CN109785285B - Insulator damage detection method based on ellipse characteristic fitting - Google Patents
Insulator damage detection method based on ellipse characteristic fitting Download PDFInfo
- Publication number
- CN109785285B CN109785285B CN201811511230.9A CN201811511230A CN109785285B CN 109785285 B CN109785285 B CN 109785285B CN 201811511230 A CN201811511230 A CN 201811511230A CN 109785285 B CN109785285 B CN 109785285B
- Authority
- CN
- China
- Prior art keywords
- insulator
- image
- fitting
- dimensional
- insulators
- 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
Links
Landscapes
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses an insulator breakage detection method based on elliptical feature fitting, which comprises the steps of firstly collecting an original image of an insulator, then carrying out image graying and image filtering treatment to remove interference noise of the image, then carrying out two-dimensional OTSU threshold segmentation on the image to obtain a global threshold value, obtaining an insulator region, and carrying out hole filling and pseudo-target removal on the image subjected to the two-dimensional OTSU threshold segmentation by combining morphological filtering and communication region marking; performing edge detection on the processed image to obtain an edge contour of the insulator, solving a central coordinate point and a long axis corner through optimal ellipse fitting to obtain a fitted ellipse of each insulator in the insulator sheet, and analyzing the insulator sheets above and below the insulator string to obtain optimal ellipse fitting of the whole insulator string; and finally, detecting damage by using a slope model. The invention solves the problems that the existing insulator has low segmentation efficiency and shadow and illumination in an insulator image have great influence on image segmentation.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an insulator breakage detection method based on ellipse characteristic fitting.
Background
The insulator is used as an important component of the power transmission line, plays an important role in mechanical support and electrical insulation of the lead, and the performance of the insulator directly influences the safe operation of a power grid. Because the insulator is exposed in the field for a long time and works in a strong electric field for a long time, the surface of the insulator is inevitably damaged in severe environments such as quenching and quenching. The damage of the porcelain body can reduce the insulation strength of the insulator, and even cause the insulator to burn out, break down and fracture, resulting in serious electric accidents. The key point of fault diagnosis is how to accurately segment the insulator, and common image segmentation methods include edge detection, threshold segmentation, region extraction and the like, but because the acquired insulator sub-images generally comprise other non-insulator backgrounds, the invention provides an elliptical characteristic fitting method based on the existing method to realize segmentation and extraction of the insulator, and combines a slope model to realize damage detection of the insulator.
Disclosure of Invention
The invention aims to provide an elliptical feature fitting-based insulator damage detection method, which solves the problems that the existing insulator segmentation efficiency is low and shadow and illumination in an insulator image have a large influence on image segmentation.
The technical scheme adopted by the invention is that the insulator breakage detection method based on elliptical characteristic fitting is implemented according to the following steps:
step 1, acquiring an original image of an insulator, and then performing image graying and image filtering treatment to remove interference noise of the image;
step 2, performing two-dimensional OTSU threshold segmentation on the image obtained in the step 1, obtaining a global threshold value to obtain an insulator region, and performing hole filling and false target removal on the image subjected to the two-dimensional OTSU threshold segmentation by combining morphological filtering and communication region marking;
step 3, carrying out edge detection on the image processed in the step 2 to obtain an edge contour of the insulator, and solving a central coordinate point and a long axis corner through optimal elliptic fitting;
step 4: obtaining fitting ellipses of each insulator in the insulator sheets through the step 3, and obtaining optimal ellipse fitting of the whole insulator string by analyzing the insulator sheets above and below the insulator string;
step 5: and (3) detecting the damage of the insulator region obtained in the step (4) by using a slope model.
The present invention is also characterized in that,
the step 1 is specifically implemented according to the following steps:
step 1.1, image graying treatment, wherein the graying formula is as follows:
Y＝0.299R+0.587G+0.114B (1)
wherein Y is the luminance calculated from the relationship between the R, G, B color component and the luminance signal Y in the color coding method YUV, R, G, B represents the red, green, and blue color components, respectively;
step 1.2, mean value filtering treatment:
let the pixel gray value before the image denoising process be f (x, y), g (x, y) be the gray value after the denoising filter, then:
where x=0, 1,2, … … N-1, y=0, 1,2, … … N-1, M denotes the selected M-th row, N denotes the selected N-th column, and M denotes the total number of pixels in the filtering template including the current pixel.
The step 2 is specifically implemented according to the following steps:
step 2.1, let the size of the image to be segmented be m×n, the gray value thereof be {0, M-1}, let P (a, b) be the pixel point in the image to be segmented, P (a, b) ∈m, and the neighborhood average gray value S (a, b) at the pixel point P (a, b) be:
in the formula (3), k represents a P (a, b) neighborhood search window size,
m represents the selected m-th row, n represents the selected n-th column;
step 2.2, use c ab The gray value representing the coordinates (a, b), the probability of occurrence over the entire gray value range is represented by P (a, b):
step 2.3, two sets of A and B are set in the two-dimensional gray level histogram, wherein A represents an image target domain to be segmented, B represents an image impurity to be segmented and a background domain thereof, A and B comprise respective probability formulas with the same gray level and related dispersion, and the probability formulas of A and B are shown as formulas (5) and (6):
target probability:
background probability:
wherein l 1 Representing the background area.
Step 2.4, presume: the S-axis represents the gray value of each pixel point in the corresponding two-dimensional gray histogram; the T axis represents the gray average value of each pixel neighborhood in the image, the two-dimensional segmentation threshold (S, T) divides the (S, T) plane into 4 parts, and the region I represents the target, the region II represents the background, the region III represents the boundary and the region IV represents the noise;
step 2.5, two-dimensional vectors of the average value corresponding to the two types of sets are as follows:
l-1 represents the ending point of the abscissa; p (P) ij Refers to the frequency of occurrence, μ, of the doublet (i, j) 0 Mean vector, μ representing background 1 Mean vector, w, representing object 0 (s, t) represents the target probability, w 1 (s, t) represents a background probability;
the average value of the two-dimensional vectors after combination is calculated by formulas (7) and (8) as follows:
wherein mu t A two-dimensional histogram total mean vector;
tr (σb) derived from the dispersion matrix is used as a measure function of the distance between the background and target portions in the image to be segmented:
tr (σb) represents a measure of dispersion between classes;
when tr (σb) is the maximum value, a two-dimensional vector is obtained, which is the two-dimensional threshold (s, t) for two-dimensional segmentation of the image:
tr(μ B (S，T))＝max{tr(μ B (S，T))} (11)。
the step 3 is specifically implemented according to the following steps:
step 3.1, extracting the insulator edge through a prewitt edge detection algorithm;
step 3.2, searching and extracting continuous edge points as an edge point set to be fitted, and judging whether the continuous edge points meet the boundary control condition or not; setting the perimeter of an ellipse as a boundary control condition according to the characteristics of the insulator string, wherein the maximum perimeter L of the ellipse max Elliptical minimum perimeter L min Carrying out ellipse fitting on the continuous edge point set meeting the condition;
step 3.3, setting fitting degree as a control condition for fitting the optimal ellipse again for all the ellipses obtained in the fitting step 3.2, and deleting, wherein the minimum sum of squares of distances from the boundary points of the ellipse to the fitting points is defined as the optimal fitting;
step 3.4, establishing a two-dimensional coordinate system, and solving the geometric characteristic parameters of the optimal ellipse to obtain an ellipse center coordinate o (x) 0 ,y 0 ) And a major axis rotation angle θ, the elliptic curve is expressed as follows:
Ax 2 +Bxy+Cy 2 +Dx+Ey+F＝0 (12)
equation coefficients (A, B, C, D, E, F) can be obtained by fitting ellipses, so that center coordinates and major axis corners are reversely obtained, and the equation is specifically calculated according to the following formula:
and overlapping the obtained fitting elliptical area on the original image, thereby accurately positioning the position of the insulator.
Step 4 is specifically implemented according to the following steps:
step 4.1, calculating the distance from the uppermost part to the lowermost part of the whole string of insulators, and expanding the distance by 1.5 times to be used as a fitting long axis of the whole string of insulators;
step 4.2, if the number of the insulator sheets is an odd number, calculating the diameter of the insulator in the middle sheet of the whole string of insulators; if the number of the insulators is even, obtaining the average value of the diameters of the two middle insulators to be used as the diameter of the middle insulator, and amplifying the average value by 1.5 times to be used as the short axis for fitting the whole string of insulators;
step 4.3, using the central coordinate of the middle insulator as the fitted central coordinate of the whole string of insulators;
and 4.4, the fitting corner of the whole string of insulators is an included angle formed by the central coordinate of the uppermost insulator sheet and the central coordinate of the lowermost insulator sheet.
Step 5 is specifically implemented according to the following steps:
the functional model of the insulator appearance is:
wherein ε is a positive constant;
the edge of each scale of the insulator image can be represented by a slope model shown in a formula (16), so that the slope function is used for representing the edge of the complete insulator, a slope model with the parameter epsilon=1.4 is selected for representing the appearance characteristics of most insulators, the positioned insulators are scanned row by row for different rows and columns respectively, the results are recorded, and if the slope model is not met after scanning, the insulator is indicated to be damaged.
The invention has the advantages that the insulators segmented by adopting the two-dimensional OTSU threshold segmentation and morphological processing not only reflect the gray distribution of the image, but also reflect the spatial information among the pixels of the image, and the invention has better effect compared with the common threshold image segmentation effect. And obtaining a fitting ellipse of each insulator in the insulator string through optimal ellipse fitting, analyzing the insulator sheets above and below the insulator string to obtain the optimal ellipse fitting of the whole insulator string, wherein most of the insulators are inside an elliptical area, and providing a good foundation for subsequent image insulator processing. And finally, carrying out damage detection by using a slope model, wherein the edge of each scale of the intact insulator image can be represented by using the slope model, and if the slope model is not satisfied after scanning, the insulator is damaged. Compared with other damage detection methods, the method provided by the invention is simple and feasible, has higher accuracy, can realize accurate positioning and damage detection of the insulator, and lays a foundation for safe and stable operation of the power transmission line.
Drawings
FIG. 1 is a flow chart of an insulator breakage detection method based on elliptical feature fitting;
FIG. 2 is an original image of an insulator in an insulator breakage detection method based on ellipse feature fitting according to the present invention;
fig. 3 is a two-dimensional gray level histogram of an insulator in the insulator breakage detection method based on ellipse characteristic fitting.
Detailed Description
The invention will be described in detail below with reference to the drawings and the detailed description.
After the insulator is accurately segmented, fault diagnosis needs to be carried out on the insulator, the occurrence of damage to the surface of the insulator is the fault with the highest occurrence frequency, the insulator damage can reduce the insulation strength of the insulator porcelain body, and even the insulator can be burnt, broken and broken to cause electric power accidents. Therefore, the damage detection of the insulator is a very critical step, and for the damage detection of the insulator, common damage detection methods include a direct observation method, an electric field measurement method, an image detection method and the like, wherein the image detection method has the advantages of high accuracy, safety, reliability, simplicity in implementation and the like compared with other detection methods. The invention utilizes the slope model to detect the damage, the edge of each scale of the sound insulator image can be characterized by the slope model, the positioned insulators are scanned row by row for different rows and columns respectively, and the results are recorded, and the interference information in the image is eliminated through various preprocessing of the image in the earlier stage, so that after the scanning, the damage of the insulators is proved if the slope model is not satisfied.
The invention discloses an insulator damage detection method based on elliptical characteristic fitting, which is implemented by a flow chart shown in figure 1 specifically according to the following steps:
step 1, an inspection robot is carried with a high-definition camera to collect an insulator image, as shown in fig. 2, an original insulator image is collected, then image graying and image filtering processing are carried out, interference noise of the image is removed, and the method is specifically implemented according to the following steps:
step 1.1, image graying treatment, wherein the graying formula is as follows:
Y＝0.299R+0.587G+0.114B (1)
wherein Y is the luminance calculated from the relationship between the R, G, B color component and the luminance signal Y in the color coding method YUV, R, G, B represents the red, green, and blue color components, respectively;
step 1.2, mean value filtering treatment:
let the pixel gray value before the image denoising process be f (x, y), g (x, y) be the gray value after the denoising filter, then:
wherein x=0, 1,2, … … N-1, y=0, 1,2, … … N-1, M represents the selected M-th row, N represents the selected N-th column, and M represents the total number of pixels in the filtering template including the current pixel;
step 2, performing two-dimensional OTSU threshold segmentation on the image obtained in the step 1, obtaining a global threshold value to obtain an insulator region, and performing hole filling and false target removal on the image subjected to the two-dimensional OTSU threshold segmentation by combining morphological filtering and communication region marking, wherein the method is implemented specifically according to the following steps:
step 2.1, let the size of the image to be segmented be m×n, the gray value thereof be {0, M-1}, let P (a, b) be the pixel point in the image to be segmented, P (a, b) ∈m, and the neighborhood average gray value S (a, b) at the pixel point P (a, b) be:
in the formula (3), k represents the size of a P (a, b) neighborhood search window, m represents the m-th row selected, and n represents the n-th column selected;
step 2.2, use c ab The probability that a gray value representing the coordinates (a, b) occurs over the entire gray value range is expressed as P (a, b):
step 2.3, two sets of A and B are set in the two-dimensional gray level histogram, wherein A represents an image target domain to be segmented, B represents an image impurity to be segmented and a background domain thereof, A and B comprise respective probability formulas with the same gray level and related dispersion, and the probability formulas of A and B are shown as formulas (5) and (6):
target probability:
background probability:
wherein l 1 Representing a background region;
step 2.4, fig. 3 is a schematic diagram of a corresponding two-dimensional gray histogram, assuming: the S-axis represents the gray value of each pixel point in the corresponding two-dimensional gray histogram; the T axis represents the gray average value of each pixel neighborhood in the image, the two-dimensional segmentation threshold (S, T) divides the (S, T) plane into 4 parts, and the region I represents the target, the region II represents the background, the region III represents the boundary and the region IV represents the noise;
step 2.5, two-dimensional vectors of the average value corresponding to the two types of sets are as follows:
l-1 represents the ending point of the abscissa, P ij Refers to the target probability of the ith row and the jth column, P ij Refers to the frequency of occurrence, μ, of the doublet (i, j) 0 Mean vector, μ representing background 1 Mean vector, w, representing object 0 (s, t) represents the target probability, w 1 (s, t) represents a background probability;
the average value of the two-dimensional vectors after combination is calculated by formulas (7) and (8) as follows:
wherein mu t A two-dimensional histogram total mean vector;
tr (σb) derived from the dispersion matrix is used as a measure function of the distance between the background and target portions in the image to be segmented:
where tr (σb) represents a measure of dispersion between classes;
when tr (σb) is the maximum value, a two-dimensional vector is obtained, which is the two-dimensional threshold (s, t) for two-dimensional segmentation of the image:
tr(μ B (S，T))＝max(tr(μ B (S，T))} (11)；
and step 3, carrying out edge detection on the image processed in the step 2 to obtain an edge contour of the insulator, solving a central coordinate point and a long axis corner through optimal elliptic fitting, and specifically implementing the following steps:
step 3.1, extracting the insulator edge through a prewitt edge detection algorithm;
step 3.2, searching and extracting continuous edge points as an edge point set to be fitted, and judging whether the continuous edge points meet the boundary control condition or not; setting the perimeter of an ellipse as a boundary control condition according to the characteristics of the insulator string, wherein the maximum perimeter L of the ellipse max Elliptical minimum perimeter L min Carrying out ellipse fitting on the continuous edge point set meeting the condition;
step 3.3, setting fitting degree as a control condition for fitting the optimal ellipse again for all the ellipses obtained in the fitting step 3.2, and deleting, wherein the minimum sum of squares of distances from the boundary points of the ellipse to the fitting points is defined as the optimal fitting;
step 3.4, establishing a two-dimensional coordinate system, and solving the geometric characteristic parameters of the optimal ellipse to obtain an ellipse center coordinate o (x) 0 ,y 0 ) And a major axis rotation angle θ, the elliptic curve is expressed as follows:
Ax 2 +Bxy+Cy 2 +Dx+Ey+F＝0 (12)
equation coefficients (A, B, C, D, E, F) can be obtained by fitting ellipses, so that center coordinates and major axis corners are reversely obtained, and the equation is specifically calculated according to the following formula:
overlapping the obtained fitting elliptical area on an original image, thereby accurately positioning the position of the insulator;
step 4: obtaining fitting ellipses of each insulator in the insulator sheets through the step 3, and obtaining the optimal ellipse fitting of the whole insulator string by analyzing the insulator sheets above and below the insulator string, wherein the fitting ellipses are implemented specifically according to the following steps:
step 4.1, calculating the distance from the uppermost part to the lowermost part of the whole string of insulators, and expanding the distance by 1.5 times to be used as a fitting long axis of the whole string of insulators;
step 4.2, if the number of the insulator sheets is an odd number, calculating the diameter of the insulator in the middle sheet of the whole string of insulators; if the number of the insulators is even, obtaining the average value of the diameters of the two middle insulators to be used as the diameter of the middle insulator, and amplifying the average value by 1.5 times to be used as the short axis for fitting the whole string of insulators;
step 4.3, using the central coordinate of the middle insulator as the fitted central coordinate of the whole string of insulators;
step 4.4, the fitting corner of the whole string of insulators is an included angle formed by the central coordinate of the uppermost insulator sheet and the central coordinate of the lowermost insulator sheet;
step 5: and (3) detecting damage to the insulator region obtained in the step (4) by using a slope model, wherein the method is specifically implemented according to the following steps:
the functional model of the insulator appearance is:
wherein ε is a positive constant;
the edge of each scale of the insulator image can be represented by a slope model shown in a formula (16), so that the slope function is used for representing the edge of the complete insulator, a slope model with the parameter epsilon=1.4 is selected for representing the appearance characteristics of most insulators, the positioned insulators are scanned row by row for different rows and columns respectively, the results are recorded, and if the slope model is not met after scanning, the insulator is indicated to be damaged.
The invention provides a method for segmenting an insulator by combining two-dimensional OTSU segmentation and morphology based on a one-dimensional threshold segmentation method, which is suitable for various image problems, so that the segmentation efficiency of the insulator is effectively improved, and the influence of shadows and illumination in an insulator image on image segmentation is reduced. After the insulator is primarily segmented for the first time, the insulator is generally elliptical, edge detection is carried out on the primarily segmented insulator image, elliptical characteristic fitting is carried out on the detected insulator edge, and then segmentation of the whole string of insulators is completed.
Compared with other damage detection methods, the insulator damage detection method based on elliptical feature fitting is simple and easy to implement and has higher accuracy. For good shooting effect and insulator appearance standard, the model can well detect and obtain ideal results. When the influence of dirt, rain, snow and the like exists, the model has a general detection effect. The method for improving the accuracy of the model is to avoid and reduce the occurrence of the above-mentioned non-ideal situation.
Claims (2)
1. The insulator breakage detection method based on elliptical characteristic fitting is characterized by comprising the following steps of:
step 1, acquiring an original image of an insulator, and then performing image graying and image filtering treatment to remove interference noise of the image;
the step 1 is specifically implemented according to the following steps:
step 1.1, image graying treatment, wherein the graying formula is as follows:
Y＝0.299R+0.587G+0.114B (1)
wherein Y is the luminance calculated from the relationship between the R, G, B color component and the luminance signal Y in the color coding method YUV, R, G, B represents the red, green, and blue color components, respectively;
step 1.2, mean value filtering treatment:
let the pixel gray value before the image denoising process be f (x, y), g (x, y) be the gray value after the denoising filter, then:
wherein x=0, 1,2, … … N-1, y=0, 1,2, … … N-1; m represents the selected M row, n represents the selected n column, and M represents the total number of pixels including the current pixel in the filtering template;
step 2, performing two-dimensional OTSU threshold segmentation on the image obtained in the step 1, obtaining a global threshold value to obtain an insulator region, and performing hole filling and false target removal on the image subjected to the two-dimensional OTSU threshold segmentation by combining morphological filtering and communication region marking;
the step 2 is specifically implemented according to the following steps:
step 2.1, let the size of the image to be segmented be m×n, the gray value thereof be {0, M-1}, let D (a, b) be the pixel point in the image to be segmented, D (a, b) ∈m, and the neighborhood average gray value S (a, b) at the pixel point D (a, b) be:
in the formula (3), k represents a D (a, b) neighborhood search window size, m represents an mth row selected, and n represents an nth column selected;
step 2.2, use c ab The probability that a gray value representing the coordinates (a, b) occurs over the entire gray value range is expressed as P (a, b):
step 2.3, two sets of A and B are set in the two-dimensional gray level histogram, wherein A represents an image target domain to be segmented, B represents an image impurity to be segmented and a background domain thereof, A and B comprise respective probability formulas with the same gray level and related dispersion, and the probability formulas of A and B are shown as formulas (5) and (6):
target probability:
background probability:
wherein l 1 Representing a background region;
step 2.4, presume: the S-axis represents the gray value of each pixel point in the corresponding two-dimensional gray histogram; the T axis represents the gray average value of each pixel neighborhood in the image, the two-dimensional segmentation threshold (S, T) divides the (S, T) plane into 4 parts, and the region I represents the target, the region II represents the background, the region III represents the boundary and the region IV represents the noise;
step 2.5, two-dimensional vectors of the average value corresponding to the two types of sets are as follows:
l-1 represents the ending point of the abscissa, P ij Refers to the target probability of row i, column j, and is also the frequency of occurrence of the doublet (i, j), μ 0 Mean vector, μ representing background 1 Mean vector, w, representing object 0 (s, t) represents the target probability, w 1 (s, t) represents a background probability;
the average value of the two-dimensional vectors after combination is calculated by formulas (7) and (8) as follows:
wherein mu t A two-dimensional histogram total mean vector;
tr (σb) derived from the dispersion matrix is used as a measure function of the distance between the background and target portions in the image to be segmented:
where tr (σb) represents a measure of dispersion between classes;
when tr (σb) is the maximum value, a two-dimensional vector is obtained, which is the two-dimensional threshold (s, t) for two-dimensional segmentation of the image:
tr(μ B (S,T))＝max{tr(μ B (S,T))} (11)；
step 3, carrying out edge detection on the image processed in the step 2 to obtain an edge contour of the insulator, and solving a central coordinate point and a long axis corner through optimal elliptic fitting;
the step 3 is specifically implemented according to the following steps:
step 3.1, extracting the insulator edge through a prewitt edge detection algorithm;
step 3.2, searching and extracting continuous edge points as an edge point set to be fitted, and judging whether the continuous edge points meet the boundary control condition or not; setting the perimeter of an ellipse as a boundary control condition according to the characteristics of the insulator string, wherein the maximum perimeter L of the ellipse max Elliptical minimum perimeter L min Carrying out ellipse fitting on the continuous edge point set meeting the condition;
step 3.3, setting fitting degree as a control condition for fitting the optimal ellipse again for all the ellipses obtained in the fitting step 3.2, and deleting, wherein the minimum sum of squares of distances from the boundary points of the ellipse to the fitting points is defined as the optimal fitting;
step 3.4, establishing a two-dimensional coordinate system, and solving the geometric characteristic parameters of the optimal ellipse to obtain an ellipse center coordinate o (x) 0 ,y 0 ) And a major axis rotation angle θ, the elliptic curve is expressed as follows:
Ax 2 +Bxy+Cy 2 +Dx+Ey+F＝0 (12)
equation coefficients (A, B, C, D, E, F) can be obtained by fitting ellipses, so that center coordinates and major axis corners are reversely obtained, and the equation is specifically calculated according to the following formula:
overlapping the obtained fitting elliptical area on an original image, thereby accurately positioning the position of the insulator;
step 4: obtaining fitting ellipses of each insulator in the insulator sheets through the step 3, and obtaining optimal ellipse fitting of the whole insulator string by analyzing the insulator sheets above and below the insulator string;
the step 4 is specifically implemented according to the following steps:
step 4.1, calculating the distance from the uppermost part to the lowermost part of the whole string of insulators, and expanding the distance by 1.5 times to be used as a fitting long axis of the whole string of insulators;
step 4.2, if the number of the insulator sheets is an odd number, calculating the diameter of the insulator in the middle sheet of the whole string of insulators; if the number of the insulators is even, obtaining the average value of the diameters of the two middle insulators to be used as the diameter of the middle insulator, and amplifying the average value by 1.5 times to be used as the short axis for fitting the whole string of insulators;
step 4.3, using the central coordinate of the middle insulator as the fitted central coordinate of the whole string of insulators;
step 4.4, the fitting corner of the whole string of insulators is an included angle formed by the central coordinate of the uppermost insulator sheet and the central coordinate of the lowermost insulator sheet;
step 5: and (3) detecting the damage of the insulator region obtained in the step (4) by using a slope model.
2. The method for detecting the breakage of the insulator based on elliptical characteristic fitting according to claim 1, wherein the step 5 is specifically implemented according to the following steps:
the functional model of the insulator appearance is:
wherein ε is a positive constant;
the edge of each scale of the insulator image can be represented by a slope model shown in a formula (16), so that the slope function is used for representing the edge of the complete insulator, a slope model with the parameter epsilon=1.4 is selected for representing the appearance characteristics of most insulators, the positioned insulators are scanned row by row for different rows and columns respectively, the results are recorded, and if the slope model is not met after scanning, the insulator is indicated to be damaged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811511230.9A CN109785285B (en) | 2018-12-11 | 2018-12-11 | Insulator damage detection method based on ellipse characteristic fitting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811511230.9A CN109785285B (en) | 2018-12-11 | 2018-12-11 | Insulator damage detection method based on ellipse characteristic fitting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109785285A CN109785285A (en) | 2019-05-21 |
| CN109785285B true CN109785285B (en) | 2023-08-08 |
Family
ID=66496726
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811511230.9A Active CN109785285B (en) | 2018-12-11 | 2018-12-11 | Insulator damage detection method based on ellipse characteristic fitting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109785285B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112116550B (en) * | 2019-06-20 | 2024-05-31 | 上海电机学院 | Insulator defect detection method based on multi-angle point registration |
| CN111028239A (en) * | 2019-08-10 | 2020-04-17 | 杭州屏行视界信息科技有限公司 | Ellipse accurate identification method for special body measuring clothes |
| CN110517265B (en) * | 2019-09-04 | 2022-03-01 | 北京理工大学 | Method and device for detecting surface defects of product and storage medium |
| CN110956641A (en) * | 2019-11-20 | 2020-04-03 | 南京拓控信息科技股份有限公司 | Train wheel tread image segmentation method based on chemical reaction optimization |
| CN110909751B (en) * | 2019-11-26 | 2022-09-02 | 长沙理工大学 | Visual identification method, system and medium for transformer substation insulator cleaning robot |
| CN110648330B (en) * | 2019-11-28 | 2020-03-10 | 征图新视(江苏)科技股份有限公司 | Defect detection method for camera glass |
| CN111260616A (en) * | 2020-01-13 | 2020-06-09 | 三峡大学 | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization |
| CN111612761B (en) * | 2020-05-20 | 2022-09-23 | 中南大学 | Breakage detection method based on insulating porcelain insulator |
| CN111964549B (en) * | 2020-07-08 | 2022-06-17 | 航天科工防御技术研究试验中心 | Detection method for judging defect of large-volume monolithic capacitor |
| CN112698380A (en) * | 2020-12-16 | 2021-04-23 | 南京大学 | Beam section processing method suitable for low-energy proton beam under strong background noise |
| CN113643234B (en) * | 2021-07-05 | 2024-05-17 | 华北电力大学 | Composite insulator damage detection method, terminal equipment and readable storage medium |
| CN114037724B (en) * | 2021-10-18 | 2024-04-09 | 哈尔滨工业大学 | Heterogeneous parallel computing system and method for threshold image segmentation and electronic equipment |
| CN114814991A (en) * | 2022-06-28 | 2022-07-29 | 成都信息工程大学 | Method and device for identifying spatial distribution area of meteorological element field |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105004972A (en) * | 2015-06-25 | 2015-10-28 | 华北电力大学（保定） | Porcelain insulator insulation state evaluation method based on solar-blind ultraviolet imaging image feature |
| CN105160669A (en) * | 2015-08-21 | 2015-12-16 | 马鞍山市安工大工业技术研究院有限公司 | Method for detecting and locating insulator defects in power transmission line image via a drone |
| CN106060466A (en) * | 2016-06-20 | 2016-10-26 | 西安工程大学 | Video image sequence-based insulator tracking monitor method |
| CN106595593A (en) * | 2016-11-10 | 2017-04-26 | 西安工程大学 | Method for monitoring windage deviation of suspension insulator strings based on ellipse fitting |
| JP2017103936A (en) * | 2015-12-02 | 2017-06-08 | 中国電力株式会社 | Insulator chain lifting device |
| CN108257138A (en) * | 2018-01-26 | 2018-07-06 | 西安工程大学 | A kind of porcelain insulator FRACTURE CHARACTERISTICS detection method based on edge detection |
-
2018
- 2018-12-11 CN CN201811511230.9A patent/CN109785285B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105004972A (en) * | 2015-06-25 | 2015-10-28 | 华北电力大学（保定） | Porcelain insulator insulation state evaluation method based on solar-blind ultraviolet imaging image feature |
| CN105160669A (en) * | 2015-08-21 | 2015-12-16 | 马鞍山市安工大工业技术研究院有限公司 | Method for detecting and locating insulator defects in power transmission line image via a drone |
| JP2017103936A (en) * | 2015-12-02 | 2017-06-08 | 中国電力株式会社 | Insulator chain lifting device |
| CN106060466A (en) * | 2016-06-20 | 2016-10-26 | 西安工程大学 | Video image sequence-based insulator tracking monitor method |
| CN106595593A (en) * | 2016-11-10 | 2017-04-26 | 西安工程大学 | Method for monitoring windage deviation of suspension insulator strings based on ellipse fitting |
| CN108257138A (en) * | 2018-01-26 | 2018-07-06 | 西安工程大学 | A kind of porcelain insulator FRACTURE CHARACTERISTICS detection method based on edge detection |
Non-Patent Citations (1)
| Title |
|---|
| 基于红外成像技术的零值绝缘子检测;苑利等;《高压电器》;20180216;第54卷(第02期);第97-102页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109785285A (en) | 2019-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109785285B (en) | Insulator damage detection method based on ellipse characteristic fitting | |
| CN104483326B (en) | High-voltage line defects of insulator detection method and system based on depth belief network | |
| CN111260616A (en) | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization | |
| CN113592861B (en) | Bridge crack detection method based on dynamic threshold | |
| CN111428748A (en) | Infrared image insulator recognition and detection method based on HOG characteristics and SVM | |
| CN106408025B (en) | Aerial Images insulator classifying identification method based on image procossing | |
| CN106157323B (en) | A kind of insulator division and extracting method of dynamic division threshold value and block search combination | |
| CN108257138B (en) | Porcelain insulator crack characteristic detection method based on edge detection | |
| CN109472788B (en) | Method for detecting flaw on surface of airplane rivet | |
| CN112819094A (en) | Target detection and identification method based on structural similarity measurement | |
| CN110660065B (en) | Infrared fault detection and identification algorithm | |
| CN108665468B (en) | Device and method for extracting tangent tower insulator string | |
| CN109559324A (en) | A kind of objective contour detection method in linear array images | |
| CN110335233B (en) | Highway guardrail plate defect detection system and method based on image processing technology | |
| CN113487563B (en) | EL image-based self-adaptive detection method for hidden cracks of photovoltaic module | |
| CN112308872B (en) | Image edge detection method based on multi-scale Gabor first derivative | |
| CN110288571B (en) | High-speed rail contact net insulator abnormity detection method based on image processing | |
| CN110276747B (en) | Insulator fault detection and fault rating method based on image analysis | |
| CN110276787B (en) | Conductor galloping monitoring method based on marker image detection | |
| CN111340833B (en) | Power transmission line extraction method for least square interference-free random Hough transformation | |
| CN111242923A (en) | Canny operator-based insulator crack detection method based on maximum entropy threshold optimization | |
| CN111008967B (en) | Insulator RTV coating defect identification method | |
| CN115601379A (en) | Surface crack accurate detection technology based on digital image processing | |
| CN114445331A (en) | Cable intermediate joint construction defect detection method, system and device based on image recognition | |
| CN116630321A (en) | Intelligent bridge health monitoring system based on artificial intelligence |
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 |