CN110715622B - Cable conductor sectional area measuring method - Google Patents

Cable conductor sectional area measuring method Download PDF

Info

Publication number
CN110715622B
CN110715622B CN201910833895.XA CN201910833895A CN110715622B CN 110715622 B CN110715622 B CN 110715622B CN 201910833895 A CN201910833895 A CN 201910833895A CN 110715622 B CN110715622 B CN 110715622B
Authority
CN
China
Prior art keywords
conductor
cable
pixel
image
super
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910833895.XA
Other languages
Chinese (zh)
Other versions
CN110715622A (en
Inventor
刘毅
沈尹扩
林福昌
李化
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Original Assignee
Huazhong University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology filed Critical Huazhong University of Science and Technology
Priority to CN201910833895.XA priority Critical patent/CN110715622B/en
Publication of CN110715622A publication Critical patent/CN110715622A/en
Application granted granted Critical
Publication of CN110715622B publication Critical patent/CN110715622B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/28—Measuring arrangements characterised by the use of optical techniques for measuring areas
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a method for measuring the sectional area of a cable conductor, which comprises the following steps: respectively shooting cross-section color images of the cable and the standard rod by adopting the same magnification ratio; clustering and segmenting each image by adopting a super-pixel segmentation method to obtain a plurality of super-pixels; determining conductor outlines in each image by adopting a region merging method based on a plurality of super pixels of each image; counting the number of pixel points in each conductor contour, calculating the conductor sectional area of the cable based on the conductor area in the cross section corresponding to the standard rod, and finishing the measurement. According to the invention, the color images of the cross sections of the cable and the standard rod with the same size are firstly collected, then a set of digital image processing mode for distinguishing the cable conductor and the nonconductor by using color difference information is introduced according to the difference of the gaps between the cable conductor and the color of the coating layer material, the whole processing process is convenient and fast to operate, has high accuracy, cannot damage the detected object, and has higher economic benefit and stronger practicability.

Description

Cable conductor sectional area measuring method
Technical Field
The invention belongs to the field of cable detection, and particularly relates to a method for measuring the sectional area of a cable conductor.
Background
The development of the cable industry is closely related to the major engineering construction of China, the application of the cable is wide, and the application of the cable has the following advantages: 1. the power transmission line of the power cable needs small space, the requirement on the laying path is low, and the power cable cannot occupy valuable urban space. 2. The cable laying is underground engineering, and the cable line can be finally covered by roads and buildings, so that the urban landscape cannot be interfered. 3. The same underground passage can accommodate multiple cable lines, and the load capacity of the power line can be greatly improved.
However, the cable industry is still faced with a number of problems, mainly with more rejected cables and difficulty in troubleshooting. Therefore, the cable pre-inspection has great significance for engineering construction.
An important index of the cable detection link is the cross-sectional area of the conductor, which is related to the direct current resistance and the current-carrying capacity of the cable. The traditional means mainly comprise three types: diameter measuring method, weighing method and method for measuring direct current resistance. The diameter measurement method calculates the sectional area of the conductor by measuring the diameter of a single copper wire, and the method has large error and is not suitable for cables made of compacted stranded conductors; the weighing method needs to intercept the power cable with the length of several meters, is accurate in measurement, is a method adopted by the national standard GB-T3048.2-2007, and is less in use in consideration of high cost of the power cable and serious influence on the sale and use of the power cable after cutting; the direct current resistance measurement method adopts direct current resistance to reversely push the cross section area of a conductor, the method is small in damage to a cable, but the direct current resistance of the conductor is also related to the length of the cable, some manufacturers make a false on the cross section area of the conductor and the length of the conductor at the same time, and the method can miss detection under the condition.
Disclosure of Invention
The invention provides a method for measuring the sectional area of a cable conductor, which is used for solving the technical problem that the method is low in practicability due to low precision and/or high cost in the conventional method for measuring the sectional area of the cable conductor.
The technical scheme for solving the technical problems is as follows: a method for measuring the sectional area of a cable conductor comprises the following steps:
step 1, respectively shooting cross-section color images of a cable and a standard rod by adopting the same magnification ratio;
step 2, clustering and segmenting each image by adopting a super-pixel segmentation method to obtain a plurality of super-pixels;
step 3, determining conductor outlines in each image by adopting a region merging method based on the multiple super pixels of each image;
and 4, counting the number of pixel points in each conductor contour, and calculating the conductor sectional area of the cable based on the conductor area in the cross section corresponding to the standard rod to finish measurement.
The invention has the beneficial effects that: the method comprises the steps of firstly, shooting high-definition color images of the cable and the standard rod by adopting the same magnification ratio in a direction perpendicular to the sections of the cable and the standard rod so as to ensure that the sizes of the two images are equal. Then, according to the difference of the colors of the gaps between the cable conductors and the color of the coating material, a digital image processing mode is introduced, wherein the cable conductors and the nonconductors are distinguished by using color difference information, specifically, each color image is subjected to super-pixel segmentation, and then, a region merging method is introduced to determine the boundary of the conductors and the nonconductors, so that the conductor contour determined by the boundary is clear, coherent and gapless, and the existence of noise points is greatly reduced. And finally, the conductor area of the cable can be determined by counting and calculating because the two images have the same size. The whole treatment process is convenient to operate, high in accuracy, free of damage to the detected object, high in economic benefit and high in practicability.
On the basis of the technical scheme, the invention can be further improved as follows.
Further, the method further comprises:
step 0, cutting the cable perpendicular to the central axis of the cable, and exposing the cross section of the cable; and cutting the standard rod perpendicular to the central axis of the standard rod, and exposing the cross section of the standard rod.
The invention has the further beneficial effects that: perpendicular to the central axis of the cable and the standard rod to provide more accurate cross section and further improve the accuracy of measuring the cross section area of the cable conductor.
Further, the lengths of the truncations are all less than 3 cm.
The invention has the further beneficial effects that: the method disclosed by the invention can be used for cutting off a small section perpendicular to the central axis of the cable to be detected to expose the cross section, and does not need to cut off a longer section, so that the problem that the sale and use of the power cable are seriously influenced after cutting is avoided, and therefore, the method is low in cost and high in practicability.
Further, the step of using the same amplification ratio specifically includes:
the same camera is used at a fixed object distance.
The invention has the further beneficial effects that: the same camera is adopted, and the sizes of two color images obtained by shooting can be accurately ensured to be equal under the fixed object distance, so that the two color images are mutually referred, and the area measurement accuracy is improved.
Further, the step 2 comprises:
s2.1, initializing an initial superpixel and a clustering center of each initial superpixel for each image;
s2.2, determining a calculation area of each clustering center, which is larger than the size of the corresponding initial superpixel, and calculating and updating the superpixels based on the distance from each pixel in each calculation area to the clustering center;
and S2.3, calculating the mass center of each updated super pixel, taking the mass center as a new clustering center of the super pixel, and repeatedly starting to execute the S2.2 until the iteration times are reached to obtain a plurality of super pixels.
The invention has the further beneficial effects that: according to the super-pixel segmentation method, the image is initially segmented through the distance measurement among the pixel points to obtain a plurality of super-pixels, the mass center of each super-pixel is calculated based on the super-pixels obtained through each iteration, next iteration is carried out based on the mass center, the difference of color information can be effectively utilized, the iteration times are reduced, the optimal super-pixel segmentation effect is obtained, and the segmentation reliability is improved.
Further, each initial super pixel is a square with the same size, and in S2.2, each calculation region is a square, and the side length of each calculation region is 2 times the side length of each super pixel.
Further, each pixel in each image is represented by a five-dimensional vector composed of R, G, B channel information and pixel coordinates;
the distance is the distance between the five-dimensional vectors and the centroid of each superpixel is calculated based on the five-dimensional vectors of all pixels in that superpixel.
The invention has the further beneficial effects that: the method combines RGB color channels and pixel coordinates of the color image to form a five-dimensional vector, performs distance measurement calculation through the five-dimensional vector of each pixel, fully considers the color difference between a conductor and a nonconductor and the difference of a space direction, and can greatly improve the extraction accuracy of the conductor outline.
Further, the step 3 comprises:
extracting a gray value matrix of a preset color of each super pixel, and calculating a gray value average value under the gray value matrix of the super pixel, wherein the preset color is the color of a conductor;
traversing the gray value of the preset color of each pixel in each image, if the gray value is greater than the gray threshold of the image, updating the gray value of the pixel corresponding to the gray value to be a non-zero value, otherwise, updating the gray value to be zero, wherein the gray threshold of each image is a preset multiple of the maximum mean value;
and determining the conductor outline in the image based on the gray value distribution corresponding to the preset color in the image.
The invention has the further beneficial effects that: the color of the conductor to be separated is set to be a preset color, the gray value matrix of the conductor color is extracted, the outline of the conductor can be conveniently marked off, and the method is convenient, rapid and accurate.
Further, in S4, the calculating a conductor sectional area of the cable based on the conductor region area in the cross section corresponding to the standard bar specifically includes:
calculating the occupied area of each pixel point based on the conductor region area in the cross section corresponding to the standard rod and the number of the pixel points in the conductor outline, and obtaining the conductor sectional area in the cable based on the occupied area of each pixel point and the number of the pixel points in the conductor outline corresponding to the cable.
The invention has the further beneficial effects that: the amplification ratios of the two color images are the same, the actual areas corresponding to the single pixels in the two images are the same, therefore, the area occupied by each pixel point of the conductor can be calculated through the conductor cross section area of the standard rod and the number of the pixel points of the conductor cross section obtained through counting, on the basis, the number of the pixel points of the cable conductor obtained through counting can be used for obtaining the area of the cable conductor cross section, and the method is high in accuracy, low in cost, easy to operate and high in practicability.
Further, the number of pixels of each image is more than 800 ten thousand; the ratio of the conductor profile diameter corresponding to the cable to the image shortest side length of the cable is 1/2-2/3.
Drawings
Fig. 1 is a flow chart of a method for measuring a cross-sectional area of a cable conductor according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a ratiometric imaging apparatus provided by an embodiment of the present invention;
FIG. 3 is a schematic diagram of an initialization image of a cluster center of a cross section of a cable conductor according to an embodiment of the present invention;
FIG. 4 is a schematic image of a cable conductor cross-section after super-pixel segmentation according to an embodiment of the present invention;
fig. 5 is a schematic image of a cross section of a cable conductor after area threshold merging according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Example one
A method 100 for measuring a cross-sectional area of a cable conductor, as shown in fig. 1, includes:
step 110, respectively shooting cross-section color images of the cable and the standard rod by adopting the same magnification ratio;
step 120, clustering and segmenting each image by adopting a super-pixel segmentation method to obtain a plurality of super-pixels;
step 130, determining conductor outlines in each image by adopting a region merging method based on a plurality of super pixels of each image;
and 140, counting the number of pixel points in each conductor contour, and calculating the conductor sectional area of the cable based on the conductor area in the cross section corresponding to the standard rod to finish measurement.
The method comprises the steps of firstly, shooting high-definition color images of the cable and the standard rod by adopting the same magnification ratio in a mode of being perpendicular to the sections of the cable and the standard rod so as to ensure that the two images are equal in size. Then, according to the difference of the colors of the gaps between the cable conductors and the color of the coating material, a digital image processing mode is introduced, wherein the cable conductors and the nonconductors are distinguished by using color difference information, specifically, each color image is subjected to super-pixel segmentation, and then, a region merging method is introduced to determine the boundary of the conductors and the nonconductors, so that the conductor contour determined by the boundary is clear, coherent and gapless, and the existence of noise points is greatly reduced. And finally, the conductor area of the cable can be determined by counting and calculating because the amplification ratios of the two images are the same. The whole treatment process is convenient to operate, high in accuracy, free of damage to the detected object, high in economic benefit and high in practicability.
Preferably, the method 100 further comprises:
step 000, cutting the cable perpendicular to the central axis of the cable to expose the cross section of the cable; and cutting the standard rod perpendicular to the central axis of the standard rod to expose the cross section of the standard rod.
Perpendicular to the central axis of the cable and the standard rod to provide more accurate cross section and further improve the accuracy of measuring the cross section area of the cable conductor.
Preferably, the lengths are less than 3cm each.
The method has the advantages that the small section is cut down to expose the cross section in the direction perpendicular to the central axis of the cable to be detected, the long section is not required to be cut, and the problem that the selling and the using of the power cable are seriously influenced after cutting is solved.
Preferably, the same amplification ratio is adopted as follows: the same camera is used at a fixed object distance.
The same camera is adopted, and the sizes of two color images obtained by shooting can be accurately ensured to be equal under the fixed object distance, so that the two color images are mutually referred, and the subsequent area measurement accuracy is improved.
Preferably, step 120 includes:
step 121, initializing an initial superpixel and a clustering center of each initial superpixel for each image;
step 122, determining a calculation area of each cluster center larger than the size of the corresponding initial superpixel, calculating and updating the initial superpixel based on the distance from each pixel in each calculation area to the cluster center of each pixel;
and 123, calculating the centroid of each updated super pixel, taking the centroid as a new clustering center of the super pixel, and repeating the step 122 until the iteration times are reached to obtain a plurality of super pixels.
It should be noted that, the platform for implementing the super-pixel segmentation method may be Matlab or openCV, and the specific implementation process may be:
(1) and (3) calibrating the region of interest: after obtaining the high-definition image, cutting out a small external square image containing the copper material.
(2) Assuming that an image has N pixels, and assuming that the number of divided super-pixels is k, each super-pixel includes N/k pixels, and the side length of each super-pixel is
Figure BDA0002191603760000071
At the beginning of the algorithm, a cluster center C is first assigned in the image according to the size of the superpixelk:[Rk,Gk,Bk,xk,yk]In the above formula, CkTo number cluster centers, R, G, B color components, x, y pixel coordinates, and spacing between adjacent cluster centers, i, then initialize the label for each pixel: cn:[kn＝-1,dn＝+∞]In the above formula, knRecord the number of cluster centers associated with each pixel, dnRecording the distance from the pixel to the cluster center;
(3) at each cluster center CkWithin an a x a square neighborhood of (preferably, a ═ 2l), the distance D of each pixel to the cluster center is calculated, if D is smaller than DnThen the value of D is assigned to DnAt the same time, k of the current pixel pointnAssigned a value of Ck. All pixels associated to the same cluster center form a superpixel, and the distance calculation formula is as follows:
Figure BDA0002191603760000072
in the formula (d)RGBThe calculation formula is that the color distance between pixels is
Figure BDA0002191603760000073
dxyFor the spatial distance of the pixel, the calculation formula is:
Figure BDA0002191603760000074
m is a normalization constant, and for an 8-bit map, m preferably takes the value 10.
(4) For each cluster center CkAnd solving the coordinates of the centroid of the corresponding super pixel, and replacing the previous clustering center by the pixel corresponding to the centroid. The calculation formula of the centroid coordinate is as follows:
Figure BDA0002191603760000081
in the above formula i represents the link to CkAll of the pixels of (1).
(5) And (4) repeating the steps (3) and (4) until the iteration times are achieved, and after the iteration is completed, forming a diagram of the super-pixel segmentation.
The initial segmentation of the image is achieved through the distance measurement between the pixel points, a plurality of superpixels are obtained, wherein the mass center of each superpixel is calculated based on the superpixel obtained through each iteration, the next iteration is carried out based on the mass center, the optimal superpixel segmentation is obtained through combination with the iteration times and the like, and the segmentation reliability is improved.
Preferably, each initial super pixel is a square with equal size, then in step 122, each calculation region is a square with a side length 2 times the side length of each initial super pixel.
Note that the super pixels are equal in size.
Preferably, each pixel in each image is represented by a five-dimensional vector composed of R, G, B channel information and pixel coordinates;
the distance is the distance between the five-dimensional vectors and the centroid of each superpixel is calculated based on the five-dimensional vectors of all pixels in the superpixel.
The RGB color channels and the pixel coordinates of the color image are combined to form a five-dimensional vector, distance measurement calculation is carried out through the five-dimensional vector of each pixel, the color difference between a conductor and a non-conductor and the difference of a space direction are fully considered, and the extraction accuracy of the conductor outline can be greatly improved.
Preferably, step 130 includes:
extracting a gray value matrix of a preset color of each super pixel, and calculating a gray value average value under the gray value matrix of the super pixel, wherein the preset color is the color of a conductor;
traversing the gray value of the preset color of each pixel in each image, if the gray value is greater than the gray threshold of the image, updating the gray value of the pixel corresponding to the gray value to be a non-zero value, otherwise, updating the gray value to be zero, wherein the gray threshold of each image is a preset multiple of the maximum mean value;
and determining the conductor outline in the image based on the gray value distribution corresponding to the preset color in the image.
The specific region merging algorithm may be as follows:
(1) the gray value matrix of a certain color component of the super pixel is extracted, and the gray value of the red component of the image is preferentially used by the algorithm because the copper conductor is red.
(2) Traversing each super pixel, calculating the mean value of the gray value of the color of each super pixel, wherein the mean value is the gray value of the super pixel, and the maximum mean value is recorded as hmaxSetting a threshold value of
Figure BDA0002191603760000091
(3) Traversing each super pixel again to find out the gray value higher than the gray value
Figure BDA0002191603760000092
Of the super pixel, willThe gray value is set to 1, and the gray value is found to be lower than
Figure BDA0002191603760000093
The super pixel of (2) has its gradation value set to 0.
(4) Counting the number of pixels with the gray value of 1, namely the total number of the pixels occupied by the copper conductor or the standard bar
The color of the conductor to be separated is set to be a preset color, the gray value matrix of the conductor color is extracted, the outline of the conductor can be conveniently marked off, and the method is convenient, rapid and accurate.
Preferably, in step 140, the calculating a conductor sectional area of the cable based on the conductor region area in the cross section corresponding to the standard rod specifically includes:
and calculating the occupied area of each pixel point based on the conductor region area in the cross section corresponding to the standard rod and the number of the pixel points in the conductor outline, and obtaining the conductor sectional area in the cable based on the occupied area of each pixel point and the number of the pixel points in the conductor outline corresponding to the cable.
The specific process for calculating the area of the cable copper material comprises the following steps:
(1) counting the number of pixels occupied by the standard rod to be n1Measured diameter of the standard rod is d1The actual area occupied by a single pixel is:
Figure BDA0002191603760000094
(2) recording the number of pixels occupied by the cable conductor as n2The actual area of the cable is:
S2＝n2S1
because the size of two color images is equal, then the size and the number of pixels of two images are equal, consequently through the conductor cross section area of standard stick and the pixel number of the conductor cross section that gets of checking, can calculate the area that obtains every pixel of conductor and account for, based on this, the pixel number of the cable conductor that gets of checking, can obtain the area of cable conductor cross section, and the accuracy is high, and is with low costs, easy operation, the practicality is strong.
Further, the number of pixels of each image is more than 800 ten thousand; the ratio of the corresponding conductor profile diameter of the cable to the length of the shortest side of the image of the cable is 1/2-2/3, and the standard rod is a cylinder made of red copper material. The method can greatly improve the accuracy of the sectional area test of the cable conductor.
To better illustrate the invention, the following examples are given:
(1) cutting off a small section of cable vertical to the central axis of the cable; in this example implementation, the test cable is a 10kV power cable, cut short (no more than 3cm) at one end;
(2) adopting the scaling image device, take the high definition image of new cross-section of cable and standard stick cross-section down, scaling image device is as shown in figure 2, scaling image device includes: the camera comprises a positioning bolt, a distance baffle, a camera and a light supplementing part; the bolt is used for fixing the cable and the standard rod, and the central axis is parallel to the main optical axis, so that imaging distortion is avoided; the distance baffle is used for controlling the distance between the cable and the imaging object on the section of the standard rod to be consistent all the time; the light compensating part is a 6-group 2W LED lamp tube. The optical parameters of the camera are as follows: the object distance is 70mm, the focal length is 16mm, the number of photosensitive CCD pixels is 1000 ten thousand, and the area size of a single pixel is 9.3276 multiplied by 10-5mm2。
(3) Cutting out a small circumscribed square image containing copper conductor or standard bar, initializing cluster center Ck:[Rk,Gk,Bk,xk,yk]And pixel label Cn:[kn＝-1,dn＝+∞]The number k of cluster centers is 900, as shown in fig. 3, the cluster centers are small points in the graph.
(4) Performing superpixel segmentation on the graph 3, wherein the segmentation algorithm is as follows: each cluster center CkIn the 2l × 2l square neighborhood, the distance D from each pixel to the cluster center is calculated, if D is smaller than DnThen the value of D is assigned to DnAt the same time, k of the current pixel pointnAssigned a value of Ck. All pixels associated to the same cluster center form a superpixel, and the distance calculation formula is as follows:
Figure BDA0002191603760000111
in the formula (d)RGBThe calculation formula is that the color distance between pixels is
Figure BDA0002191603760000112
dxyFor the spatial distance of the pixel, the calculation formula is:
Figure BDA0002191603760000113
m is a normalization constant, and the image is an 8-bit map in the example, so that the value of m is 8; and after traversing the clustering center, calculating the centroid of each super pixel, and replacing the previous clustering center with the centroid.
(5) Repeating the step (4) and iterating for 10 times, wherein the image after iteration is shown in FIG. 4, and each small block in the image represents each super pixel after the segmentation is completed.
(6) And (3) carrying out region threshold combination on the graph 4, wherein the combination algorithm is as follows: extracting a gray value matrix of a certain color component of the segmented image, wherein the gray value matrix of the red component of the image is preferentially used by the algorithm because the copper conductor is red; traversing each super pixel, calculating the mean value of the gray value of the color of each super pixel, wherein the mean value is the gray value of the super pixel, and the maximum value is recorded as hmaxSetting a threshold value of
Figure BDA0002191603760000114
Traversing each super pixel again to find out the gray value higher than the gray value
Figure BDA0002191603760000115
The gray value of the super-pixel is set to 1, and the gray value lower than the gray value is found
Figure BDA0002191603760000116
The super pixel of (2) has its gradation value set to 0. The combined graph is shown in FIG. 5.
(7) The number of pixels with a gray scale value of 1 (white) in the dot map 5 is the total number of pixels occupied by the copper conductors or the standard bars, which is the target in this exampleNumber of quasi-rod pixels n11315489, diameter d1Is 10 mm; the actual area for a single pixel is:
Figure BDA0002191603760000117
pixel value n of copper conductor22961474, the actual area of the copper conductor is:
S2＝n2S1＝176.8mm2
the cross-sectional area measured by a weighing method specified in the national standard GB/T3048.2-2007 is 175.3mm2The difference of the result measured by the method is 0.85 percent, so that the method has higher accuracy.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (9)

1. A method for measuring the sectional area of a cable conductor is characterized by comprising the following steps:
step 1, respectively shooting cross-section color images of a cable and a standard rod by adopting the same magnification ratio;
step 2, clustering and segmenting each image by adopting a super-pixel segmentation method to obtain a plurality of super-pixels;
step 3, determining conductor outlines in each image by adopting a region merging method based on the multiple super pixels of each image;
step 4, counting the number of pixel points in each conductor contour, and calculating the conductor sectional area of the cable based on the conductor area in the cross section corresponding to the standard rod to finish measurement;
the step 2 comprises the following steps:
s2.1, initializing an initial superpixel and a clustering center of each initial superpixel for each image;
s2.2, determining a calculation area of each clustering center, which is larger than the size of the corresponding initial superpixel, and calculating and updating the superpixels based on the distance from each pixel in each calculation area to the clustering center;
and S2.3, calculating the mass center of each updated super pixel, taking the mass center as a new clustering center of the super pixel, and repeatedly starting to execute the S2.2 until the iteration times are reached to obtain a plurality of super pixels.
2. A method of measuring a cross-sectional area of a conductor of an electrical cable according to claim 1, the method further comprising:
step 0, cutting the cable perpendicular to the central axis of the cable, and exposing the cross section of the cable; and cutting the standard rod perpendicular to the central axis of the standard rod, and exposing the cross section of the standard rod.
3. A method as claimed in claim 2, wherein the lengths of the cuts are less than 3 cm.
4. The method for measuring the sectional area of the cable conductor according to claim 1, wherein the same amplification ratio is specifically adopted: the same camera is used at a fixed object distance.
5. A method as claimed in claim 1, wherein each of the initial super pixels is a square with equal size, and in S2.2, each of the calculation regions is a square with a side length 2 times as long as that of each of the initial super pixels.
6. A cable conductor cross-sectional area measuring method according to claim 1, wherein each pixel in each image is represented by a five-dimensional vector consisting of R, G, B channel information and pixel coordinates;
the distance is the distance between the five-dimensional vectors and the centroid of each superpixel is calculated based on the five-dimensional vectors of all pixels in that superpixel.
7. A cable conductor cross-sectional area measuring method according to claim 1, wherein the step 3 comprises:
extracting a gray value matrix of a preset color of each super pixel, and calculating a gray value average value under the gray value matrix of the super pixel, wherein the preset color is the color of a conductor;
traversing the gray value of the preset color of each pixel in each image, if the gray value is greater than the gray threshold of the image, updating the gray value of the pixel corresponding to the gray value to be a non-zero value, otherwise, updating the gray value to be zero, wherein the gray threshold of each image is a preset multiple of the maximum average value, and the preset multiple is one fourth;
and determining the conductor outline in the image based on the updated gray value distribution corresponding to the preset color in the image.
8. The method for measuring a conductor sectional area of a cable according to claim 1, wherein in the step 4, the conductor sectional area of the cable is calculated based on an area of a conductor region in a cross section corresponding to the standard bar, specifically:
calculating the occupied area of each pixel point based on the conductor area in the cross section corresponding to the standard rod and the number of the pixel points in the conductor outline, and obtaining the conductor sectional area in the cable based on the occupied area of each pixel point and the number of the pixel points in the conductor outline corresponding to the cable.
9. A cable conductor cross-sectional area measuring method according to any one of claims 1 to 8, wherein the number of pixels per image is more than 800 ten thousand; the ratio of the conductor profile diameter corresponding to the cable to the image shortest side length of the cable is 1/2-2/3.
CN201910833895.XA 2019-09-04 2019-09-04 Cable conductor sectional area measuring method Active CN110715622B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910833895.XA CN110715622B (en) 2019-09-04 2019-09-04 Cable conductor sectional area measuring method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910833895.XA CN110715622B (en) 2019-09-04 2019-09-04 Cable conductor sectional area measuring method

Publications (2)

Publication Number Publication Date
CN110715622A CN110715622A (en) 2020-01-21
CN110715622B true CN110715622B (en) 2020-12-29

Family

ID=69209623

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910833895.XA Active CN110715622B (en) 2019-09-04 2019-09-04 Cable conductor sectional area measuring method

Country Status (1)

Country Link
CN (1) CN110715622B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111504232B (en) * 2020-07-01 2020-10-02 上海建工集团股份有限公司 Digital detection method for concrete cracks
CN111899198A (en) * 2020-08-06 2020-11-06 北京科技大学 Defogging method and device for marine image
CN113390465B (en) * 2021-07-02 2024-05-14 国网江西省电力有限公司电力科学研究院 Device and method for rapidly detecting quality of whole cable conductor
CN113701651A (en) * 2021-10-20 2021-11-26 国网天津市电力公司电力科学研究院 Cable insulation core size detection method, device and system
CN115435714A (en) * 2022-09-28 2022-12-06 无锡华澄线缆有限公司 Analysis system for uniform detection of cable core

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9025880B2 (en) * 2012-08-29 2015-05-05 Disney Enterprises, Inc. Visual saliency estimation for images and video
CN103353987B (en) * 2013-06-14 2015-10-28 山东大学 A kind of superpixel segmentation method based on fuzzy theory
CN104881681B (en) * 2015-05-22 2018-08-14 浙江大学 Image sequence type labeling based on mixing graph model
CN104933707B (en) * 2015-07-13 2018-06-08 福建师范大学 A kind of super-pixel reconstruct segmentation and method for reconstructing based on multiphoton confocal microscopic cell images
CN109724557A (en) * 2019-03-07 2019-05-07 武汉脉源电气有限公司 A kind of method of nondestructive measurement power cable conductor cross-sectional area

Also Published As

Publication number Publication date
CN110715622A (en) 2020-01-21

Similar Documents

Publication Publication Date Title
CN110715622B (en) Cable conductor sectional area measuring method
WO2018086299A1 (en) Image processing-based insulator defect detection method and system
CN103957771B (en) Image processing apparatus and image processing method
CN105445607B (en) A kind of electrical equipment fault detection method drawn based on thermoisopleth
CN109523528B (en) Power transmission line extraction method based on unmanned aerial vehicle binocular vision SGC algorithm
DE102013113490A1 (en) Method and system for estimating a position of a camera
WO2012111906A1 (en) Apparatus and method for generating image feature data
CN109345536A (en) A kind of image superpixel dividing method and its device
CN115046653A (en) Cross-regional stored grain temperature variation detection method based on temperature field cloud picture
CN114266893A (en) Smoke and fire hidden danger identification method and device
CN112884795A (en) Power transmission line inspection foreground and background segmentation method based on multi-feature significance fusion
CN107705330B (en) Visibility intelligent estimation method based on road camera
CN109741337B (en) Region merging watershed color remote sensing image segmentation method based on Lab color space
CN112509033B (en) Automatic iron tower inclination detection method and device based on image processing
CN108564076A (en) Visual control system in power wiring in a kind of intelligent building
AU2003233534B9 (en) Systems and methods for analyzing target contrast features in images of biological samples
CN108830834B (en) Automatic extraction method for video defect information of cable climbing robot
CN116543036A (en) Cable bending radius calculation method based on algebraic reconstruction of mask center line
CN114638805B (en) Track slab crack detection method, system and storage medium
CN113486873B (en) Transformer substation equipment inspection method and system based on big data and artificial intelligence
CN114842060A (en) Interweaving contour characteristic differential extraction method for yarn evenness visual inspection
CN106023191B (en) A kind of optics delineation character edge extraction and edge fitting method based on structure feature
US6373897B1 (en) Moving quantity detection apparatus and method
CN110211053B (en) Rapid and accurate phase matching method for three-dimensional measurement
CN112037201A (en) Directional growth structure light center extraction method based on line structure light image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant