CN108734709B - Insulator flange shape parameter identification and damage detection method - Google Patents
Insulator flange shape parameter identification and damage detection method Download PDFInfo
- Publication number
- CN108734709B CN108734709B CN201810533013.3A CN201810533013A CN108734709B CN 108734709 B CN108734709 B CN 108734709B CN 201810533013 A CN201810533013 A CN 201810533013A CN 108734709 B CN108734709 B CN 108734709B
- Authority
- CN
- China
- Prior art keywords
- flange
- insulator
- area
- circularity
- region
- 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
- 239000012212 insulator Substances 0.000 title claims abstract description 75
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 30
- 230000007547 defect Effects 0.000 claims abstract description 15
- 230000002950 deficient Effects 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims abstract description 7
- 230000011218 segmentation Effects 0.000 claims abstract description 6
- 238000009434 installation Methods 0.000 claims abstract description 5
- 238000012360 testing method Methods 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims abstract description 5
- 238000013507 mapping Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 5
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000005315 distribution function Methods 0.000 claims description 3
- 230000000630 rising effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 230000005540 biological transmission Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 210000002364 input neuron Anatomy 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T7/187—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
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/64—Analysis of geometric attributes of convexity or concavity
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a method for identifying and detecting shape parameters and damage of an insulator flange, which comprises the following steps: step 1: adjusting the installation position and the shooting angle of the camera according to the position of the insulator string; carrying out histogram equalization, segmentation, filtering and connected domain marking on the collected field insulator image to finally obtain an insulator flange region; step 2: by analyzing the characteristics of the intact flange and the defective flange of the insulator in the image, extracting four shape characteristics of the perimeter C, the area A, the distance d from the center of a circle and the circularity e of the flange region as characteristic parameters of the defect detection of the flange, and further determining whether the insulator is damaged; and step 3: and (3) constructing an insulator flange damage detection model, taking the four characteristic parameters obtained in the step (2) as input, taking the flange with good/defective as output, training and testing the model, and finally realizing the detection and identification of the insulator flange damage. The method can realize non-contact and on-line monitoring of the damage of the flange of the insulator.
Description
Technical Field
The invention belongs to the technical field of image processing of power transmission lines, and particularly relates to a method for identifying and detecting shape parameters of an insulator flange.
Background
With the rapid development of economy, the power demand is gradually increased, high-voltage and ultrahigh-voltage overhead power lines become the main modes of long-distance power transmission and distribution, and the safe and stable operation of the overhead power lines is an important guarantee for economic construction in China. Insulators occupy an important position in high-voltage transmission lines: firstly, provide mechanical support for the wire of transmission current, secondly prevent that the electric current from to the ground formation passageway ground connection, its operating condition direct relation is to the normal operating of whole electric power system. The insulator bears the action of working voltage and various overvoltage in the operation process, and also bears the self weight of the insulator, the weight of a lead, wind power, mechanical force, weather change and corrosion of chemical substances, and the insulator is severe in working condition and is easy to have various faults such as cracking, skirt edge defect, flange defect, bulb corrosion and the like. Although the local flange defect does not necessarily cause an accident, it is preferable to find a replacement in time because the defect spreads into a crack. However, the existing flange damage mainly depends on manual detection, and has the problems of poor safety, time consumption, labor consumption and the like, so an intelligent insulator flange defect detection technology needs to be provided urgently.
Disclosure of Invention
The invention aims to provide a method for identifying and detecting the shape parameters of an insulator flange, which can establish a model for detecting the damage of the insulator flange, take characteristic parameters such as the perimeter, the area, the distance between the relative circle centers and the circularity of the insulator flange, which are obtained by extracting through an image processing technology, as input, finally output the complete damage condition of the insulator flange and realize the non-contact and on-line monitoring of the damage of the flange.
The technical scheme adopted by the invention is that the method for identifying the shape parameters and detecting the damage of the flange of the insulator is implemented according to the following steps:
step 1: adjusting the installation position and the shooting angle of the camera according to the position of the insulator string; carrying out histogram equalization, segmentation, filtering and connected domain marking on the collected field insulator image to finally obtain an insulator flange region;
step 2: by analyzing the characteristics of the intact flange and the defective flange of the insulator in the image, extracting four shape characteristics of the perimeter C, the area A, the distance d from the center of a circle and the circularity e of the flange region as characteristic parameters of the defect detection of the flange, and further determining whether the insulator is damaged;
and step 3: and (3) constructing an insulator flange damage detection model, taking the four characteristic parameters obtained in the step (2) as input, taking the flange with good/defective as output, training and testing the model, and finally realizing the detection and identification of the insulator flange damage.
The present invention is also characterized in that,
step 1 is carried out according to the following steps:
step 1.1: the contrast is limited, and the image is enhanced by a self-adaptive histogram equalization algorithm;
step 1.2: extracting an insulator region by a maximum inter-class variance method;
step 1.3: and removing noise by adopting morphology and Gaussian filtering, and screening out a flange area of the insulator by a connected domain marking method.
Step 1.1 is carried out according to the following steps:
step 1.1.1, dividing the insulator image into a plurality of sub-blocks, calculating a histogram of each sub-block and cutting the histogram by using a predefined threshold value; assuming that the number of gray levels in the histogram is F, the clipping value is limit, solving the partial processes higher than the value in the histogram and assuming that the partial processes are evenly divided into all the gray levels, and solving the process/F of the height of the whole rising of the histogram; taking upper-limit-process/F as a limit, if the amplitude is higher than the limit, directly setting the limit, filling the amplitude between the upper and the limit to the limit, and directly filling the amplitude to the limit if the amplitude is lower than the upper to directly fill the process/F pixel points;
step 1.1.2, assuming that the sub-block size is N × N, the local cumulative distribution function of the jth sub-block is cdf (j), and the derivative of cdf (j) is a local histogram, the local mapping function is:
the interpolation operation is used, namely the value at each pixel point is obtained by carrying out bilinear interpolation on the mapping function values of four sub-blocks around the pixel point, except for the boundary point.
In step 2, the specific calculation method of the perimeter C, the area a, the distance d from the center of the circle and the circularity e is as follows:
(1) perimeter C: the sum of the pixels of all the pixel points on the boundary of the connected region is the perimeter;
(2) area A: moment invariant refers to a feature quantity that remains unchanged after translation, rotation, or scaling of the position of the target object in the image. The p + q moment of the pixel (x, y) in the connected domain is expressed as:
zero order momentRepresents the sum of the pixels in the connected component, i.e., the area a of the connected component;
(3) distance d from circle center: the first moment can be obtained according to the formula (2-1)Anddivided by the zero-order moment M, respectively00Obtained thereafterAndthat is, the centroid coordinate of the connected domain, the circle center position (a, b) of the flange region can be obtained by using Hough elliptic transformation, and the distance d from the circle center can be expressed as:
(4) circularity e: the characteristic quantity of the relevant region shape calculated based on the region area and the perimeter is used for describing the proximity degree of the target region and the circle; based on the circumference C and the area a calculated in (1) and (2), a calculation formula of the circularity e can be obtained:
the value range of e is between 0 and 1, the circularity of the circle is 1, the circularity of other shapes is less than 1, and the value of e is larger as the shape of the connected domain is closer to the circle.
In the step 3, the constructed insulator flange damage detection model is a convolutional neural network, an input layer of the convolutional neural network comprises 4 neurons, the input is respectively the perimeter, the area, the distance from a relative circle center and the circularity, the output is 0 or 1, the flange outputs 0 in a complete state, and the defect outputs 1.
The invention has the beneficial effects that: the method comprises the steps of extracting shape characteristics such as perimeter, area, distance relative to circle center, circularity and the like of an insulator flange region through image enhancement, segmentation and the like, establishing an insulator flange damage detection model, taking characteristic parameters as input, finally outputting damage conditions of the insulator flange, and realizing non-contact and on-line monitoring of flange damage.
Drawings
FIG. 1 is a flow chart of an insulator flange shape parameter identification and damage detection method of the present invention;
fig. 2 is a block diagram of histogram equalization 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 provides a method for identifying and detecting shape parameters and damage of an insulator flange, which is implemented according to the following steps as shown in figure 1:
step 1: the insulator flanges are distributed on the lower surface of the insulator, so that the installation position and the shooting angle of the camera need to be adjusted according to the position of the insulator string. And carrying out histogram equalization, segmentation, filtering, connected domain marking and other processing on the collected field insulator image to finally obtain the flange region of the insulator.
Step 1.1: the invention adopts a Contrast-Limited Adaptive Histogram Equalization (CLAHE) algorithm to enhance the Contrast of a target pixel and a background pixel in an image, and the method specifically comprises the following steps:
step 1.1.1, the insulator image is divided into a number of sub-blocks, the histogram of each sub-block is calculated and the histogram is clipped with a predefined threshold. Assuming that the number of gray levels in the histogram is F and the clipping value is limit, the excess of the part of the histogram higher than the value is obtained and is divided equally to all the gray levels, and the height excess/F of the whole rising of the histogram is obtained. Taking upper-limit-process/F as a limit, if the amplitude is higher than the limit, directly setting the limit, filling the amplitude between the upper and the limit to the limit, and directly filling the amplitude to the limit if the amplitude is lower than the upper to directly fill the process/F pixel points;
step 1.1.2, assuming that the sub-block size is N × N, and the local cumulative distribution function of the jth sub-block is cdf (j) (the derivative of cdf (j) is a local histogram), the local mapping function is:
if the pixel point in each sub-block is only transformed by the local mapping function of the block, the block effect is caused, so that the interpolation operation is utilized, namely the value at each pixel point is obtained by carrying out bilinear interpolation on the mapping function values of four sub-blocks around the pixel point (except for the boundary point). In fig. 2, taking 4 × 4 sub-blocks as an example, central pixel points of four sub-blocks at the upper left corner, the lower left corner, the upper right corner and the lower right corner of an image form a dashed quadrangle, and a boundary pixel point refers to a pixel point outside the quadrangle. And performing mapping transformation and bilinear interpolation on the pixel points in the dotted line quadrangle by using four sub-blocks around the pixel points, performing mapping transformation on shadow pixel points by using one sub-block directly for the pixel points outside the dotted line quadrangle, performing mapping transformation and linear interpolation on the rest pixel points by using two sub-blocks, and respectively calculating to obtain a mapping function of each sub-block in the image.
Step 1.2: the image area can be roughly divided into an insulator area and a non-insulator area (background area), and the invention adopts a maximum between-class variance segmentation method to segment the image. Assuming that L gray levels exist in an image, a pixel point is divided into C through a gray level threshold s0And C1Two kinds, C0Pixel points representing gray levels 1 to s, C1The pixel points with the gray levels from s to L are represented, and the probability of occurrence of the two types is respectively as follows:
in the formula PiAnd expressing the probability of the pixel point with the gray level i. The average gray levels of the two classes are:
optimum threshold s*:
All gray values in the image are compared with s*Making a comparison to be less than s*The pixel of (2) is set to 0 (black), and the rest pixels are set to 255 (white), so that a binary image of the insulator can be obtained.
Step 1.3: and local holes in the image are eliminated by adopting morphological closed operation, and irrelevant information in the image is further removed by adopting Gaussian filtering, so that the accuracy of a subsequent algorithm is ensured. The obtained target area is formed by combining a plurality of mutually communicated pixels. The method adopts an 8-connected domain marking method for marking, wherein 8-connected pixels comprise upper, lower, left, right and diagonal pixels with a common vertex angle with a central pixel, and after marking is finished, a flange area of the insulator is obtained by screening.
Step 2: by analyzing the characteristics of the intact flange and the defective flange of the insulator in the image, extracting four shape characteristics of the perimeter C, the area A, the distance d from the center of a circle and the circularity e of the flange region as characteristic parameters of the defect detection of the flange, and further determining whether the insulator is damaged;
the flange of the insulator is actually distributed on the lower surface of the insulator in a concentric circle shape, but due to the influence of factors such as the field installation position of the insulator, the shooting angle and the like, a camera cannot shoot an image right below the insulator, so the flange area is generally distributed in a concentric ellipse shape in the image. By analyzing the insulator image, the intact insulator flange is a continuous ellipse, the perimeter and the area are maximum, only one perimeter value and one area value exist, the centroid and the circle center of the ellipse are superposed (the distance relative to the circle center is 0), and the circularity is less than 1; however, when the flange has defects, the original continuous ellipse can be broken, one connected domain can be divided into two or more, the perimeter and the area can be correspondingly reduced, the number of the connected domains is increased, the position of the center of mass deviates, and the circularity of the broken part is greater than that of other areas of the flange. Therefore, through analysis and experiments, the four shape characteristics of the perimeter, the area, the distance relative to the circle center and the circularity are finally selected as characteristic parameters of defect detection, and the specific calculation method is as follows:
(1) perimeter C: the sum of the pixels of all the pixel points on the boundary of the connected region is the perimeter.
(2) Area A: moment invariant refers to a feature quantity that remains unchanged after translation, rotation, or scaling of the position of the target object in the image. The p + q moment of the pixel (x, y) in the connected domain is expressed as:
zero order momentRepresenting the sum of the pixels in the connected component, i.e., the area a of the connected component.
(3) Distance d from circle center: the first moment can be obtained according to the formula (2-1)Anddivided by the zero-order moment M, respectively00Obtained thereafterAndi.e. the centroid coordinates of the connected domain. The position (a, b) of the center of the flange area can be obtained by using Hough elliptic transformation, and the distance d from the center of the circle can be expressed as:
(4) circularity e: the feature value of the shape of the region is calculated based on the area and the perimeter of the region, and is generally used to describe the proximity of the target region to a circle. Based on the circumference C and the area a calculated in (1) and (2), a calculation formula of the circularity e can be obtained:
the value range of e is between 0 and 1, the circularity of the circle is 1, the circularity of other shapes is less than 1, and the value of e is larger as the shape of the connected domain is closer to the circle.
And step 3: and (3) constructing an insulator flange damage detection model, taking the four characteristic parameters obtained in the step (2) as input, taking the flange with good/defective as output, training and testing the model, and finally realizing the detection and identification of the insulator flange damage.
In a relatively simple case, it is possible to determine whether there is a breakage of the insulator flange using one of the four characteristic parameters obtained in step 2 or using a linear method. In practical application, however, the flange damage conditions are various, the judgment of the damage condition of the insulator only through one characteristic has certain limitation, and if the damage condition of the flange of the insulator is accurately judged, the four characteristics are combined to carry out comprehensive judgment, namely, a recognition model is established. The convolutional neural network is a feedforward neural network, and has unique superiority in the aspects of speech recognition, image processing and the like due to the special structure of local weight sharing. The basic structure comprises two layers: the input of each neuron is connected with the local receiving domain of the previous layer and the local feature is extracted; each calculation layer of the feature mapping layer is composed of a plurality of feature mappings, each feature mapping is a plane, and the weights of all neurons on the plane are equal and are shared, so that the number of free parameters of the network is reduced. Therefore, the convolutional neural network is used as a detection model for the damage of the flange of the insulator, an input layer of the convolutional neural network comprises 4 neurons, the input neurons are respectively the perimeter, the area, the distance relative to the circle center and the circularity, the output is 0 or 1, the flange outputs 0 in a complete state, and the defect output is 1. Selecting a plurality of insulator images, extracting the four required characteristic parameters through the steps 1 and 2, carrying out network training and testing on the obtained data in a ratio of 2:1, outputting corresponding recognition results, comparing the recognition results with actual conditions, and judging the accuracy of the model.
Claims (1)
1. The method for recognizing the shape parameters and detecting the damage of the flange of the insulator is characterized by comprising the following steps of:
step 1: adjusting the installation position and the shooting angle of the camera according to the position of the insulator string; carrying out histogram equalization, segmentation, filtering and connected domain marking on the collected field insulator image to finally obtain an insulator flange region;
step 2: by analyzing the characteristics of the intact flange and the defective flange of the insulator in the image, extracting four shape characteristics of the perimeter C, the area A, the distance d from the center of a circle and the circularity e of the flange region as characteristic parameters of the defect detection of the flange, and further determining whether the insulator is damaged;
and step 3: constructing an insulator flange damage detection model, taking the four characteristic parameters obtained in the step 2 as input, taking the flange which is intact or has defects as output, training and testing the model, and finally realizing the detection and identification of the insulator flange damage;
the step 1 is implemented according to the following steps:
step 1.1: the contrast is limited, and the image is enhanced by a self-adaptive histogram equalization algorithm;
step 1.2: extracting an insulator region by a maximum inter-class variance method;
step 1.3: removing noise by adopting morphology and Gaussian filtering, and screening out a flange area of the insulator by a connected domain marking method;
step 1.1 is carried out according to the following steps:
step 1.1.1, dividing the insulator image into a plurality of sub-blocks, calculating a histogram of each sub-block and cutting the histogram by using a predefined threshold value; assuming that the number of gray levels in the histogram is F, the clipping value is limit, solving the partial processes higher than the value in the histogram and assuming that the partial processes are evenly divided into all the gray levels, and solving the process/F of the height of the whole rising of the histogram; taking upper-limit-process/F as a limit, if the amplitude is higher than the limit, directly setting the limit, filling the amplitude between the upper and the limit to the limit, and directly filling the amplitude to the limit if the amplitude is lower than the upper to directly fill the process/F pixel points;
step 1.1.2, assuming that the sub-block size is N × N, the local cumulative distribution function of the jth sub-block is cdf (j), and the derivative of cdf (j) is a local histogram, the local mapping function is:
carrying out bilinear interpolation on the value of each pixel point by using the mapping function values of four sub-blocks around the pixel point, except for boundary points, by utilizing interpolation operation;
in step 2, the specific calculation method of the perimeter C, the area a, the distance d from the center of the circle and the circularity e is as follows:
(1) perimeter C: the sum of the pixels of all the pixel points on the boundary of the connected region is the perimeter;
(2) area A: moment invariant refers to a feature quantity which remains unchanged after the position of a target object in an image is translated, rotated or scaled; the p + q moment of the pixel (x, y) in the connected domain is expressed as:
zero order momentRepresents the sum of the pixels in the connected component, i.e., the area a of the connected component;
(3) distance d from circle center: the first moment can be obtained according to the formula (2-1)Anddivided by the zero-order moment M, respectively00Obtained thereafterAndthat is, the centroid coordinate of the connected domain, the circle center position (a, b) of the flange region can be obtained by using Hough elliptic transformation, and the distance d from the circle center can be expressed as:
(4) circularity e: the characteristic quantity of the relevant region shape calculated based on the region area and the perimeter is used for describing the proximity degree of the target region and the circle; based on the circumference C and the area a calculated in (1) and (2), a calculation formula of the circularity e can be obtained:
the value range of e is between 0 and 1, the circularity of the circle is 1, the circularity of other shapes is less than 1, and the value of e is larger as the shape of the connected domain is closer to the circle;
in the step 3, the constructed insulator flange damage detection model is a convolutional neural network, an input layer of the convolutional neural network comprises 4 neurons, the input is respectively the perimeter, the area, the distance from a relative circle center and the circularity, the output is 0 or 1, the flange outputs 0 in a complete state, and the defect outputs 1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810533013.3A CN108734709B (en) | 2018-05-29 | 2018-05-29 | Insulator flange shape parameter identification and damage detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810533013.3A CN108734709B (en) | 2018-05-29 | 2018-05-29 | Insulator flange shape parameter identification and damage detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108734709A CN108734709A (en) | 2018-11-02 |
CN108734709B true CN108734709B (en) | 2022-03-25 |
Family
ID=63936629
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810533013.3A Active CN108734709B (en) | 2018-05-29 | 2018-05-29 | Insulator flange shape parameter identification and damage detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108734709B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111062919B (en) * | 2019-12-12 | 2023-11-03 | 苏州韦士肯检测科技有限公司 | Bearing ring appearance defect detection method |
CN111289853A (en) * | 2020-02-25 | 2020-06-16 | 安徽炬视科技有限公司 | Channel-space attention mechanism-based insulator detection system and algorithm |
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 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105300323A (en) * | 2015-11-18 | 2016-02-03 | 江苏省电力公司连云港供电公司 | Shedding area detection method of RTV coating of insulator and detection device |
CN105957081A (en) * | 2016-04-28 | 2016-09-21 | 华北电力大学(保定) | Glass insulator string dropping fault detection method |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101113145B1 (en) * | 2007-04-05 | 2012-03-13 | 후지쯔 세미컨덕터 가부시키가이샤 | Surface morphology sensor and method for manufacture thereof |
CN105740844A (en) * | 2016-03-02 | 2016-07-06 | 成都翼比特自动化设备有限公司 | Insulator cracking fault detection method based on image identification technology |
CN106290388B (en) * | 2016-08-03 | 2018-09-28 | 国网山东省电力公司电力科学研究院 | A kind of insulator breakdown automatic testing method |
CN106228172B (en) * | 2016-09-20 | 2019-09-24 | 西安工程大学 | Damaged insulator image extraction method based on clustering |
CN107145846B (en) * | 2017-04-26 | 2018-10-19 | 贵州电网有限责任公司输电运行检修分公司 | A kind of insulator recognition methods based on deep learning |
CN107680068A (en) * | 2017-08-31 | 2018-02-09 | 田其冲 | A kind of digital image enhancement method for considering image naturalness |
-
2018
- 2018-05-29 CN CN201810533013.3A patent/CN108734709B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105300323A (en) * | 2015-11-18 | 2016-02-03 | 江苏省电力公司连云港供电公司 | Shedding area detection method of RTV coating of insulator and detection device |
CN105957081A (en) * | 2016-04-28 | 2016-09-21 | 华北电力大学(保定) | Glass insulator string dropping fault detection method |
Non-Patent Citations (1)
Title |
---|
High-speed railway rod-insulator detection using segment clustering and deformable part models;Ye Han et.al;《 2016 IEEE International Conference on Image Processing (ICIP)》;20160819;第3852-3856页 * |
Also Published As
Publication number | Publication date |
---|---|
CN108734709A (en) | 2018-11-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108765373B (en) | Insulator abnormity automatic detection method based on integrated classifier online learning | |
CN107507194B (en) | Insulator string fault detection method based on infrared image temperature distribution rule and BP neural network | |
CN111428748B (en) | HOG feature and SVM-based infrared image insulator identification detection method | |
CN109523529B (en) | Power transmission line defect identification method based on SURF algorithm | |
CN108734709B (en) | Insulator flange shape parameter identification and damage detection method | |
CN108734689B (en) | Method for detecting scattered strands of conducting wires based on region growth | |
CN113205063A (en) | Visual identification and positioning method for defects of power transmission conductor | |
CN108986104B (en) | Insulator image segmentation method based on distance transformation and watershed marking algorithm | |
CN108108772B (en) | Insulator pollution flashover state detection method based on aerial image of distribution line | |
CN107895376A (en) | Based on the solar panel recognition methods for improving Canny operators and contour area threshold value | |
CN111814686A (en) | Vision-based power transmission line identification and foreign matter invasion online detection method | |
CN105957088B (en) | Transformer composite insulator casing monitoring method and system based on computer vision | |
CN112560634B (en) | Power line insulator rapid detection and positioning method and system based on field image | |
CN107179479B (en) | Visible light image-based transmission conductor strand scattering defect detection method | |
CN111126253A (en) | Knife switch state detection method based on image recognition | |
CN111783773B (en) | Correction method for angle-inclined telegraph pole signboard | |
CN110276787B (en) | Conductor galloping monitoring method based on marker image detection | |
CN110766016A (en) | Code spraying character recognition method based on probabilistic neural network | |
CN110874825B (en) | Method for extracting binary image of water trace on surface of composite insulator | |
CN111008967B (en) | Insulator RTV coating defect identification method | |
CN115690012A (en) | Detection method for wrong connection line of electric energy meter | |
CN107886493A (en) | A kind of wire share split defect inspection method of transmission line of electricity | |
CN110363749A (en) | Stockbridge damper corrosion degree evaluation method based on image procossing | |
CN116630321B (en) | Intelligent bridge health monitoring system based on artificial intelligence | |
CN116758423A (en) | Power transmission line foreign matter detection method based on white point rate method |
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 |