, and the calculation formula is as follows:

[0062] T(x)=|y upper (x)-y lower (x)|

[0063] In the formula, T(x) represents the thickness of the waterproof tape at position x, y upper (x) represents the vertical coordinate value of the upper edge pixel at the x position, y lower (x) represents the vertical coordinate value of the lower edge pixel at the x position.

[0064] S5 Waterproof tape thickness uniformity evaluation: Design an indicator to quantify the uniformity of the waterproof tape, that is, the degree of smoothness of the change in the thickness of the waterproof tape in the horizontal direction (that is, the thickness value of each column). After obtaining the thickness information of each position of the segmented cable waterproof tape area, it is recorded as {T(1), T(2), ..., T(W)}, where W is the width of the image area. After obtaining the thickness sequence, the present invention quantifies the degree of smoothness of the change in the thickness of the waterproof tape in the horizontal direction of the image by calculating the uniformity score U of the waterproof tape thickness. The mathematical expression of the uniformity score U is as follows:

[0065]

[0066] Where U is the thickness uniformity score, G is the normalization function, W is the width of the image, and T irepresents the thickness of the waterproof tape in the i-th column of the image, Represents the mean thickness. For the obtained uniformity score U, a threshold is set. The threshold setting can be adjusted according to the specific needs of the actual application scenario to meet the detection accuracy requirements in different scenarios. In this example, the threshold of the uniformity score is set to 0.1. When the calculated uniformity score U is less than the threshold, it means that the thickness distribution of the waterproof tape is relatively uniform and meets the specifications and quality requirements; otherwise, it means that the thickness distribution of the waterproof tape is uneven, and there is a large fluctuation in thickness, which requires further optimization of the wrapping process or adjustment of the installation process.

[0067] Through the above image-based uniformity detection method of the waterproof tape wrapping, the process quality control of the uniformity of the waterproof tape wrapping can be achieved.

[0068] Example 2

[0069] As a second aspect of the present invention, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned cable wrapping waterproof tape uniformity detection method based on image measurement. In addition to the above-mentioned processor, memory and interface, any device with data processing capability in which the device in the embodiment is located may also include other hardware according to the actual function of the device with data processing capability, which will not be described in detail.

[0070] Example 3

[0071] As a third aspect of the present invention, the present application also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implements the above-mentioned cable wrapping waterproof tape uniformity detection method based on image measurement. The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), an SD card, a flash card (Flash Card), etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capability. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.

[0072] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for detecting uniformity of cable wrapping waterproof tape based on image measurement, characterized in that the steps include: Obtain a front-view image of the cable-wrapped waterproof tape, and use the trained segmentation model to segment the waterproof tape area from the front-view image; Identify the upper and lower boundaries of the waterproof belt and divide them into upper and lower edge point sets; The segmented waterproofing zone is geometrically calibrated, and the thickness of the waterproofing zone is calculated based on the identified upper and lower boundaries of the waterproofing zone: The uniformity of the thickness of the waterproof tape is evaluated based on the smoothness of the thickness change of the waterproof tape along the lateral direction.

2. The method for detecting uniformity of cable wrapping waterproof tape based on image measurement according to claim 1 is characterized in that: When obtaining the front-view image of the cable-wrapped waterproof tape, the camera is shot parallel to the surface of the waterproof tape; the width of the waterproof tape in the image should occupy more than 70% of the image width, and a complete section of the waterproof tape should be displayed; the upper and lower edges of the waterproof tape are located in the center of the image.

3. The method for detecting uniformity of cable wrapping waterproof tape based on image measurement according to claim 1 is characterized in that: The specific process of identifying the upper and lower boundaries of the waterproof tape is as follows: Gaussian filtering is used to smooth the image and remove texture noise outside the waterproof zone area in the image; For the image after noise removal, calculate the horizontal and vertical gradients of each pixel; By comparing the gradient amplitude of the pixel point with the two pixels before and after it in the gradient direction of the pixel point, the pixels of non-local maximum values ​​among all the pixels on the edge are suppressed, and only the pixels of local maximum values ​​are retained as edge points. If the pixel point is not greater than the gradient amplitude of the two pixels before and after it, the pixel point is suppressed to 0; Set the gradient value threshold of two pixels, the high threshold T h and low threshold T l ; When the gradient value of the pixel point is higher than the higher threshold T h When the gradient value of the pixel point is between the high threshold T h and low threshold T l When the gradient value of the pixel point is lower than the low threshold T l , it is classified as non-edge; strong edge points will be directly retained, while weak edge points are further determined to belong to the edge based on edge connections; Calculate whether all weak edge points are connected to strong edge points. If so, mark the weak edge points as edge points, otherwise suppress them to 0. After obtaining a series of edge pixels, a binary image with only edge pixels is output, where the pixel values ​​of the edge pixels are 255 and the pixel values ​​of other non-edge pixels are 0.

4. The method for detecting uniformity of cable wrapping waterproof tape based on image measurement according to claim 3 is characterized in that: The division of the upper and lower edge point sets is specifically as follows: extracting the index value of each edge pixel in the edge pixel binary image; sorting and clustering all edge points according to their vertical coordinates, traversing the pixel indexes of all edge points in the binary image, scanning one by one by column, clustering edge points with adjacent and continuous vertical coordinate index values ​​into the same set, and dividing all edge points into upper and lower edge point sets.

5. The method for detecting uniformity of cable wrapped waterproof tape based on image measurement according to claim 1 is characterized in that: The geometric calibration of the segmented waterproof zone area is specifically as follows: Perform straight line fitting on the upper edge point set, and calculate the offset angle θ between the fitting line and the horizontal correction line based on the slope K of the fitting line offset ; According to the offset angle θ offset The image of the waterproof belt area is rotated and corrected so that the upper edge of the waterproof belt is parallel to the horizontal correction line.

6. The method for detecting uniformity of cable wrapped waterproof tape based on image measurement according to claim 1 is characterized in that: The thickness of the waterproof belt is specifically calculated as follows: the upper and lower edges of the waterproof belt in the upper and lower edge point sets obtained by the division are extracted as a set of pixel points, and the corresponding upper edge pixel points and lower edge pixel points are found at each column position, and their row difference is calculated as the thickness of the waterproof belt at that position.

7. The method for detecting uniformity of cable wrapped waterproof tape based on image measurement according to claim 1 is characterized in that: The uniformity of the waterproof tape thickness is evaluated and the waterproof tape thickness uniformity score U is calculated as follows: Where U is the thickness uniformity score, G is the normalization function, W is the width of the image, and T i represents the thickness of the waterproof tape in the i-th column of the image, Indicates the mean thickness.

8. The method for detecting uniformity of cable wrapped waterproof tape based on image measurement according to claim 7 is characterized in that: The method sets a threshold for the obtained average score U; When the calculated uniformity score U is less than the threshold, the thickness of the waterproofing tape meets the specifications and quality requirements; On the contrary, it indicates that the thickness of the waterproof tape is unevenly distributed, and the wrapping process should be further optimized or the installation process should be adjusted.

9. A cable wrapping waterproof tape uniformity detection device based on image measurement, comprising a memory, a processor, and a program stored in the memory, characterized in that: When the processor executes the program, the method for detecting uniformity of cable wrapping waterproof tape based on image measurement as described in any one of claims 1 to 8 is implemented.

10. A storage medium having a program stored thereon, characterized in that: When the program is executed, the method for detecting uniformity of cable wrapping waterproof tape based on image measurement as described in any one of claims 1 to 8 is implemented.

Images