A method and device for measuring structural dimensions of high voltage cables

Abstract

The present invention discloses a method and device for measuring the structural dimensions of a high-voltage cable. The method collects small field-of-view images of various regions, splices the small field-of-view images, obtains a complete high-voltage cable image, obtains the diameter of the high-voltage cable through edge extraction and fitting algorithms, obtains the thickness of the high-voltage cable through a matching algorithm, and obtains the number of conductors of the high-voltage cable through a region filling algorithm, thereby completing the measurement of the structural dimensions of the high-voltage cable and solving the problems of low precision and low efficiency of existing manual measurement.

Description

Technical Field

[0001] The present application relates to the field of cable measurement, and in particular to a method for measuring the structural dimensions of a high-voltage cable, and also to a device for measuring the structural dimensions of a high-voltage cable. Background Art

[0002] The measurement of high-voltage cable structure dimensions is one of the most basic inspection items for discovering quality defects in high-voltage cable manufacturing. If the structural dimension parameters of the high-voltage cable, such as the thickness of the insulation layer and the thickness of the shielding layer, do not meet the design requirements, it will bring hidden dangers to the reliability of power supply after it is put into use. At present, when conducting structural inspection tests on high-voltage cables, most of the measurements are made with the help of instruments or manual sampling, which makes the structural dimension inspection of high-voltage cables cumbersome and inefficient. Due to the large size range of different structures of high-voltage cables, the measurement methods specified in the existing standards rely heavily on the experience of the test personnel, resulting in certain difficulties in accurate measurement. Summary of the invention

[0003] The present application provides a method and device for measuring the structural dimensions of a high-voltage cable, which solve the problems of low precision and low efficiency of existing manual measurements.

[0004] The present application provides a method for measuring the structural dimensions of a high-voltage cable, comprising:

[0005] Collect small field of view images of various areas of high voltage cables;

[0006] Image stitching technology is used to stitch small field of view images to obtain a complete high-voltage cable image;

[0007] The diameter of the high-voltage cable is obtained through edge extraction and fitting algorithms; the thickness of the high-voltage cable is obtained through matching algorithms; the number of conductors of the high-voltage cable is obtained through area filling algorithms, thereby completing the measurement of the structural dimensions of the high-voltage cable.

[0008] Preferably, collecting small field of view images of various areas of the high-voltage cable includes:

[0009] Use an optical system to capture small field-of-view images of various areas of the high-voltage cable.

[0010] Preferably, the small field of view images are stitched using image stitching technology to obtain a complete high-voltage cable image, including:

[0011] Select any small field of view image as the reference image;

[0012] The SURF algorithm is used to extract the feature points of the reference image and the image to be stitched, and the feature points of the reference image and the image to be stitched are matched;

[0013] Using the RANSAC algorithm to obtain the perspective transformation matrix of the paired feature points, the image to be stitched is transformed into the coordinate system of the reference image through the perspective transformation matrix;

[0014] The image to be spliced ​​after being transformed by the perspective transformation matrix is ​​copied to the reference image, and the image to be spliced ​​and the reference image are fused to obtain a complete high-voltage cable image.

[0015] Preferably, after the step of obtaining a complete high-voltage cable image, the method further comprises:

[0016] The complete high-voltage cable image is subjected to a large law to obtain a binary complete high-voltage cable image.

[0017] Preferably, it also includes:

[0018] Use the Canny operator to extract the edge contour of the complete binary high-voltage cable image.

[0019] Preferably, the diameter of the high-voltage cable is obtained by edge extraction and fitting algorithm, including:

[0020] The least squares method circular fitting is performed on each edge contour of the high-voltage cable image in turn to obtain the diameter of each contour of the high-voltage cable.

[0021] Preferably, the thickness of the high-voltage cable is obtained by a matching algorithm, including:

[0022] For each edge contour of the high-voltage cable image, contours belonging to the same layer are found by searching for matching point pairs to obtain the thickness of each contour of the high-voltage cable.

[0023] Preferably, the number of conductors of the high-voltage cable is obtained by a region filling algorithm, including:

[0024] Calculating the total pixel area of ​​the conductor region in the high-voltage cable image;

[0025] The total pixel area of ​​the conductor region is divided by the pixel area corresponding to a single conductor to obtain the number of conductors of the high-voltage cable.

[0026] The present application also provides a device for measuring the structural dimensions of a high-voltage cable, including:

[0027] A small field of view image acquisition unit, which acquires small field of view images of various areas of the high-voltage cable;

[0028] Small field of view image stitching unit, which uses image stitching technology to stitch small field of view images to obtain a complete high-voltage cable image;

[0029] The high-voltage cable structure dimension measurement unit obtains the diameter of the high-voltage cable through edge extraction and fitting algorithms; obtains the thickness of the high-voltage cable through matching algorithms; and obtains the number of conductors of the high-voltage cable through area filling algorithms, thereby completing the measurement of the high-voltage cable structure dimension.

[0030] Preferably, the small field of view image stitching unit comprises:

[0031] The reference image selection subunit selects any small field of view image as the reference image;

[0032] The paired feature point acquisition subunit uses the SURF algorithm to extract the feature points of the reference image and the image to be stitched, and matches the feature points of the reference image and the image to be stitched;

[0033] The image conversion subunit to be stitched uses the RANSAC algorithm to obtain the perspective transformation matrix of the paired feature points, and transforms the image to be stitched into the coordinate system of the reference image through the perspective transformation matrix;

[0034] The complete high-voltage cable image acquisition subunit copies the image to be spliced ​​after being transformed by the perspective transformation matrix to the reference image, fuses the image to be spliced ​​with the reference image, and obtains a complete high-voltage cable image.

[0035] The present application provides a method and device for measuring the structural dimensions of a high-voltage cable. The method collects small field-of-view images of various regions, splices the small field-of-view images, obtains a complete high-voltage cable image, obtains the diameter of the high-voltage cable through edge extraction and fitting algorithms, and obtains the thickness of the high-voltage cable through a matching algorithm. The method also obtains the number of conductors of the high-voltage cable through a region filling algorithm, thereby completing the measurement of the structural dimensions of the high-voltage cable and solving the problems of low precision and low efficiency of existing manual measurements. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of a method for measuring the structural dimensions of a high-voltage cable provided in the present application;

[0037] Figure 2 It is a local small field image of the high-voltage cable involved in this application;

[0038] Figure 3 It is a complete high-voltage cable image obtained by stitching the small-field-of-view images of the high-voltage cable involved in this application;

[0039] Figure 4 is a complete high-voltage cable image after binarization processing involved in this application;

[0040] Figure 5 It is a schematic diagram of the outline of a certain layer of the high-voltage cable involved in this application;

[0041] Figure 6 It is a schematic diagram of a device for measuring the structural dimensions of a high-voltage cable provided in this application. DETAILED DESCRIPTION

[0042] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0043] Figure 1 This is a flow chart of a method for measuring the structural dimensions of a high-voltage cable provided in this application. Figure 1 The method provided in this application is described in detail.

[0044] Step S101, collecting small field of view images of various areas of the high voltage cable.

[0045] Use the optical system to collect small field images of each area of ​​the high-voltage cable. Figure 2 As shown, the optical system used in the present application includes a camera.

[0046] Step S102, using image stitching technology to stitch the small field of view images to obtain a complete high-voltage cable image.

[0047] From the small field of view images, select any small field of view image as the reference image, use the SURF algorithm to extract the feature points of the reference image and the image to be stitched, and match the feature points of the reference image and the image to be stitched.

[0048] The SURF algorithm uses the approximation of the Hessian matrix determinant to construct a pyramid image, thereby achieving feature extraction at multiple scales. The purpose of selecting the Hessian matrix is ​​to generate stable edge points of the image. A Hessian matrix can be calculated for each pixel in the image:

[0049]

[0050] When the discriminant of the Hessian matrix reaches a local maximum, the current point is determined to be brighter or darker than other points in the surrounding area, thereby defining the position of the feature point. The discriminant of the Hessian matrix is:

[0051]

[0052] Then count the Haar wavelets around the feature points and select the direction with the largest sum of wavelet features as the main direction to achieve rotation invariance of the feature.

[0053] After obtaining the feature points of the image, the relevant feature point matching algorithm can be used to pair the feature points.

[0054] The perspective transformation matrix of the paired feature points is obtained using the RANSAC algorithm, and the image to be stitched is transformed into the coordinate system of the reference image through the perspective transformation matrix.

[0055] Image registration is to use the paired feature points to calculate the perspective transformation matrix between the image to be stitched and the reference image. Assuming that the pixel coordinates of a feature point in the image to be stitched are (u, v, w) in the coordinate system of the image to be stitched, and the pixel coordinates of the corresponding feature point in the reference image are (x', y', w') in the coordinate system of the reference image, there is the following corresponding relationship:

[0056]

[0057] in This is the desired perspective transformation matrix. The present invention uses the RANSAC method to solve the perspective transformation matrix corresponding to the feature point pair. Finally, the image to be spliced ​​is transformed into the coordinate system of the reference image through the perspective transformation matrix.

[0058] The image to be spliced ​​after the perspective transformation matrix transformation is copied to the reference image, and the image to be spliced ​​and the reference image are merged to obtain a complete high-voltage cable image, such as Figure 3 shown.

[0059] Due to the differences in optical systems, image stitching may have discontinuous edges. A weighted fusion method can be used to synthesize the pixel values ​​of the overlapping areas of the images by taking the pixel values ​​of the reference image and the image to be stitched respectively according to certain weights and adding them together.

[0060] Value dst =a*Value src1 +(1-a)*Value src2

[0061] Where Value dst is the pixel value of the stitched image, Value src1 is the pixel value of the reference image, Value src2 is the pixel value of the image to be stitched, a and 1-a are the corresponding weights, and the specific values ​​can be allocated according to the distance between the pixel point and the boundary of the two images.

[0062] After obtaining the complete high-voltage cable image, the complete high-voltage cable image is subjected to a large law to obtain a binary complete high-voltage cable image, such as Figure 4As shown. There are only two colors of information in the binary image, usually black and white. The purpose of image binarization is to facilitate the extraction of information in the image. Binarized images can increase recognition efficiency when performing computer recognition. Then, the Canny operator is used to extract the edge contour of the complete binary high-voltage cable image.

[0063] Step S103, obtaining the diameter of the high-voltage cable through edge extraction and fitting algorithms; obtaining the thickness of the high-voltage cable through a matching algorithm; and obtaining the number of conductors of the high-voltage cable through an area filling algorithm, thereby completing the measurement of the structural dimensions of the high-voltage cable.

[0064] The least squares method circular fitting is performed on each edge contour of the high-voltage cable image in turn to obtain the diameter of each contour of the high-voltage cable.

[0065] For each edge contour of the high-voltage cable image, contours belonging to the same layer are found by searching for matching point pairs to obtain the thickness of each contour of the high-voltage cable.

[0066] Figure 5 This is a schematic diagram of the contour of a certain layer of a high-voltage cable, where the outer contour is N, the inner contour is M, and A, B, and C are three points on the inner and outer contours respectively. The specific process of finding matching point pairs is as follows:

[0067] 1) Starting from a point on N, find the "matching point" with the shortest distance on M, traverse all points on N, and record the "matching point" of each point.

[0068] 2) Starting from a point on M, find the "matching point" with the shortest distance to that point on N, traverse all points on M, and record the "matching point" of each point.

[0069] 3) Take point A on N, and assume that the "matching point" of point A on M is point B. If the "matching point" of point B on N is point A, the match is considered successful, and the matching point pair of A and B is recorded. If the "matching point" of point B on N is C, the match is considered unsuccessful.

[0070] 4) Repeat 3) for all points on N to obtain all matching point pairs. Calculate the distance between each matching point pair, and then calculate the maximum, minimum and average values ​​of all matching point pairs as the maximum, minimum and average values ​​of the output thickness.

[0071] Finally, the number of conductors of the high-voltage cable is obtained through the area filling algorithm. The total pixel area of ​​the conductor area in the high-voltage cable image is calculated and the total pixel area of ​​the conductor area is divided by the pixel area corresponding to a single conductor to obtain the number of conductors of the high-voltage cable, thereby completing the measurement of the structural size of the high-voltage cable.

[0072] Based on the same inventive concept, the present application also provides a device 600 for measuring the structural dimensions of a high-voltage cable. Figure 6 Shown include:

[0073] A small field of view image acquisition unit 610, which acquires small field of view images of various areas of the high voltage cable;

[0074] A small field of view image stitching unit 620, which stitches small field of view images using image stitching technology to obtain a complete high-voltage cable image;

[0075] The high-voltage cable structure dimension measurement unit 630 obtains the diameter of the high-voltage cable through edge extraction and fitting algorithms; obtains the thickness of the high-voltage cable through a matching algorithm; and obtains the number of conductors of the high-voltage cable through an area filling algorithm, thereby completing the measurement of the high-voltage cable structure dimension.

[0076] Preferably, the small field of view image stitching unit comprises:

[0077] The reference image selection subunit selects any small field of view image as the reference image;

[0078] The paired feature point acquisition subunit uses the SURF algorithm to extract the feature points of the reference image and the image to be stitched, and matches the feature points of the reference image and the image to be stitched;

[0079] The image conversion subunit to be stitched uses the RANSAC algorithm to obtain the perspective transformation matrix of the paired feature points, and transforms the image to be stitched into the coordinate system of the reference image through the perspective transformation matrix;

[0080] The complete high-voltage cable image acquisition subunit copies the image to be spliced ​​after being transformed by the perspective transformation matrix to the reference image, fuses the image to be spliced ​​with the reference image, and obtains a complete high-voltage cable image.

[0081] The present application provides a method and device for measuring the structural dimensions of a high-voltage cable. The method collects small field-of-view images of various regions, splices the small field-of-view images, obtains a complete high-voltage cable image, obtains the diameter of the high-voltage cable through edge extraction and fitting algorithms, and obtains the thickness of the high-voltage cable through a matching algorithm. The method also obtains the number of conductors of the high-voltage cable through a region filling algorithm, thereby completing the measurement of the structural dimensions of the high-voltage cable. The method can significantly improve the detection efficiency of the structural dimensions of the high-voltage cable, and further improve the measurement accuracy and precision of the cable structural dimension parameters, thereby solving the problems of low precision and low efficiency of existing manual measurements.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for measuring the structural dimensions of a high voltage cable, characterized in that: include: Collect small field of view images of various areas of high voltage cables; Image stitching technology is used to stitch small field of view images to obtain a complete high-voltage cable image, including: Select any small field of view image as the reference image; The SURF algorithm is used to extract the feature points of the reference image and the image to be stitched, and the feature points of the reference image and the image to be stitched are matched; Use the RANSAC algorithm to obtain the perspective transformation matrix of the matched feature points, and transform the image to be stitched into the coordinate system of the reference image through the perspective transformation matrix; The image to be spliced ​​after the perspective transformation matrix transformation is copied to the reference image, and the image to be spliced ​​and the reference image are merged to obtain a complete high-voltage cable image; The diameter of the high-voltage cable is obtained through edge extraction and fitting algorithms; the thickness of the high-voltage cable is obtained through matching algorithms; the number of conductors of the high-voltage cable is obtained through area filling algorithms, thereby completing the measurement of the structural dimensions of the high-voltage cable, including: Calculating the total pixel area of ​​the conductor region in the high-voltage cable image; The total pixel area of ​​the conductor region is divided by the pixel area corresponding to a single conductor to obtain the number of conductors of the high-voltage cable.

2. The method according to claim 1, characterized in that: Collect small field of view images of various areas of high voltage cables, including: Use an optical system to capture small field-of-view images of various areas of the high-voltage cable.

3. The method according to claim 1, characterized in that After the steps to obtain a complete high voltage cable image, it also includes: The complete high-voltage cable image is subjected to a large law to obtain a binary complete high-voltage cable image.

4. The method according to claim 3, characterized in that: Also includes: Use the Canny operator to extract the edge contour of the complete binary high-voltage cable image.

5. The method according to claim 1 or 3, characterized in that: The diameter of the high-voltage cable is obtained through edge extraction and fitting algorithms, including: The least squares method circular fitting is performed on each edge contour of the high-voltage cable image in turn to obtain the diameter of each contour of the high-voltage cable.

6. The method according to claim 1 or 3, characterized in that: The thickness of the high-voltage cable is obtained through the matching algorithm, including: For each edge contour of the high-voltage cable image, contours belonging to the same layer are found by searching for matching point pairs to obtain the thickness of each contour of the high-voltage cable.

7. A device for measuring the structural dimensions of a high voltage cable, characterized in that: include: A small field of view image acquisition unit, which acquires small field of view images of various areas of the high-voltage cable; The small field of view image stitching unit uses image stitching technology to stitch small field of view images to obtain a complete high-voltage cable image, including: The reference image selection subunit selects any small field of view image as the reference image; The paired feature point acquisition subunit uses the SURF algorithm to extract the feature points of the reference image and the image to be stitched, and matches the feature points of the reference image and the image to be stitched; The image conversion subunit to be stitched uses the RANSAC algorithm to obtain the perspective transformation matrix of the matched feature points, and transforms the image to be stitched into the coordinate system of the reference image through the perspective transformation matrix; The complete high-voltage cable image acquisition subunit copies the image to be spliced ​​after the perspective transformation matrix conversion to the reference image, and fuses the image to be spliced ​​with the reference image to obtain a complete high-voltage cable image; The high-voltage cable structure dimension measurement unit obtains the diameter of the high-voltage cable through edge extraction and fitting algorithms; obtains the thickness of the high-voltage cable through matching algorithms; obtains the number of conductors of the high-voltage cable through area filling algorithms, thereby completing the measurement of the high-voltage cable structure dimensions, including: Calculating the total pixel area of ​​the conductor region in the high-voltage cable image; The total pixel area of ​​the conductor region is divided by the pixel area corresponding to a single conductor to obtain the number of conductors of the high-voltage cable.

Images