CN116612138A - Electrical equipment on-line monitoring method based on image processing - Google Patents
Electrical equipment on-line monitoring method based on image processing Download PDFInfo
- Publication number
- CN116612138A CN116612138A CN202310860813.7A CN202310860813A CN116612138A CN 116612138 A CN116612138 A CN 116612138A CN 202310860813 A CN202310860813 A CN 202310860813A CN 116612138 A CN116612138 A CN 116612138A
- Authority
- CN
- China
- Prior art keywords
- data
- gray scale
- size
- data set
- image
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000012545 processing Methods 0.000 title claims abstract description 29
- 238000012544 monitoring process Methods 0.000 title claims abstract description 19
- 230000002159 abnormal effect Effects 0.000 claims abstract description 95
- 230000005856 abnormality Effects 0.000 claims abstract description 55
- 230000011218 segmentation Effects 0.000 claims abstract description 32
- 238000003064 k means clustering Methods 0.000 claims abstract description 14
- 238000010586 diagram Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 9
- 230000000903 blocking effect Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000001931 thermography Methods 0.000 abstract description 17
- 238000010438 heat treatment Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 15
- 238000003709 image segmentation Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000013021 overheating Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Geometry (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to an on-line monitoring method of electrical equipment based on image processing, which comprises the following steps: acquiring a gray level image of an electrical equipment circuit image; dividing the gray image into blocks, obtaining pixel abnormality degree according to pixel characteristics in the blocks, and correspondingly generating a data set according to the pixel abnormality degree; adjusting the data in the data set according to the data necessity and the pixel gray gradient of the image pixels in the data set; obtaining a data set segmentation result diagram by using a K-means clustering algorithm for the adjusted data set; judging whether the circuit is abnormal or not according to the data set segmentation result diagram. The invention solves the problems that the thermal imaging can only monitor the related conditions of overheat of the circuit surface and the like, but cannot monitor the conditions of short-circuit heating and the like of the rear-row wires with overlapped circuits in time; and the area with problems in the rear row is rapidly and accurately monitored.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an on-line monitoring method of electrical equipment based on image processing.
Background
With the continuous development of society, electric power is the most important energy supply, and electric equipment is operated continuously in a normal state, so that a fault problem is usually caused, and the stability of an electric power system is affected. With the development of the age, the infrared thermal imaging technology is widely applied to electrical equipment monitoring, but the technology is limited at present, the thermal imaging can only monitor relevant conditions such as overheating of the circuit surface and the like and cannot be applied to all scenes, and the conditions such as short-circuit heating of a rear-row wire with overlapped circuits and the like cannot be timely monitored, so that potential safety hazards and economic losses which cannot be timely found by the circuit are caused.
The prior art is limited, thermal imaging can only monitor relevant conditions such as overheating of the circuit surface and cannot be applied to all scenes, and the conditions such as short-circuit heating of a rear-row wire with overlapping lines cannot be timely monitored. In the image segmentation of the k-means clustering algorithm, because abnormal areas in the thermal imaging images with the short-circuit and heating of the rear-row wires with overlapped circuits are extremely unobvious, the data sets of the images are scattered and disordered, and the required abnormal data cannot be gathered into one type, so that the image segmentation effect is poor, and whether the abnormality exists cannot be judged.
Disclosure of Invention
The invention provides an on-line monitoring method of electrical equipment based on image processing, which aims to solve the existing problems.
The on-line monitoring method of the electrical equipment based on image processing adopts the following technical scheme:
an embodiment of the present invention provides an on-line monitoring method for an electrical device based on image processing, the method including the steps of:
acquiring a gray level image of an electrical equipment circuit image;
dividing the gray image into blocks to obtain all large-size gray blocks and small-size gray blocks; obtaining the abnormal degree of the pixels in the large-size gray scale block and the abnormal degree of the pixels in the small-size gray scale block; establishing a rectangular coordinate system according to the abnormal degree of the pixels in the large-size gray scale block and the abnormal degree of the pixels in the small-size gray scale block; generating a data set by corresponding all pixel points in the gray level image to a rectangular coordinate system;
acquiring the data necessity of the pixel points corresponding to the data in the data set; obtaining a spacing adjustment value of data in the data set according to the data necessity of the pixel points corresponding to the data in the data set; adjusting all data in the data set according to the data interval adjustment value of the pixel point corresponding to the data in the data set to obtain an adjusted data set;
clustering the adjusted data sets to obtain a data set segmentation result diagram; judging whether the circuit is abnormal or not according to the data set segmentation result diagram.
Preferably, the step of blocking the gray image to obtain all the large-size gray blocks and the small-size gray blocks includes the following specific steps:
uniformly dividing the gray image according to large-size gray blocks with preset sizes until the gray image is uniformly divided into a plurality of large-size gray blocks; thereby obtaining all large-size gray blocks of the gray image; and dividing all large-size gray blocks in the gray image into small-size gray blocks with preset sizes.
Preferably, the step of obtaining the degree of abnormality of the pixels in the large-size gray scale block and the degree of abnormality of the pixels in the small-size gray scale block includes the following specific steps:
the pixel abnormality degree calculation expression in the large-size gray scale block is as follows:
in the method, in the process of the invention,indicate->The degree of abnormality of pixel points in a large-size gray scale block; />Indicate->Pixel approximation in a large-size gray scale block; />Indicate->Large blockSize intra-gray block variance;
the pixel abnormality degree calculation expression in the small-size gray scale block is:
in the method, in the process of the invention,indicate->The large-sized gray scale block is divided into +.>The degree of abnormality of pixel points in the small-size gray scale blocks; />Indicate->Pixel approximation in a large-size gray scale block; />Indicate->The large-sized gray scale block is divided into +.>Intra-block variance of small-size gray scale; />Indicate->The large-sized gray scale block is divided into +.>The pixel gray average value in the small-size gray scale block.
Preferably, the rectangular coordinate system is established according to the degree of abnormality of the pixels in the large-size gray scale block and the degree of abnormality of the pixels in the small-size gray scale block, and the specific steps include:
establishing a rectangular coordinate system according to the abnormal degree of the pixels in the large-size gray scale blockThe horizontal axis indicates the degree of abnormality of pixels in the small-sized gray scale block +.>Is the vertical axis.
Preferably, the obtaining the pitch adjustment value of the data in the data set according to the data necessity of the pixel point corresponding to the data in the data set includes the following specific steps:
data in a data setThe data pitch adjustment value of (2) is:
in the method, in the process of the invention,representing the coordinates in the dataset as +.>A data pitch adjustment value of the data of (a); />Representing data ∈in a dataset>Initial distance from origin; />Representing data ∈in a dataset>A value normalized by a pixel gray gradient value corresponding to a pixel point in the gray image; />Representing the coordinates in the dataset as +.>The abscissa of the data of (2); />Representing the coordinates in the dataset as +.>Ordinate of data of>Representing the coordinates in the dataset as +.>Data necessity of the pixel point corresponding to the data of (a).
Preferably, the data set after adjustment uses a K-means clustering algorithm to obtain a data set segmentation result graph, which comprises the following specific steps:
and carrying out a K-means clustering algorithm with the K value of 2 on the adjusted data set to obtain a data set segmentation result graph.
Preferably, the determining whether the circuit is abnormal according to the data set segmentation result graph includes the following specific steps:
when the electrical equipment circuit image is abnormal, the data set segmentation result graph is displayed in white, and when the electrical equipment circuit image is abnormal, the data set segmentation result graph is displayed in black, and whether the electrical equipment circuit image has the circuit abnormality is judged according to the data set segmentation result graph.
The technical scheme of the invention has the beneficial effects that: the invention solves the problems that the thermal imaging can only monitor relevant conditions such as overheat of the circuit surface, but cannot monitor the conditions such as short circuit and heating of the rear-row wires with overlapped wires in time, and the temperature propagation influence among the wires is extremely unobvious in the thermal imaging, and the existing thermal imaging image processing technology cannot accurately monitor the area with the rear row problem. According to the invention, according to the change of the front-row wires affected by the temperature of the rear-row abnormal wires, the abnormal degree of each pixel is obtained through size and dimension blocking, the pixel point is converted into data according to the abnormal degree, and then each data is adjusted through the data necessity and the gray gradient of the pixel, so that the data corresponding to the target pixel are more aggregated, thereby achieving a better K-means clustering algorithm segmentation effect.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of steps of an on-line monitoring method of an electrical device based on image processing.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of the on-line monitoring method for the electrical equipment based on image processing according to the invention, which is provided by the invention, with reference to the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the on-line monitoring method for the electrical equipment based on image processing provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of an on-line monitoring method for an electrical device based on image processing according to an embodiment of the present invention is shown, the method includes the following steps: it should be noted that, aiming at the situation that thermal imaging can only monitor the surface overheat of a circuit and other relevant situations, the thermal imaging cannot be applied to all scenes, and the situations such as short-circuit heating of a rear-row wire with overlapping wires cannot be monitored in time, because the temperature propagation influence among wires is extremely insignificant, the existing thermal imaging image processing technology cannot accurately monitor the area with problems of the rear row, thereby causing potential safety hazards and economic losses which cannot be found in time by the circuit. In the image segmentation of the k-means clustering algorithm, because abnormal areas in the thermal imaging images with short-circuit and heating of the rear-row wires with overlapped circuits are extremely unobvious, the data sets of the images are scattered and disordered, and the required abnormal data cannot be gathered into one type, so that the problem that whether the images are abnormal or not cannot be judged due to poor image segmentation effect is solved. The present embodiment can realize local processing of the front and rear double-row electric wire regions of the gradation image of the electric device circuit image by large-size blocking of the gradation image of the electric device circuit image, so that the processing operation is applied only to a specific image block. By independently processing the image blocks of the front and rear double-row wire areas, abnormal characteristics can be more accurately captured; for the purpose of carrying out small-size blocking on the large-size gray scale blocks, in order to carry out abnormal judgment on details of the abnormal large-size gray scale blocks, normal pixel points exist in the large-size gray scale blocks, meanwhile, the large-size gray scale blocks cannot be analyzed to obtain the self abnormal degree due to the size limitation of the large-size gray scale blocks, after the small-size blocks are blocked, whether the small-size gray scale blocks are in an abnormal region in a gray scale image of an electric device circuit image can be analyzed, and if all the small-size gray scale blocks after the large-size gray scale blocks are blocked are divided into the abnormal region, whether the large-size gray scale blocks are abnormal can be distinguished by analyzing the pixel abnormal degree characteristics of the small-size gray scale blocks; the pixel points in the small-size gray scale blocks with large abnormal degrees can also have normal pixel points, and the K-means clustering effect is poor due to the fact that the normal pixel points exist, so that the distance between the data is required to be adjusted according to the abnormal degrees of the small-size gray scale blocks and the large-size gray scale blocks, normal pixel points and abnormal pixel points can be better distinguished when the K-means algorithm is used for classifying, and the detection result of the electrical equipment circuit image with unobvious abnormality is more obvious.
Step S001: a gray scale image of an electrical device circuit image is acquired. Specifically, front and rear double-row wires are connected between the electrical devices, the wires are closely arranged, an infrared thermal imaging camera is used for acquiring thermal imaging images of the front and rear double-row wires connected between the electrical devices, and then the acquired thermal imaging images of the front and rear double-row wires connected between the electrical devices are subjected to a grey-scale algorithm to obtain grey-scale images of circuit images of the electrical devices.
Step S002: and carrying out block processing on the gray level image, obtaining pixel abnormality degree according to pixel characteristics in the block, and correspondingly generating a data set according to the pixel abnormality degree.
When the rear-row wire is abnormal in the gray-scale image of the electric device circuit image, the temperature of the front-row wire is extremely slightly increased, so that the gray-scale image of the electric device circuit image is also extremely slightly increased, and the gray-scale image of the electric device circuit image is directly integrated, so that the detail change cannot be accurately captured.
Local processing of the front and rear double-row wire regions of the grayscale image of the electrical device circuit image can be achieved by large-size blocking of the grayscale image of the electrical device circuit image so that the processing operation is applied only to specific image blocks. By independently processing the image blocks of the front and rear double-row wire areas, the abnormal characteristics can be captured more accurately. The size of the large-size blocks cannot be too small, and if the size of the large-size blocks is too small, abnormal areas in the gray level image of the circuit image of the electrical equipment are divided into a plurality of large-size gray level blocks, the overall effect is lost, and if the overall effect is lost, whether any large-size gray level block is abnormal or not cannot be accurately judged.
For large-size gray scaleThe block is divided into small-size blocks in order to judge the details of the abnormal large-size gray blocks, normal pixel points exist in the large-size gray blocks, meanwhile, the large-size gray blocks cannot be analyzed to obtain the abnormal degree of the large-size gray blocks, after the small-size blocks are divided into small-size blocks, whether the small-size gray blocks are in the abnormal region in the gray map of the electric equipment circuit image can be analyzed, and if all the small-size gray blocks after the large-size gray blocks are divided into the abnormal region, whether the large-size gray blocks are abnormal can be distinguished by analyzing the abnormal degree characteristics of the pixels of the small-size gray blocks. The present embodiment sets a preset large-size gray scale blockThe small-sized gray block is set to +.>The operator can adjust the device according to the actual situation, and the device is not limited to this.
And uniformly dividing the gray image by using the large-size gray blocks until the gray image is uniformly divided into a plurality of large-size gray blocks. The size of the large-size gray scale blocks is set to be combined with the size of the gray scale image, so that the large-size gray scale blocks in the gray scale image are integer numbers, and the large-size gray scale blocks can completely divide the gray scale image. For the special case that the boundary pixel points in the gray level image cannot form a large-size gray level block, overlapping the large-size gray level block of the boundary pixel points with other large-size gray level blocks; all large-size gray blocks in the gray image are divided into small-size gray blocks. So far, all large-size gray scale blocks and small-size gray scale blocks in the gray scale image are acquired.
The pixel approximation degree in the large-size gray scale block is obtained through the maximum and minimum gray scale values in the large-size gray scale block area, and the pixel abnormality degree in the large-size block is obtained through the pixel approximation degree in the block and the variance in the block, so that the probability of short-circuiting the electric wire is higher when the abnormality degree is higher. The pixels in the small-size gray scale blocks pass through, the similarity in the large-size gray scale blocks and the variance in the small-size gray scale blocks, the importance degree in the small-size gray scale blocks is obtained, the higher the importance degree is, the higher the possibility of abnormality in an image is, and then the abnormality degree is obtained according to the gray scale mean value and the importance degree of the pixels in the small-size gray scale blocks, and the higher the abnormality degree is, the higher the possibility of short-circuit wires is.
Specifically, the pixel abnormality degree calculation expression in the large-size gray scale block is:
in the method, in the process of the invention,indicate->The degree of abnormality of the pixel points in the large-size gray scale block is higher, and the possibility of the abnormal pixel points in the block is higher; />Indicate->Pixel approximation in large-scale gray scale block, namely +.>The ratio of the maximum value of the pixel gray level to the minimum value of the pixel gray level in the large-size gray level block is higher, the higher the pixel approximation degree in the block is, the lower the pixel approximation degree in the block is, and the higher the possibility that abnormal pixel points exist in the block is; />Indicate->The intra-block variance of the large-size gray scale.
The larger the variance in the large-size gray scale block and the pixel approximation degree in the large-size gray scale block, the larger the gray scale of the large sizeThe difference of gray values of pixel points in the block is large, and the large difference does not exist in the normal circuit image of the electrical equipment, soCan represent->Degree of pixel abnormality in a large-size gray scale block.
The data can be initially separated by dividing the gray level image of the circuit image of the electrical equipment into large-size blocks, so that the abnormality of the large-size gray level block has certain integrity. Because the large-size blocks have large sizes, even the large-size gray blocks with large abnormal degrees contain a plurality of normal pixel points, and the existing normal pixel points have poor K-means clustering effect, so that the K-means clustering is required to be adjusted according to the abnormal degrees of the small-size gray blocks, normal pixel points and abnormal pixel points can be better distinguished during classification, and the detection result of the circuit images of the electrical equipment with unobvious abnormality is more obvious.
It should be further noted that, since the local abnormality degree of the gray image of the electric device circuit image is not integrated only by the gray variance of the small-size gray block, but is insufficient to determine the abnormality degree of the small-size gray block, and since the larger the gray mean value of the small-size gray block is, the higher but inaccurate the abnormality degree in the small-size gray block is, the approximation degree of the large-size gray block where the small-size gray block is located needs to be used again to comprehensively determine the abnormality degree of the small-size gray block.
The pixel abnormality degree calculation expression in the small-size gray scale block is:
in the method, in the process of the invention,indicate->The large-sized gray scale block is divided into +.>The degree of abnormality of the pixel points in the small-size gray scale blocks is higher, and the probability that the pixel points in the blocks are abnormal pixel points is higher; />Indicate->The larger the pixel approximation degree in the block is, the lower the pixel approximation degree in the block is, and the higher the possibility of abnormal pixel points in the block is; />Indicate->The large-sized gray scale block is divided into +.>Intra-block variance of small-size gray scale; />Indicate->The large-sized gray scale block is divided into +.>The pixel gray average value in the small-size gray scale block.
Finally, establishing a rectangular coordinate system, wherein the degree of abnormality of the pixels in the large-size gray scale block is usedFor the horizontal axis (+)>Axis), degree of abnormality of pixels in small-size gray scale block +.>For the vertical axis (+)>Axes), all pixel points in the gray scale image are corresponding to a rectangular coordinate system to generate a data set.
The mapping operation between the pixel points and the data set is that, for any one large-size gray scale block, the abnormal degree of the large-size gray scale block and the abnormal degree of a plurality of small-size gray scale blocks divided by the large-size gray scale block are assigned to each pixel point in the corresponding small-size gray scale block to be used as the abnormal degree value corresponding to each pixel point.
So far, the abnormal degree of the pixels in all large-size gray scale blocks is obtained, and the abnormal degree of the pixels in all small-size gray scale blocks and all pixel points in the gray scale image correspond to a data set generated by a rectangular coordinate system.
Step S003: and adjusting the data in the data set according to the data necessity and the pixel gray gradient of the pixel points corresponding to the data in the data set.
It should be noted that, the data set generated by the coordinate system corresponding to all the pixels in the gray image is obtained through the above steps, and the data set is obtained through image segmentation, so that part of pixel data needs to be further optimized. In the data set, there are cases that a small number of pixels are abnormal pixels, that is, there are few pixels in the large-size gray scale block that are abnormal pixels and there are few pixels in the large-size block that are not abnormal pixels, and these pixels may be slightly scattered, so that the k-means algorithm segmentation effect is affected.
The pixel points in the small-size gray scale blocks with large abnormal degrees can also have normal pixel points, and the K-means clustering effect is poor due to the fact that the normal pixel points exist, so that the distance between the data is required to be adjusted according to the abnormal degrees of the small-size gray scale blocks and the large-size gray scale blocks, normal pixel points and abnormal pixel points can be better distinguished when the K-means algorithm is used for classifying, and the detection result of the electrical equipment circuit image with unobvious abnormality is more obvious.
It should be noted that, according to the analysis of the data in the data set, it is necessary to aggregate the data in which the pixel points in the large-size gray scale block have few abnormal pixel points in the direction in which the pixel abnormality degree in the large-size gray scale block and the pixel abnormality degree in the small-size gray scale block increase, and aggregate the data in which the pixel abnormality degree in the large-size gray scale block and the pixel abnormality degree in the small-size gray scale block decrease in the direction in which the pixel abnormality degree in the small-size gray scale block does not have few abnormal pixels in the large-size gray scale block, so that it is determined which part of the data in the data set needs to be aggregated by the ratio of the pixel abnormality degree in the large-size gray scale block and the pixel abnormality degree in the small-size gray scale block, and then adjust the data pitch in the data set by the pixel point gray scale gradient corresponding to the data in the data set.
In particular, data in a datasetThe data pitch adjustment value of (2) is:
in the method, in the process of the invention,representing the coordinates in the dataset as +.>A data pitch adjustment value of the data of (a); />Representing data ∈in a dataset>Initial distance from origin；/>Representing data ∈in a dataset>The value after normalization of the pixel gray gradient value of the pixel point in the corresponding gray image is the pixel point +.>Difference from the gray average value of the 8 neighborhood pixel point of the pixel point; />Representing the coordinates in the dataset as +.>The abscissa of the data of (2) also indicates the coordinate +.>The data of the (a) corresponds to the abnormal degree of the pixel point in the large-size gray scale block where the pixel point in the gray scale image is positioned; />Representing the coordinates in the dataset as +.>The ordinate of the data of (2) also indicates the coordinate +.>The data of the (a) corresponds to the abnormal degree of the pixel point in the small-size gray scale block where the pixel point in the gray scale image is positioned; />Representing the coordinates in the dataset as +.>The data necessity of the pixel point corresponding to the data of the image is determined according to the data abnormality degree and the pixel gray gradient of the pixel point in the gray image corresponding to the data.
According to the analysis of the data set, the effect that the embodiment wants to achieve is that the abnormal data are divided together, the normal data are divided together, so that the data with few pixels in the large-size gray scale block as abnormal pixels are required to be gathered in the direction in which the abnormal degree of the large-size gray scale block and the abnormal degree of the small-size gray scale block are increased, the data with few pixels in the large-size gray scale block as non-abnormal pixels are gathered in the direction in which the abnormal degree of the large-size gray scale block and the abnormal degree of the small-size gray scale block are reduced, the data need to be gathered in the direction is judged by the ratio of the abnormal degree of the large-size gray scale block to the abnormal degree of the small-size gray scale block, then the original pitch of the data is adjusted by the pixel gray scale gradient corresponding to the data, the abnormal pixels are gathered in the direction in which the abnormal degree of the large-size gray scale block and the abnormal degree of the small-size gray scale block are increased, the normal pixels are gathered in the direction in which the abnormal degree of the large-size gray scale block and the abnormal degree of the small-size gray scale block are reduced, and finally the pitch after the data are obtained, and the pixels in the gathered region are not gathered together. Therefore, normal pixel points and abnormal pixel points can be distinguished, the clustering effect is better, the distinguishing effect of the abnormal region and the normal region of the circuit image of the electrical equipment is better, and the display of the abnormal region of the circuit image of the electrical equipment is more obvious.
And according to the data interval adjustment value of the pixel point corresponding to the data in the data set, adjusting all the data in the data set on the connection line between the data and the origin of the coordinate system, and obtaining an adjusted data set.
Thus, an adjusted dataset is obtained.
Step S004: obtaining a data set segmentation result diagram by using a K-means clustering algorithm for the adjusted data set; judging whether the circuit is abnormal or not according to the data set segmentation result diagram.
In order to most intuitively determine whether or not an abnormal electric wire exists in the infrared thermal imaging gray level image of the electric equipment connecting line, the display segmentation effect is abnormal and abnormal, and the k value is set to 2, so that an infrared thermal imaging image segmentation result of the electric equipment is obtained. Wherein the present embodiment uses two colors of black and white with obvious chromatic aberration to display the segmentation result.
Specifically, a K-means clustering algorithm with a K value of 2 is performed on the adjusted data set to obtain a data set segmentation result diagram, a class cluster of the data set segmentation result diagram, which is close to the origin, is marked as a normal class, and a class cluster of the data set segmentation result diagram, which is far from the origin, is marked as an abnormal class.
And the pixel points corresponding to the normal class in the gray level image of the electric equipment circuit image in the data set segmentation result diagram are marked as white, and the pixel points corresponding to the abnormal class in the gray level image of the electric equipment circuit image in the data set segmentation result diagram are marked as abnormal pixel points as black.
When the electrical equipment circuit image is abnormal, the data set segmentation result graph is displayed in white, and when the electrical equipment circuit image is abnormal, the data set segmentation result graph is displayed in black, and whether the electrical equipment circuit image has the circuit abnormality is judged according to the data set segmentation result graph.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (7)
1. The on-line monitoring method of the electrical equipment based on the image processing is characterized by comprising the following steps:
acquiring a gray level image of an electrical equipment circuit image;
dividing the gray image into blocks to obtain all large-size gray blocks and small-size gray blocks; obtaining the abnormal degree of the pixels in the large-size gray scale block and the abnormal degree of the pixels in the small-size gray scale block; establishing a rectangular coordinate system according to the abnormal degree of the pixels in the large-size gray scale block and the abnormal degree of the pixels in the small-size gray scale block; generating a data set by corresponding all pixel points in the gray level image to a rectangular coordinate system;
acquiring the data necessity of the pixel points corresponding to the data in the data set; obtaining a spacing adjustment value of data in the data set according to the data necessity of the pixel points corresponding to the data in the data set; adjusting all data in the data set according to the data interval adjustment value of the pixel point corresponding to the data in the data set to obtain an adjusted data set;
clustering the adjusted data sets to obtain a data set segmentation result diagram; judging whether the circuit is abnormal or not according to the data set segmentation result diagram.
2. The method for on-line monitoring of electrical equipment based on image processing according to claim 1, wherein the step of blocking the gray scale image to obtain all large-size gray scale blocks and small-size gray scale blocks comprises the following specific steps:
uniformly dividing the gray image according to large-size gray blocks with preset sizes until the gray image is uniformly divided into a plurality of large-size gray blocks; thereby obtaining all large-size gray blocks of the gray image; and dividing all large-size gray blocks in the gray image into small-size gray blocks with preset sizes.
3. The method for on-line monitoring of electrical equipment based on image processing according to claim 1, wherein the steps of obtaining the degree of abnormality of the pixels in the large-size gray scale block and the degree of abnormality of the pixels in the small-size gray scale block comprise the following specific steps:
the pixel abnormality degree calculation expression in the large-size gray scale block is as follows:
in the method, in the process of the invention,indicate->Large rulerDegree of abnormality of pixel points in the inch gray scale block; />Indicate->Pixel approximation in a large-size gray scale block; />Indicate->Intra-block variance of large-size gray scale;
the pixel abnormality degree calculation expression in the small-size gray scale block is:
in the method, in the process of the invention,indicate->The large-sized gray scale block is divided into +.>The degree of abnormality of pixel points in the small-size gray scale blocks; />Indicate->Pixel approximation in a large-size gray scale block; />Indicate->The large-sized gray scale block is divided into +.>Intra-block variance of small-size gray scale; />Indicate->The large-sized gray scale block is divided into +.>The pixel gray average value in the small-size gray scale block.
4. The method for on-line monitoring of electrical equipment based on image processing according to claim 1, wherein the establishing a rectangular coordinate system according to the degree of abnormality of the pixels in the large-size gray scale block and the degree of abnormality of the pixels in the small-size gray scale block comprises the following specific steps:
establishing a rectangular coordinate system according to the abnormal degree of the pixels in the large-size gray scale blockThe horizontal axis indicates the degree of abnormality of pixels in the small-sized gray scale block +.>Is the vertical axis.
5. The method for on-line monitoring of electrical equipment based on image processing according to claim 1, wherein the step of obtaining the pitch adjustment value of the data in the data set according to the data necessity of the pixel point corresponding to the data in the data set comprises the following specific steps:
data in a data setThe data pitch adjustment value of (2) is:
in the method, in the process of the invention,representing the coordinates in the dataset as +.>A data pitch adjustment value of the data of (a); />Representing data ∈in a dataset>Initial distance from origin; />Representing data ∈in a dataset>A value normalized by a pixel gray gradient value corresponding to a pixel point in the gray image; />Representing the coordinates in the dataset as +.>The abscissa of the data of (2); />Representing the coordinates in the dataset as +.>Ordinate of data of>Representing dataConcentrated coordinates of +.>Data necessity of the pixel point corresponding to the data of (a).
6. The method for on-line monitoring of electrical equipment based on image processing according to claim 1, wherein the data set segmentation result graph is obtained by using a K-means clustering algorithm on the adjusted data set, and the method comprises the following specific steps:
and carrying out a K-means clustering algorithm with the K value of 2 on the adjusted data set to obtain a data set segmentation result graph.
7. The method for on-line monitoring of electrical equipment based on image processing according to claim 1, wherein the step of judging whether the circuit abnormality exists according to the data set segmentation result graph comprises the following specific steps:
when the electrical equipment circuit image is abnormal, the data set segmentation result graph is displayed in white, and when the electrical equipment circuit image is abnormal, the data set segmentation result graph is displayed in black, and whether the electrical equipment circuit image has the circuit abnormality is judged according to the data set segmentation result graph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310860813.7A CN116612138B (en) | 2023-07-14 | 2023-07-14 | Electrical equipment on-line monitoring method based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310860813.7A CN116612138B (en) | 2023-07-14 | 2023-07-14 | Electrical equipment on-line monitoring method based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116612138A true CN116612138A (en) | 2023-08-18 |
| CN116612138B CN116612138B (en) | 2023-09-19 |
Family
ID=87676763
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310860813.7A Active CN116612138B (en) | 2023-07-14 | 2023-07-14 | Electrical equipment on-line monitoring method based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116612138B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883411A (en) * | 2023-09-08 | 2023-10-13 | 浙江诺电电力科技有限公司 | Intelligent remote monitoring system for switch cabinet |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1091782A (en) * | 1996-09-10 | 1998-04-10 | Olympus Optical Co Ltd | Method for extracting specific site for gradation picture |
| US5982920A (en) * | 1997-01-08 | 1999-11-09 | Lockheed Martin Energy Research Corp. Oak Ridge National Laboratory | Automated defect spatial signature analysis for semiconductor manufacturing process |
| WO2007105107A2 (en) * | 2006-03-14 | 2007-09-20 | Agency For Science, Technology And Research | Methods, apparatus and computer-readable media for image segmentation |
| CN110264448A (en) * | 2019-06-06 | 2019-09-20 | 中南大学 | A kind of insulator breakdown detection method based on machine vision |
| CN111814711A (en) * | 2020-07-15 | 2020-10-23 | 中国矿业大学 | Image feature fast matching method and system applied to mine machine vision |
| WO2021068330A1 (en) * | 2019-10-12 | 2021-04-15 | 平安科技（深圳）有限公司 | Intelligent image segmentation and classification method and device and computer readable storage medium |
| CN115049835A (en) * | 2022-08-16 | 2022-09-13 | 众烁精密模架(南通)有限公司 | Data preprocessing method based on die-casting die defect identification |
-
2023
- 2023-07-14 CN CN202310860813.7A patent/CN116612138B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1091782A (en) * | 1996-09-10 | 1998-04-10 | Olympus Optical Co Ltd | Method for extracting specific site for gradation picture |
| US5982920A (en) * | 1997-01-08 | 1999-11-09 | Lockheed Martin Energy Research Corp. Oak Ridge National Laboratory | Automated defect spatial signature analysis for semiconductor manufacturing process |
| WO2007105107A2 (en) * | 2006-03-14 | 2007-09-20 | Agency For Science, Technology And Research | Methods, apparatus and computer-readable media for image segmentation |
| CN110264448A (en) * | 2019-06-06 | 2019-09-20 | 中南大学 | A kind of insulator breakdown detection method based on machine vision |
| WO2021068330A1 (en) * | 2019-10-12 | 2021-04-15 | 平安科技（深圳）有限公司 | Intelligent image segmentation and classification method and device and computer readable storage medium |
| CN111814711A (en) * | 2020-07-15 | 2020-10-23 | 中国矿业大学 | Image feature fast matching method and system applied to mine machine vision |
| CN115049835A (en) * | 2022-08-16 | 2022-09-13 | 众烁精密模架(南通)有限公司 | Data preprocessing method based on die-casting die defect identification |
Non-Patent Citations (3)
| Title |
|---|
| 严晨曦;熊凌;: "基于改进的Otsu方法的钢渣彩色图像分割", 计算机工程与设计, no. 09 * |
| 苑玮琦;姚双;: "基于骨架特征的柔性电路板缺陷检测方法", 计算机应用, no. 1 * |
| 陈放;杨艳;: "基于超像素和模糊聚类的医学超声图像分割算法", 半导体光电, no. 01 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883411A (en) * | 2023-09-08 | 2023-10-13 | 浙江诺电电力科技有限公司 | Intelligent remote monitoring system for switch cabinet |
| CN116883411B (en) * | 2023-09-08 | 2023-12-19 | 浙江诺电电力科技有限公司 | Intelligent remote monitoring system for switch cabinet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116612138B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116612138B (en) | Electrical equipment on-line monitoring method based on image processing | |
| US8374437B2 (en) | Information processing apparatus, information processing method, image identifying apparatus, image identifying method, and program | |
| US9311526B2 (en) | Image processing system and method of improving human face recognition | |
| CN115841493B (en) | Cable detection method based on image processing | |
| US20020051578A1 (en) | Method and apparatus for object recognition | |
| CN109738450B (en) | Method and device for detecting notebook keyboard | |
| CN113205063A (en) | Visual identification and positioning method for defects of power transmission conductor | |
| CN116721107B (en) | Intelligent monitoring system for cable production quality | |
| KR102336030B1 (en) | Electric vehicle charger fire detection and charger condition prediction system | |
| CN117274722B (en) | Intelligent detection method for distribution box based on infrared image | |
| CN114170552A (en) | Natural gas leakage real-time early warning method and system based on infrared thermal imaging | |
| CN114331997A (en) | Wire defect monitoring method based on image recognition | |
| CN115239626A (en) | Defect detection method, optical detection device, electronic device, and storage medium | |
| KR20210086303A (en) | Pattern inspection apparatus based on deep learning and inspection method using the same | |
| CN116543238B (en) | Image detection method for cable insulating layer | |
| US20230342903A1 (en) | Abnormality detection device, abnormality detection method, and abnormality detection system | |
| Zeng et al. | Automated extraction of PCB components based on specularity using layered illumination | |
| CN115830106B (en) | Auxiliary positioning method for electrified cleaning of equipment in machine room | |
| CN110633657B (en) | Blast furnace material collapse prediction method | |
| JP4267103B2 (en) | Appearance image classification apparatus and method | |
| CN107845080B (en) | Card image enhancement method | |
| WO2021217445A1 (en) | Image processing method, device and system, and storage medium | |
| CN109636797B (en) | Method for detecting deformation of non-contact flexible direct current converter valve module | |
| CN107092855A (en) | Vehicle part recognition methods and equipment, vehicle identification method and equipment | |
| CN117152421B (en) | Transmission line foreign matter detection method, device, computer equipment and storage medium |
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 |