CN110349128B - Method for detecting icicle bridging degree between glass insulator umbrellas - Google Patents

Method for detecting icicle bridging degree between glass insulator umbrellas Download PDF

Info

Publication number
CN110349128B
CN110349128B CN201910543745.5A CN201910543745A CN110349128B CN 110349128 B CN110349128 B CN 110349128B CN 201910543745 A CN201910543745 A CN 201910543745A CN 110349128 B CN110349128 B CN 110349128B
Authority
CN
China
Prior art keywords
image
insulator
maximum
bridging
maximum value
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
CN201910543745.5A
Other languages
Chinese (zh)
Other versions
CN110349128A (en
Inventor
黄新波
章小玲
张烨
朱永灿
高玉菡
伍逸群
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.)
Xian Polytechnic University
Original Assignee
Xian Polytechnic University
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 Xian Polytechnic University filed Critical Xian Polytechnic University
Priority to CN201910543745.5A priority Critical patent/CN110349128B/en
Publication of CN110349128A publication Critical patent/CN110349128A/en
Application granted granted Critical
Publication of CN110349128B publication Critical patent/CN110349128B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/10—Image enhancement or restoration by non-spatial domain filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration by the use of local operators
    • G06T5/30—Erosion or dilatation, e.g. thinning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/0008—Industrial image inspection checking presence/absence
    • 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
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N2021/8411—Application to online plant, process monitoring
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20048—Transform domain processing
    • G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]

Abstract

The invention discloses a method for detecting the icicle bridging degree between glass insulator umbrellas, which comprises the following steps: step 1, obtaining an image I through image acquisition equipment, and then carrying out gray processing to obtain the image I1(ii) a Step 2, by comparing the image I1Processing to obtain binary image I3(ii) a Step 3, gray level image I1Otsu threshold segmentation is carried out to obtain a binary image I4For image I4Processed to obtain image I5‑I7Finally, obtaining a binary image I of the insulator umbrella surface area by threshold segmentation8(ii) a Step 4, passing the imageI3And image I8Extracting icicle area image I by difference9Processed to obtain an image I10(ii) a Step 5, carrying out area image I on the insulator string5Transversely overlapping to obtain an image I11Obtaining a threshold value Tc; and 6, dividing the bridging degree between the glass insulator umbrellas into: no bridging, hidden danger and bridging. The method solves the problem that the existing insulator umbrella icicle bridging degree has no better characterization and measurement method.

Description

Method for detecting icicle bridging degree between glass insulator umbrellas
Technical Field
The invention belongs to the technical field of power transmission line monitoring, and particularly relates to a method for detecting icicle bridging degree between glass insulator umbrellas.
Background
Icing of transmission lines poses a serious challenge to safe operation of power grids, and attracts the attention of scholars all over the world from the 60 s in the 19 th century. Mass-produced icing flashover is the main cause of the reduction of the stability of the power grid. In 2008, 58% of 500kV line trip accidents in large-area ice disasters in southern China are caused by ice flashover. The ice coating of the power transmission line is a comprehensive physical phenomenon influenced by factors such as microscopic weather, tiny terrain temperature change, humidity, air convection, circulation and wind. Natural ice coating is generally divided into soft rime, hard rime and slush according to weather conditions and ice coating characteristics. Soft rime is usually a small point attached to the surface of an object and forms a white and opaque triangle shape when facing the wind; hard rime is composed of hard transparent or opaque small blocks, and is not like soft rime in a granular structure; slush is a clear column of ice deposited on a smooth surface. The ice is usually transparent and has strong adhesion capability, ice columns are often formed among insulator umbrellas, bridging is easy to occur, and ice flashover accidents are caused, so the ice is the most serious form of all ice coating forms.
China is one of the most serious countries with ice coating problems of power transmission lines, the west and east power transmission projects, and severe insulator flashover is often caused by the ice coating problem in the south, north and middle corridors. In 2019, 1 and 3 days, extreme freezing rain weather occurs in the Emei mountain area, so that multiple 500kV power transmission lines in the country are seriously iced, and the overhead ground wires on the left side of 580-582 # sections of 500kV east sky lines are broken. In 12 months in 2018, the ice coating of 94 power transmission lines in Guangxi is continuously influenced by cold tide; in 2008, 1 to 2 months, the continuous occurrence of freezing disasters in china has affected the operation of power grids of 13 provinces, and at least 7541 transmission lines with voltage of more than 10kV are destroyed, and 859 transformer substations with voltage of more than 35kV are paralyzed. The event greatly shocks the power grid, not only influences the power supply requirements and normal life of people, but also causes huge economic loss.
And the ice coating probability is highest in the high-incidence period of the ice coating of the power transmission line from 11 months to 3 months in the next year, particularly in winter and late spring cold. Therefore, timely ice coating troubleshooting and fault early warning of the power transmission line are very important in the period. The ice coating is considered as the most serious ice coating condition and is extremely harmful to the insulator, and because the ice coating is easily formed into transparent ice columns, when the ice coating reaches a certain length, two insulators are easily bridged to cause flashover accidents. In addition, if ice coating with high density exists on the line, the ice coating is increased in a wet mode, has strong adhesion capability and is not easy to fall off the line, and then the ice coating is accumulated, and the ice coating part can generate self-excited vibration with low frequency and large amplitude under the excitation of wind. When the self-excited vibration of a wire, an insulator, a hardware fitting, a pole tower and the like is carried out for a long time, fatigue damage is caused by unbalanced impact, and the safe and stable operation of a power grid is easily damaged. Aiming at the problem that the existing icicle bridging degree between insulator umbrellas is not well characterized and measured, the invention provides an image detection method.
Disclosure of Invention
The invention aims to provide a method for detecting the bridging degree of icicles among glass insulator umbrellas.
The technical scheme adopted by the invention is that the method for detecting the icicle bridging degree between glass insulator umbrellas is specifically implemented according to the following steps:
step 1, acquiring an image I of a glass insulator through image acquisition equipment arranged on an unmanned aerial vehicle or a tower camera, and carrying out graying processing on the acquired image I of the glass insulator to obtain a grayed image I1；
Step 2, by comparing the image I1Carrying out significance detection to obtain an image I of a connecting piece for eliminating an insulator string2And obtaining a binary image I after significance detection through threshold segmentation3；
Step 3, gray level image I1Otsu threshold segmentation is carried out to obtain a binary image I4For image I4Sequentially carrying out area screening and color reversal treatment, and carrying out area screening and color reversal treatment again, wherein the area threshold values of the two times are respectively T1And T2Obtaining an insulator string segmentation image I5Dividing the complete insulator string into image I5Mapping back to the image I to obtain an RGB image I of the insulator string region6Then three-channel image decomposition is carried out, and an insulator umbrella surface area image I is obtained through color characteristic processing7Finally, obtaining a binary image I of the insulator umbrella surface area by threshold segmentation8；
Step 4, passing the image I3And image I8Extracting icicle area image I by difference9Using structural elements of linear structure to ice column region I9Carrying out corrosion treatment to obtain an ice column connected domain image I without adhesion10；
Step 5, carrying out area image I on the insulator string5Transversely overlapping to obtain an image I11Then passing through a distance threshold T3And a maximum threshold value T4Removing local maxima to determine a global maximum corresponding to the insulator long shaft, namely a row where the center position is located; in picture I5Searching two boundary pixel points of the row coordinate, and determining the column coordinate of the central point through the two boundary pixel points so as to determine the insulatorThe central point is used for obtaining a threshold value Tc of ice column bridging between the umbrella insulators by calculating the average value of Euclidean distances of the central points between the adjacent insulators;
step 6, calculating an icicle connected domain image I10The minimum external rectangle and the vertical length Jc of the icicle connected domain in the (1) take Tc/3 and Tc as threshold values, and the bridging degree between the glass insulator umbrellas is divided into: no bridging, hidden danger and bridging.
The present invention is also characterized in that,
the specific implementation steps of the step 2 are as follows:
step 2.1, to image I1And (3) carrying out significance detection to protrude the umbrella surface part and the icicle part of the insulator, removing the insulator connecting piece, and carrying out the image significance detection process as follows:
first, the grayed image I is subjected to a graying process1Performing two-dimensional discrete Fourier transform, converting the two-dimensional discrete Fourier transform from a space domain to a frequency domain, and calculating an amplitude spectrum A (f) and a phase spectrum P (f);
Figure BDA0002103360990000041
P(f)＝arctan(Im(F(I1))/Re(F(I1))) (2-2)
where Re is the real part of the function going to the frequency domain; im is the imaginary part of the function converted to the frequency domain; f represents a two-dimensional discrete Fourier transform;
then, taking logarithm of the amplitude A (f) to obtain log spectrum L (f)
L(f)＝log(A(f)) (2-3)
Since the log curve satisfies the local linearity condition, a local averaging filter h is usedn(f) Smoothing it to obtain an average spectrum, where hnIs a convolution kernel for n × n mean filtering, so the spectrum residual is the difference between the log spectrum and its mean filtering, and is calculated according to the following equation:
R(f)＝L(f)-hn(f)*L(f) (2-4)
where, denotes convolution, r (f) is the Spectral Residual spectrum, which is used as the new amplitude and the original phase spectrum p (f) to reconstruct a new frequency domain spectrum;
finally, the reconstructed frequency domain spectrum is subjected to Fourier inverse transformation and then to Gaussian fuzzy filtering,
s(x,y)＝g(x,y)*F-1[exp(R(f)+iP(f))]2 (2-5)
wherein g (x, y) is a gaussian kernel of 5 × 5,. sigma.1.5, F-1Representing an inverse fourier transform;
obtaining a salient region image I through Fourier inverse transformation and Gaussian fuzzy filtering2；
Step 2.2, threshold segmentation binarization to obtain a binary image I3。
The specific implementation steps of the step 3 are as follows:
step 3.1, to image I1Otsu threshold segmentation is carried out to obtain an image I4For image I4Sequentially carrying out area screening and color reversal treatment, and carrying out area screening and color reversal treatment again, wherein the area threshold values of the two times are respectively T1And T2When the image I4When the area of the connected domain area is larger than the area threshold value, the connected domain is reserved, and the image I is traversed4Obtaining an insulator string segmentation image I by all connected domains5；
Step 3.2, mixing I5Mapping back to image I, taking out insulator string region, i.e. image I6And image I6Decomposed into R, G, B three-channel images, respectively denoted as I6_R、I6G and I6_B；
Step 3.3, finding out through image tests that when G-R is larger than 0, the image is green, and the larger the absolute value of G-R is, the darker the color is, so that the image I comprising the insulator string region is subjected to the expression (3-1)6The pixel points are processed one by one to obtain an image I7，I7The specific calculation method of the pixel point at the (x, y) position at any position is as follows:
Figure BDA0002103360990000051
obtaining an image I through threshold segmentation8。
The specific implementation of step 4 is as follows:
step 4.1, differencing the images obtained in step 2 and step 3, namely, image I3And image I8The difference result of the pixel values of the same pixel positions of the two images is given to the ice column area image I9Corresponding image position I9(x, y), the computational expression of which is as follows:
I9(x,y)＝I3(x,y)-I8(x,y) (4-1)；
step 4.2, taking a linear structure with the 90-degree direction, namely the vertical value direction and the length of L pixels as a structural element, and carrying out ice column region image I9Carrying out corrosion treatment to obtain an icicle connected domain image I10。
In step 5, because the insulator is a circular structure, the diameter of the insulator must pass through the center of the circle, the insulator appears as an ellipse in the image, the midpoint of the longest axis of each insulator in the image is the center of the insulator, based on the midpoint, the insulator string area is transversely overlapped and the maximum value point is obtained, because of the influence of ice coating of the insulator, the overlapped curve is not smooth, a large number of local maximum value points exist, the local maximum value points are not the centers of the insulators, the existence of the local maximum value points can make the calculation value of the ice column bridging threshold Tc between the insulator umbrellas go wrong, the local maximum values must be removed, and the method for removing the local maximum values to obtain the global maximum value is as follows: by distance threshold T3Grouping the maximum value points, determining the maximum value of the maximum value in each group as the maximum value of the group, and then utilizing the threshold value T of the maximum value4Eliminating local maximum value and obtaining global maximum value and its local image I5Calculating the center point of the insulator sheet through two boundary points of the position, wherein the icicle bridging threshold Tc between the insulator umbrellas is equivalent through the Euclidean distance of the center point between the insulators;
the specific implementation of the step 5, namely the solving process of the ice column bridging threshold Tc between the insulator umbrellas is as follows:
step 5.1, because the longest axis of the insulator must include the center of the insulator, finding the position of the longest axis of the insulator and averaging two boundaries to obtain the coordinate of the center;
therefore to image I5Performing transverse superposition to obtain a transverse superposition curve I of the insulator string region11：
Figure BDA0002103360990000061
And 5.2, according to the definition of the maximum value: if to x0All points nearby, all having f (x)<f(x0) Then f (x)0) Is a maximum of the function f (x), so that the insulator string region is superimposed laterally by the curve I11The number of the superimposed pixels is a maximum value point if the number of the superimposed pixels simultaneously satisfies the following conditions:
Figure BDA0002103360990000062
obtaining a transverse superposition curve I of the insulator string region11Respectively recording the maximum value and the corresponding coordinates of the maximum value by using the arrays a and b;
step 5.3, due to the influence of ice coating, the smoothness of the superposed curve is not enough, a large number of local maximum value points exist, and the distance passes through a distance threshold T3And a maximum threshold value T4Removing local maximum values;
taking the first extreme point as the maximum point of the first maximum value group, then taking the second maximum point, and calculating the coordinate distance between the first and second maximum points and the distance threshold value T3Comparing if it is less than the distance threshold T3If not, establishing a new maximum value group and taking the second extreme point as the maximum value point of the newly established maximum value group, and completing the grouping of all the detected maximum value points by analogy;
after grouping is completed, taking the maximum value in the group as the final maximum value of the group, and respectively recording the grouping maximum value and the corresponding coordinate thereof by using the arrays ma and md;
setting a maximum threshold T4The long shaft filter is used for filtering a false insulator long shaft;
firstly, the maximum value in the array ma is taken, and then the maximum values are compared with other maximum values in the array one by one, if the difference value is less than the maximum value threshold value T4Then the insulator is considered as a long axis; if the difference is larger than the maximum threshold value T4Considered as a long axis of the dummy insulator; the finally determined long axis of the insulator is represented by an array jr;
step 5.4, taking out the line coordinate corresponding to the first long axis, and obtaining the line coordinate in the image I5Finding two boundary pixel points of the insulator, solving the mean value of the sum of the two boundary point row coordinates to be the row coordinate of the center of the insulator, locking the center position of the insulator through the row and column coordinates, and finding the centers of the rest insulators by analogy;
if the image to be detected only comprises two glass insulators, Tc is the Euclidean distance between the two central points; if the number of insulators included is more than two, Tc is the average value of the euclidean distance between every two adjacent central points.
In step 6, the bridging degree between the glass insulator umbrellas is divided into three grades: no bridging, hidden danger and bridging; the specific process of classification is as follows:
calculating icicle connected domain image I10The minimum external rectangle and the vertical length Jc of the icicle connected domain are sequentially judged:
if Jc is less than Tc/3, no bridge connection exists, and the minimum circumscribed rectangle of the icicle connected domain is framed and selected by a black dotted line in the image I;
if Tc/3 is less than or equal to Jc and less than Tc, hidden danger exists, and the minimum circumscribed rectangle of the icicle connected domain is framed and selected by a black solid line in the image I;
and if the Jc is larger than or equal to the Tc, bridging, and using a red solid line to frame and select the minimum circumscribed rectangle of the icicle connected domain in the image I.
The invention has the beneficial effects that: the method disclosed by the invention is used for identifying and detecting the insulator-umbrella icicle bridging defect caused by ice-covered ice attached to the glass insulator, judging the bridging degree by taking the center distance of the insulator as a threshold value on the basis of identifying the insulator-umbrella icicle, and identifying and early warning the insulator-umbrella icicle bridging degree so as to take remedial measures in time, solve the problem that the existing insulator-umbrella icicle bridging degree is not well characterized and measured, and have important significance in ensuring the reliable transmission of electric energy.
Drawings
FIG. 1 is a flow chart of an algorithm of a method for detecting the bridging degree of icicles between glass insulator umbrellas according to the present invention;
FIG. 2 is an original view of a glass insulator image;
FIG. 3 is a graph of the result of the segmentation of a glass insulator image in the method of the present invention;
FIG. 4 is an image of icicle connected domains in a glass insulator image in accordance with the method of the present invention;
FIG. 5 is a graph showing the lateral pixel overlap of glass insulators in the method of the present invention;
FIG. 6 is a graph showing the maximum detection result of the overlay curve in the method of the present invention;
FIG. 7 is the result of the method of the present invention after the grouping of the maxima of the overlay curves;
FIG. 8 is a result of local maxima elimination in the method of the present invention;
FIG. 9 is a graph showing the threshold Tc in the method of the present invention;
FIG. 10 is a graph showing the results of measuring the bridging degree of icicles between glass insulator umbrellas in the method of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
The invention discloses a method for detecting the icicle bridging degree between glass insulator umbrellas, which is specifically implemented according to the following steps as shown in figures 1-10:
step 1, obtaining an image I (a used sample image is shown in figure 2) of a glass insulator through image acquisition equipment arranged on an unmanned aerial vehicle or a tower camera, and performing graying processing on the obtained glass insulator image I to obtain an image I1；
Step 2, by comparing the image I1Carrying out significance detection to obtain an image I of a connecting piece for eliminating an insulator string2And obtaining a binary image I after significance detection through threshold segmentation3；
The specific implementation steps of the step 2 are as follows:
step 2.1, to image I1And (3) carrying out significance detection to protrude the umbrella surface part and the icicle part of the insulator, removing the insulator connecting piece, and carrying out the image significance detection process as follows:
firstly, the image I after the graying processing is carried out1Performing two-dimensional discrete Fourier transform, converting the two-dimensional discrete Fourier transform from a space domain to a frequency domain, and calculating an amplitude spectrum A (f) and a phase spectrum P (f);
Figure BDA0002103360990000091
P(f)＝arctan(Im(F(I1))/Re(F(I1))) (2-2)
where Re is the real part of the function going to the frequency domain; im is the imaginary part of the function converted to the frequency domain; f represents a two-dimensional discrete Fourier transform;
logarithm of amplitude A (f) to obtain log spectrum L (f)
L(f)＝log(A(f)) (2-3)
Since the log curve satisfies the local linearity condition, a local averaging filter h is usedn(f) Smoothing it to obtain an average spectrum, where hnIs a convolution kernel of n × n mean filtering, and n is 3, so the spectrum residual is the difference between the log spectrum and the mean filtering, and can be calculated according to the following equation:
R(f)＝L(f)-hn(f)*L(f) (2-4)
where denotes convolution, r (f) is the Spectral Residual spectrum, which is used as the new amplitude and the original phase spectrum p (f) to reconstruct a new frequency domain spectrum.
And carrying out Fourier inverse transformation on the reconstructed frequency domain spectrum, and then carrying out Gaussian fuzzy filtering.
s(x,y)＝g(x,y)*F-1[exp(R(f)+iP(f))]2 (2-5)
Wherein g (x, y) is a gaussian kernel of 5 × 5,. sigma.1.5, F-1Representing an inverse fourier transform;
obtaining a salient region image I through Fourier inverse transformation and Gaussian fuzzy filtering2。
Step 2.2, threshold segmentation binarization to obtain a binary image I3。
Step 3, gray level image I1Otsu threshold segmentation is carried out to obtain a binary image I4For image I4Sequentially carrying out area screening and color reversal treatment, and carrying out area screening and color reversal treatment again, wherein the area threshold values of the two times are respectively T1And T2Obtaining an insulator string segmentation image I5Dividing the complete insulator string into image I5Mapping back to the image I to obtain an RGB image I of the insulator string region6Then three-channel image decomposition is carried out, and an insulator umbrella surface area image I is obtained through color characteristic processing7Finally, obtaining a binary image I of the insulator umbrella surface area by threshold segmentation8；
The umbrella surface of the glass insulator is generally light green, so that the umbrella surface of the glass insulator can be effectively extracted through color characteristics, and the extraction method of the umbrella surface of the glass insulator is as follows:
the specific implementation steps of the step 3 are as follows:
step 3.1, to image I1Otsu threshold segmentation is carried out to obtain an image I4(ii) a For image I4Sequentially carrying out area screening and color reversal treatment, and carrying out area screening and color reversal treatment again, wherein the area threshold values of the two times are respectively T1And T2When the image I4When the area of the connected domain area is larger than the area threshold value, the connected domain is reserved, and the image I is traversed4Obtaining an insulator string segmentation image I by all connected domains5The processing result is shown in figure 3;
step 3.2, mixing I5Mapping back to image I, taking out insulator string region, i.e. image I6. And image I6Decomposed into R, G, B three-channel images, respectively denoted as I6_R、I6G and I6_B；
Step 3.3, finding out through image tests that when G-R is larger than 0, the color is green, and the larger the absolute value of G-R is, the darker the color is, so that the image I comprising the insulator string region is subjected to (3-1)6The pixel points are processed one by one to obtainThe image is I7，I7The specific calculation method of the pixel point at the (x, y) position at any position is as follows:
Figure BDA0002103360990000111
since the icicle portion is generally transparent, it will also appear as a lighter green color. The image test finds that when the set threshold value is 0.3, a better segmentation effect can be achieved, and the image I is obtained through threshold segmentation8。
Step 4, passing the image I3And image I8Extracting icicle area image I by difference9Using structural elements of linear structure to ice column region I9Carrying out corrosion treatment to obtain an ice column connected domain image I without adhesion10；
The specific steps of step 4 are as follows:
step 4.1, differencing the images obtained in step 2 and step 3, namely, image I3And image I8The result of subtraction of the same pixel position pixel values of the two images is given to the icicle area image I9Corresponding image I of9(x, y) position, the computational expression is as follows:
I9(x,y)＝I3(x,y)-I8(x,y) (4-1)
step 4.2, taking a linear structure with the 90-degree direction, namely the vertical value direction and the length of L pixels as a structural element, and carrying out ice column region image I9Carrying out corrosion treatment to obtain an icicle connected domain image I10The processing result is shown in figure 4.
Step 5, carrying out area image I on the insulator string5Transversely overlapping to obtain an image I11Then passing through a distance threshold T3And a maximum threshold value T4Eliminating local maximum values to determine a global maximum value corresponding to the insulator long shaft, namely a row where the center position of the insulator is located; in picture I5Searching two boundary pixel points of the row coordinate, determining the column coordinate of the central point through the two boundary pixel points to determine the central point of the insulator, and calculating the middle between the adjacent insulatorsObtaining a threshold Tc of ice column bridging between the insulator umbrellas by the average value of the Euclidean distance of the center points;
because the insulator is a circular structure, the diameter of the insulator must pass through the center of the circle, the insulator appears as an ellipse in the image, and the midpoint of the longest axis of each insulator in the image is the center of the insulator. According to the method, firstly, insulator string regions are transversely overlapped and maximum points are obtained, due to the influence of insulator icing, an overlapping curve is not smooth, a large number of local maximum points exist, the local maximum points are not the centers of the insulators, the calculated value of the ice column bridging threshold Tc between the insulator umbrellas can be wrong due to the existence of the local maximum points, and therefore the local maximum values are required to be removed. The method for eliminating the local maximum value to obtain the global maximum value comprises the following steps: by distance threshold T3Grouping the maximum value points, determining the maximum value of the maximum value in each group as the maximum value of the group, and then utilizing the threshold value T of the maximum value4Eliminating local maximum value and obtaining global maximum value and its local image I5And calculating the center point of the insulator sheet through two boundary points of the position. The ice column bridging threshold Tc between the insulator umbrellas is equivalent through the Euclidean distance of the center points between the insulators;
the specific implementation of the step 5, namely the solving process of the ice column bridging threshold Tc between the insulator umbrellas is as follows:
step 5.1, because the longest axis of the insulator must include the center of the insulator, finding the position of the longest axis of the insulator and averaging two boundaries to obtain the coordinate of the center;
therefore to image I5Performing transverse superposition to obtain a transverse superposition curve I of the insulator string region11：
Figure BDA0002103360990000121
And obtaining a transverse superposition curve of the insulator string region, as shown in the figure 5.
And 5.2, according to the definition of the maximum value: if to x0All points nearby, all having f (x)<f(x0) Then f (x)0) Is a maximum of the function f (x)Value, so the insulator chain region transverse superposition curve I11The number of the superimposed pixels is a maximum value point if the number of the superimposed pixels simultaneously satisfies the following conditions:
Figure BDA0002103360990000122
obtaining a transverse superposition curve I of the insulator string region11Respectively recording the corresponding coordinates of the maximum value and the maximum value by using the arrays a and b, and transversely superposing a curve I in the insulator string region11The detected maximum point is noted above, as in fig. 6.
Step 5.3, due to the influence of ice coating, the smoothness of the superposed curve is not enough, a large number of local maximum value points exist, and the distance passes through a distance threshold T3And a maximum threshold value T4Removing local maximum values;
taking the first extreme point as the maximum point of the first maximum value group, then taking the second maximum point, and calculating the coordinate distance between the first and second maximum points and the distance threshold value T3Comparing if it is less than the distance threshold T3And if not, establishing a new maximum value group and taking the second extreme point as the maximum value point of the newly established maximum value group, and completing the grouping of all the detected maximum value points by analogy.
After grouping is finished, taking the maximum value in the group as the final maximum value of the group, superposing the results of the grouping of the maximum values of the map, respectively recording the grouping maximum values and the corresponding coordinates thereof by using the arrays ma and md, and transversely superposing a curve I in the insulator string region11The maximum points of each group are marked as in fig. 7.
Because the insulators are all produced according to the unified standard, the diameter numerical values of the insulators are not greatly different, namely, the numerical value of the longest axis of the insulator in the image, which corresponds to the global maximum, is also not greatly different. Setting a maximum threshold value T in consideration of the deviation of the major axis value in the image possibly caused by the ice-coated insulator4The long axis filter is used for filtering the false insulator long axis.
Firstly, the maximum value in the array ma is taken, and then the maximum value and the maximum value in the array ma are gradually addedIf the difference is less than the maximum threshold T4Then the insulator is considered as a long axis; if the difference is larger than the maximum threshold value T4Considered as a long axis of the dummy insulator; the finally determined long axis of the insulator is represented by an array jr, and a curve I is transversely superposed in the insulator string region11The detected global maximum point is marked, namely the line where the long axis of the insulator is located is as shown in the attached figure 8.
Step 5.4, taking out the line coordinate corresponding to the first long axis, and obtaining the line coordinate in the image I5Finding two boundary pixel points of the insulator, solving the mean value of the sum of the two boundary point row coordinates to be the row coordinate of the center of the insulator, locking the center position of the insulator through the row and column coordinates, and finding the centers of the rest insulators by analogy;
if the image to be detected only comprises two glass insulators, Tc is the Euclidean distance between the two central points; if the number of insulators included is more than two, Tc is an average value of euclidean distances between every two adjacent central points, and Tc determined by the sample image is shown in fig. 9.
Step 6, "state grid company substation operation and maintenance management rules (trial) 4 th minute manual isolating switch operation and maintenance detailed rule, 2.4.2 fifth tour in abnormal weather: when the ice weather is carried out, the ice thickness and the ice bridging degree of the external insulation are observed, the ice thickness is not more than 10mm, the ice bridging length is not more than 1/3 of the dry arc distance, the creepage is not more than that of the second umbrella skirt, and the creepage phenomenon of the middle umbrella skirt is avoided. ". The dry arc distance is also called as the arc distance, is the shortest distance between the external air of the insulator between two metal accessories which normally have operating voltage, and is equivalent to Tc in the image detection process;
the degree of bridging between the glass insulator ribs is classified into three levels according to the above regulations: no bridging, hidden danger and bridging; the specific process of classification is as follows:
calculating icicle connected domain image I10The minimum external rectangle and the vertical length Jc of the icicle connected domain are sequentially judged:
if Jc is less than Tc/3, no bridge connection exists, and the minimum circumscribed rectangle of the icicle connected domain is framed and selected by a black dotted line in the image I;
if Tc/3 is less than or equal to Jc and less than Tc, hidden danger exists, and the minimum circumscribed rectangle of the icicle connected domain is framed and selected by a black solid line in the image I;
if Jc is larger than or equal to Tc, bridging, and using a red solid line to frame and select the minimum circumscribed rectangle of the icicle connected domain in the image I;
the detection result corresponding to the sample image is shown in fig. 10.

Claims (6)

1. A method for detecting the bridging degree of icicles among glass insulator umbrellas is characterized by comprising the following steps:
step 1, acquiring an image I of a glass insulator through image acquisition equipment arranged on an unmanned aerial vehicle or a tower camera, and carrying out graying processing on the acquired image I of the glass insulator to obtain a grayed image I1；
Step 2, by comparing the image I1Carrying out significance detection to obtain an image I of a connecting piece for eliminating an insulator string2And obtaining a binary image I after significance detection through threshold segmentation3；
Step 3, gray level image I1Otsu threshold segmentation is carried out to obtain a binary image I4For image I4Sequentially carrying out area screening and color reversal treatment, and carrying out area screening and color reversal treatment again, wherein the area threshold values of the two times are respectively T1And T2Obtaining an insulator string segmentation image I5Dividing the complete insulator string into image I5Mapping back to the image I to obtain an RGB image I of the insulator string region6Then three-channel image decomposition is carried out, and an insulator umbrella surface area image I is obtained through color characteristic processing7Finally, obtaining a binary image I of the insulator umbrella surface area by threshold segmentation8；
Step 4, passing the image I3And image I8Extracting icicle area image I by difference9Using structural elements of linear structure to ice column region I9Carrying out corrosion treatment to obtain an ice column connected domain image I without adhesion10；
Step 5, segmenting the insulator string into images I5Transversely overlapping to obtain an image I11Then passing through a distance threshold T3And a maximum threshold value T4Removing local maxima to determine a global maximum corresponding to the insulator long shaft, namely a row where the center position is located; in picture I5Searching two boundary pixel points of the row coordinate, determining the column coordinate of a central point through the two boundary pixel points so as to determine the central point of the insulator, and calculating the average value of Euclidean distances of the central points between adjacent insulators to obtain a threshold value Tc of ice column bridging between insulator umbrellas;
step 6, calculating an icicle connected domain image I10The minimum external rectangle and the vertical length Jc of the icicle connected domain in the (1) take Tc/3 and Tc as threshold values, and the bridging degree between the glass insulator umbrellas is divided into: no bridging, hidden danger and bridging.
2. The method for detecting the bridging degree of the icicles among the glass insulator umbrellas as claimed in claim 1, wherein the step 2 is implemented by the following steps:
step 2.1, to image I1And (3) carrying out significance detection to protrude the umbrella surface part and the icicle part of the insulator, removing the insulator connecting piece, and carrying out the image significance detection process as follows:
first, the grayed image I is subjected to a graying process1Performing two-dimensional discrete Fourier transform, converting the two-dimensional discrete Fourier transform from a space domain to a frequency domain, and calculating an amplitude spectrum A (f) and a phase spectrum P (f);
Figure FDA0002927500150000021
P(f)＝arctan(Im(F(I1))/Re(F(I1))) (2-2)
where Re is the real part of the function going to the frequency domain; im is the imaginary part of the function converted to the frequency domain; f represents a two-dimensional discrete Fourier transform;
then, taking logarithm of the amplitude A (f) to obtain log spectrum L (f)
L(f)＝log(A(f)) (2-3)
Since the log curve satisfies the local linearity condition, the local usePartial averaging filter hn(f) Smoothing it to obtain an average spectrum, where hnIs a convolution kernel for n × n mean filtering, so the spectrum residual is the difference between the log spectrum and its mean filtering, and is calculated according to the following equation:
R(f)＝L(f)-hn(f)*L(f) (2-4)
where, denotes convolution, r (f) is the Spectral Residual spectrum, which is used as the new amplitude and the original phase spectrum p (f) to reconstruct a new frequency domain spectrum;
finally, the reconstructed frequency domain spectrum is subjected to Fourier inverse transformation and then to Gaussian fuzzy filtering,
s(x,y)＝g(x,y)*F-1[exp(R(f)+i×P(f))]2 (2-5)
wherein g (x, y) is a gaussian kernel of 5 × 5,. sigma.1.5, F-1Representing an inverse fourier transform;
obtaining a salient region image I through Fourier inverse transformation and Gaussian fuzzy filtering2；
Step 2.2, threshold segmentation binarization to obtain a binary image I3。
3. The method for detecting the bridging degree of icicles between glass insulator umbrellas as claimed in claim 1,
the method is characterized in that the specific implementation steps of the step 3 are as follows:
step 3.1, to image I1Otsu threshold segmentation is carried out to obtain an image I4For image I4Sequentially carrying out area screening and color reversal treatment, and carrying out area screening and color reversal treatment again, wherein the area threshold values of the two times are respectively T1And T2When the image I4When the area of the connected domain area is larger than the area threshold value, the connected domain is reserved, and the image I is traversed4Obtaining an insulator string segmentation image I by all connected domains5；
Step 3.2, mixing I5Mapping back to image I, taking out insulator string region, i.e. image I6And image I6Decomposed into R, G, B three-channel images, respectively denoted as I6_R、I6G and I6_B；
Step 3.3, finding out through image tests that when G-R is larger than 0, the image is green, and the larger the absolute value of G-R is, the darker the color is, so that the image I comprising the insulator string region is subjected to the expression (3-1)6The pixel points are processed one by one to obtain an image I7，I7The specific calculation method of the pixel point at the (x, y) position at any position is as follows:
Figure FDA0002927500150000031
obtaining an image I through threshold segmentation8。
4. The method for detecting the bridging degree of the icicles between the glass insulator umbrellas as claimed in claim 1, wherein the specific implementation manner of the step 4 is as follows:
step 4.1, differencing the images obtained in step 2 and step 3, namely, image I3And image I8The difference result of the pixel values of the same pixel positions of the two images is given to the ice column area image I9Corresponding image position I9(x, y), the computational expression of which is as follows:
I9(x,y)＝I3(x,y)-I8(x,y) (4-1)；
step 4.2, taking a linear structure with the length of L pixels in the vertical direction which is the 90-degree direction as a structural element, and carrying out ice column region image I9Carrying out corrosion treatment to obtain an icicle connected domain image I10。
5. The method for detecting the bridging degree of the icicles between the glass insulator umbrellas as claimed in claim 1, wherein in the step 5, since the insulator is a circular structure, the diameter of the insulator must pass through the center of the circle, the insulator appears as an ellipse in the image, and then the midpoint of the longest axis of each insulator in the image is the center of the insulator, and based on this, the insulator string region is transversely superimposed and the maximum point is obtained, and due to the influence of the ice coating of the insulator, the superimposed curve is not smooth, and the stored maximum point is storedAt a large number of local maximum points, the local maximum points are not the centers of the insulators, and the existence of the local maximum points can cause errors in the calculated value of the ice column bridging threshold Tc between the umbrella-shaped insulators, so that the local maximum values must be removed, and the method for removing the local maximum values to obtain the global maximum values comprises the following steps: by distance threshold T3Grouping the maximum value points, determining the maximum value of the maximum value in each group as the maximum value of the group, and then utilizing the threshold value T of the maximum value4Eliminating local maximum value and obtaining global maximum value and its local image I5Calculating the center point of the insulator sheet through two boundary points of the position, wherein the icicle bridging threshold Tc between the insulator umbrellas is equivalent through the Euclidean distance of the center point between the insulators;
the solving process of the ice column bridging threshold Tc between the insulator umbrellas in the step 5 is as follows:
step 5.1, because the longest axis of the insulator must include the center of the insulator, finding the position of the longest axis of the insulator and averaging two boundaries to obtain the coordinate of the center;
therefore to image I5Performing transverse superposition to obtain a transverse superposition curve I of the insulator string region11：
Figure FDA0002927500150000041
And 5.2, according to the definition of the maximum value: if to x0All points nearby, all having f (x)<f(x0) Then f (x)0) Is a maximum of the function f (x), so that the insulator string region is superimposed laterally by the curve I11The number of the superimposed pixels is a maximum value point if the number of the superimposed pixels simultaneously satisfies the following conditions:
Figure FDA0002927500150000051
obtaining a transverse superposition curve I of the insulator string region11Respectively recording the maximum value and the corresponding coordinates of the maximum value by using the arrays a and b;
step 5.3, due to the influence of ice coating, the smoothness of the superposed curve is not enough, a large number of local maximum value points exist, and the distance passes through a distance threshold T3And a maximum threshold value T4Removing local maximum values;
taking the first extreme point as the maximum point of the first maximum value group, then taking the second maximum point, and calculating the coordinate distance between the first and second maximum points and the distance threshold value T3Comparing if it is less than the distance threshold T3If not, establishing a new maximum value group and taking the second extreme point as the maximum value point of the newly established maximum value group, and completing the grouping of all the detected maximum value points by analogy;
after grouping is completed, taking the maximum value in the group as the final maximum value of the group, and respectively recording the grouping maximum value and the corresponding coordinate thereof by using the arrays ma and md;
setting a maximum threshold T4The long shaft filter is used for filtering a false insulator long shaft;
firstly, the maximum value in the array ma is taken, and then the maximum values are compared with other maximum values in the array one by one, if the difference value is less than the maximum value threshold value T4Then the insulator is considered as a long axis; if the difference is larger than the maximum threshold value T4Considered as a long axis of the dummy insulator; the finally determined long axis of the insulator is represented by an array jr;
step 5.4, taking out the line coordinate corresponding to the first long axis, and obtaining the line coordinate in the image I5Finding two boundary pixel points of the insulator, solving the mean value of the sum of the two boundary point row coordinates to be the row coordinate of the center of the insulator, locking the center position of the insulator through the row and column coordinates, and finding the centers of the rest insulators by analogy;
if the image to be detected only comprises two glass insulators, Tc is the Euclidean distance between the two central points; if the number of insulators included is more than two, Tc is the average value of the euclidean distance between every two adjacent central points.
6. The method for detecting the icicle bridging degree between glass insulator umbrellas as claimed in claim 1, wherein in the step 6, the glass insulator umbrellas bridging degree is divided into three grades: no bridging, hidden danger and bridging; the specific process of classification is as follows:
calculating icicle connected domain image I10The minimum external rectangle and the vertical length Jc of the icicle connected domain are sequentially judged:
if Jc is less than Tc/3, no bridge connection exists, and the minimum circumscribed rectangle of the icicle connected domain is framed and selected by a black dotted line in the image I;
if Tc/3 is less than or equal to Jc and less than Tc, hidden danger exists, and the minimum circumscribed rectangle of the icicle connected domain is framed and selected by a black solid line in the image I;
and if the Jc is larger than or equal to the Tc, bridging, and using a red solid line to frame and select the minimum circumscribed rectangle of the icicle connected domain in the image I.
CN201910543745.5A 2019-06-21 2019-06-21 Method for detecting icicle bridging degree between glass insulator umbrellas Active CN110349128B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910543745.5A CN110349128B (en) 2019-06-21 2019-06-21 Method for detecting icicle bridging degree between glass insulator umbrellas

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910543745.5A CN110349128B (en) 2019-06-21 2019-06-21 Method for detecting icicle bridging degree between glass insulator umbrellas

Publications (2)

Publication Number Publication Date
CN110349128A CN110349128A (en) 2019-10-18
CN110349128B true CN110349128B (en) 2021-04-06

Family

ID=68182726

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910543745.5A Active CN110349128B (en) 2019-06-21 2019-06-21 Method for detecting icicle bridging degree between glass insulator umbrellas

Country Status (1)

Country Link
CN (1) CN110349128B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111047561A (en) * 2019-11-22 2020-04-21 国网江西省电力有限公司电力科学研究院 Method for identifying cracks of shed tortoise of composite insulator
CN111289532B (en) * 2020-02-24 2023-04-07 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 Discrimination method for fake post insulator
CN112150495B (en) * 2020-09-24 2023-03-14 西安工程大学 Insulator ice flashover risk prediction method based on deduction rule
CN112816043B (en) * 2020-12-30 2023-01-06 北京万集科技股份有限公司 Method and system for determining vehicle type of vehicle
CN113235504A (en) * 2021-05-31 2021-08-10 国能朔黄铁路发展有限责任公司 Tunnel ice hanging identification method, device and equipment and tunnel deicing system
CN116797604B (en) * 2023-08-28 2023-12-26 中江立江电子有限公司 Glass insulator defect identification method, device, equipment and medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5276286A (en) * 1991-08-28 1994-01-04 Gec Alsthom Energie Inc. Exposed high voltage outdoor disconnector for operation under icing conditions
CN102175203A (en) * 2011-01-30 2011-09-07 西安工程大学 Method for analyzing icing prominent influence factors of power transmission line
CN102346015A (en) * 2011-06-24 2012-02-08 西安工程大学 Method for measuring icing thickness of insulator of power transmission line based on video difference analysis
CN204807083U (en) * 2015-07-10 2015-11-25 中国电力科学研究院 Electrified insulator chain icing bridging degree intelligence image analysis device
CN105513039A (en) * 2015-07-10 2016-04-20 中国电力科学研究院 Charged insulator string icing bridging degree intelligent image analysis method
CN106060466A (en) * 2016-06-20 2016-10-26 西安工程大学 Video image sequence-based insulator tracking monitor method
CN106340006A (en) * 2016-08-22 2017-01-18 华南理工大学 Icing degree assessment method based on insulator image umbrella stretch-out
CN109215020A (en) * 2018-08-30 2019-01-15 国网黑龙江省电力有限公司佳木斯供电公司 Ultra-high-tension power transmission line fault recognition method based on computer vision

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5276286A (en) * 1991-08-28 1994-01-04 Gec Alsthom Energie Inc. Exposed high voltage outdoor disconnector for operation under icing conditions
CN102175203A (en) * 2011-01-30 2011-09-07 西安工程大学 Method for analyzing icing prominent influence factors of power transmission line
CN102346015A (en) * 2011-06-24 2012-02-08 西安工程大学 Method for measuring icing thickness of insulator of power transmission line based on video difference analysis
CN204807083U (en) * 2015-07-10 2015-11-25 中国电力科学研究院 Electrified insulator chain icing bridging degree intelligence image analysis device
CN105513039A (en) * 2015-07-10 2016-04-20 中国电力科学研究院 Charged insulator string icing bridging degree intelligent image analysis method
CN106060466A (en) * 2016-06-20 2016-10-26 西安工程大学 Video image sequence-based insulator tracking monitor method
CN106340006A (en) * 2016-08-22 2017-01-18 华南理工大学 Icing degree assessment method based on insulator image umbrella stretch-out
CN109215020A (en) * 2018-08-30 2019-01-15 国网黑龙江省电力有限公司佳木斯供电公司 Ultra-high-tension power transmission line fault recognition method based on computer vision

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
Hybrid method for analysis of operation state of the composite insulator;Yi Tian et al;《2016 International Conference on Condition Monitoring and Diagnosis (CMD)》;20160928;第448-451页 *
Modeling of AC flashover on ice-covered composite insulators with different shed configurations;Lichun Shu et al;《 IEEE Transactions on Dielectrics and Electrical Insulation》;20141231;第21卷(第6期);第2642-2651页 *
冰柱模型和支柱绝缘子覆冰直流闪络特性与机理研究;苑吉河;《中国优秀博硕士学位论文全文数据库 (硕士)工程科技Ⅱ辑》;20050315;第C042-446页 *
复杂环境下覆冰绝缘子识别检测技术;黄新波等;《高电压技术》;20170310;第195-202页 *
玻璃绝缘子覆冰图像分析;蒋晓蓝;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;20180615;第C042-21页 *
输电线路绝缘子覆冰厚度图像识别算法;张烨等;《电力系统自动化》;20161130;第891-899页 *

Also Published As

Publication number Publication date
CN110349128A (en) 2019-10-18

Similar Documents

Publication Publication Date Title
CN110349128B (en) Method for detecting icicle bridging degree between glass insulator umbrellas
CN105004972B (en) Porcelain insulator Condition assessment of insulation method based on day blind ultraviolet imagery characteristics of image
CN106680285B (en) Method for recognizing insulator contamination state based on infrared image assisted visible light image
CN108734689B (en) Method for detecting scattered strands of conducting wires based on region growth
CN102520286B (en) Hyperspectrum-based composite insulator operation state classification method
CN108986104B (en) Insulator image segmentation method based on distance transformation and watershed marking algorithm
CN103047938B (en) Electric power line ice-covering thickness detection method and pick-up unit
CN111754465B (en) Insulator positioning and string dropping detection method
CN102519846B (en) Hyperspectrum-based composite insulator hydrophobicity detection method
CN107506768A (en) A kind of stranded recognition methods of transmission line wire based on full convolutional neural networks
CN111695506B (en) Early warning method and system for wind-induced foreign matter short-circuit fault of transmission line
CN106340006B (en) A kind of icing degree assessment method stretched out based on insulation subgraph umbrella
CN108108772A (en) A kind of insulator contamination condition detection method based on distribution line Aerial Images
CN104849287A (en) Composite insulator contamination degree non-contact detection method
CN109146852A (en) A kind of porcelain insulating substring chapeau de fer disk edge detection method based on infrared image
CN102519974B (en) Method for detecting cracking of composite insulator based on intelligent image identification
CN104615811A (en) Mapping method for electric transmission line freezing weather risk level distribution diagram
CN109523543A (en) A kind of wire strand breakage detection method based on Edge Distance
CN110853089A (en) Multi-factor-based simulation wire icing thickness algorithm
CN107729397B (en) Method for judging ice and snow covering type by using RGB (red, green and blue) components of image
CN112150495A (en) Insulator ice flashover risk prediction method based on deduction rule
CN112270234A (en) Power transmission line insulation sub-target identification method based on aerial image
CN112016641A (en) Method and device for alarming line short circuit fault caused by foreign matter
CN110223346B (en) Shape prior-based multi-insulator positioning method
CN110109195B (en) Lightning approach prediction method based on radar and sounding data

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