CN115965623B - Surface flaw detection method and system in transformer production - Google Patents
Surface flaw detection method and system in transformer production Download PDFInfo
- Publication number
- CN115965623B CN115965623B CN202310251600.4A CN202310251600A CN115965623B CN 115965623 B CN115965623 B CN 115965623B CN 202310251600 A CN202310251600 A CN 202310251600A CN 115965623 B CN115965623 B CN 115965623B
- Authority
- CN
- China
- Prior art keywords
- flaw
- value
- detected area
- pixel point
- crack
- 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
Images
Abstract
The invention relates to the technical field of image data processing, in particular to a method and a system for detecting surface flaws in transformer production, wherein the method acquires surface images of all parts of a transformer bushing so as to obtain a flaw waiting area in the surface images; combining the values, gray values and corresponding edge distances of each pixel point in the flaw detection area in the RGB three channels to obtain azimuth gloss characteristic values of the flaw detection area; combining the fitting goodness, the corner number, the crack width and the first number of clusters corresponding to the flaw to-be-detected area to obtain the hot crack tortuosity of the flaw to-be-detected area; and combining the azimuth gloss characteristic value and the thermal cracking tortuosity of the flaw to-be-detected area to obtain thermal cracking significance, and detecting the surface flaws of the corresponding part according to the thermal cracking significance of each flaw to-be-detected area. The invention improves the accuracy and convenience of detecting the surface flaws of each component of the transformer bushing, and reduces the loss of manpower and resources.
Description
Technical Field
The invention relates to the technical field of image data processing, in particular to a method and a system for detecting surface flaws in transformer production.
Background
The transformer bushing is a current-carrying element of the transformer, is a main insulation device of outgoing lines of a transformer winding, and has the function of insulating between the outgoing lines and a transformer shell and simultaneously fixing the outgoing lines. In order to ensure normal operation of the transformer and durable use of the transformer bushing, prevent accidents such as ground short circuit and insulation fault of power equipment, the fully sealed vacuum inside the transformer bushing is ensured, so that the air tightness of the transformer bushing is required to be detected in time.
At present, the air tightness detection of the transformer bushing mainly depends on infrared detection, sound discrimination and electric experiment methods, wherein the infrared detection is mainly sensitive to temperature change, the sound discrimination is easily influenced by environment to cause insufficient accuracy, and the electric experiment method has more accurate results but complex process and consumes a large amount of manpower and material resources. These methods are mainly used to detect the surface quality and air tightness of the assembled transformer and are not applicable during assembly and welding. If each part of the transformer bushing has cracks on the surface after production and before bushing assembly, the quality of the transformer bushing assembled by the parts is poor, the service life is shortened, and the tightness is poor; if the inspection is performed after the completion of the assembly, a lot of time and raw materials are wasted, and thus a crack flaw detection method for a production and assembly process of a transformer bushing, which causes a reduction in service life, is required.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a method and a system for detecting surface flaws in transformer production, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for detecting a surface defect in transformer production, the method comprising:
collecting surface images of all parts of the transformer bushing, wherein the surface images are RGB images, and obtaining surface gray images corresponding to the RGB images; performing edge detection on the surface gray level image to obtain an area corresponding to the closed edge line as a flaw to-be-detected area;
obtaining the edge distance between any pixel point in the flaw to-be-detected area and the image edge corresponding to the surface image, and obtaining the azimuth gloss characteristic value of the flaw to-be-detected area by combining the value and gray value corresponding to each pixel point in the flaw to-be-detected area in three RGB channels and the corresponding edge distance;
performing straight line fitting on all pixel points in the flaw to-be-detected area to obtain a fitting goodness; performing corner detection on the surface gray level image, and counting the number of corner points in a flaw to-be-detected area; refining the flaw to-be-detected area to obtain a corresponding skeleton line, making a straight line for each pixel point on the skeleton line, and obtaining the crack width of the corresponding pixel point according to the intersection point of the straight line and the flaw to-be-detected area; clustering pixel points based on the crack width to obtain a first number of clusters, and obtaining thermal crack tortuosity of a flaw to-be-detected area by combining the fitting goodness, the number of corner points, the crack width and the first number of clusters;
combining the azimuth gloss characteristic value of the flaw to-be-detected area and the hot cracking tortuosity to obtain the hot cracking significance; and detecting the surface flaws of the corresponding component according to the hot crack significance of each flaw to-be-detected area.
Further, the method for obtaining the edge distance from any one pixel point in the flaw to-be-detected area to the image edge corresponding to the surface image includes:
and taking any pixel point in the flaw detection area as a target pixel point, respectively calculating the difference value between the corresponding Manhattan distance and the Chebyshev distance between the target pixel point and any pixel point on the image edge corresponding to the surface image based on the coordinates of the target pixel point, and taking the minimum difference value as the edge distance from the target pixel point to the image edge corresponding to the surface image.
Further, the method for acquiring the azimuth gloss characteristic value of the flaw to-be-detected area by combining the corresponding value, gray value and corresponding edge distance of each pixel point in the flaw to-be-detected area in three RGB channels comprises the following steps:
arranging the edge distances of each pixel point in the flaw detection area in a sequence from small to large to obtain a sequence, and acquiring the pixel points corresponding to the first M edge distances in the sequence as first pixel points; m is a positive integer;
calculating an average edge distance according to the edge distance of the first pixel point; the corresponding values of the three RGB channels are R value, B value and G value, for each first pixel point, a first difference value between the maximum value and the R value, a second difference value between the maximum value and the B value and a third difference value between the maximum value and the G value are calculated, the addition results of the first difference value, the second difference value and the third difference value are calculated, the addition results are molecules, the gray values of the corresponding first pixel points are denominators to obtain corresponding ratio values, and the ratio values of all the first pixel points are added to obtain ratio value summation;
and taking the product of the inverse of the average edge distance and the sum of the ratios as an azimuth gloss characteristic value of the flaw to-be-detected area.
Further, the method for obtaining the crack width of the corresponding pixel point according to the intersection point of the straight line and the flaw to-be-detected area by making the straight line for each pixel point on the skeleton line comprises the following steps:
numbering each pixel point on the skeleton line in sequence, wherein for the ith pixel point, i=1, 2 and 3 … N-1, N is the number of the pixel points on the skeleton line, and N is a positive integer; connecting the ith pixel point and the (i+1) th pixel point to obtain a straight line serving as the ith straight line, passing the ith pixel point to make a target straight line perpendicular to the ith straight line, obtaining two intersection points of the target straight line in a flaw to-be-detected area, and taking the Euclidean distance between the two intersection points as the crack width of the ith pixel point.
Further, the method for clustering pixel points based on the crack width to obtain a first number of clusters includes:
and constructing a plane rectangular coordinate system by taking the crack width of the pixel points corresponding to the numbers with the numbers as the abscissa and taking the crack width of the pixel points corresponding to the numbers as the ordinate, and clustering all the pixel points by using a DBSCAN algorithm according to each coordinate point in the plane rectangular coordinate system to obtain a first number of clusters.
Further, the method for obtaining the thermal cracking tortuosity of the flaw to-be-detected area by combining the goodness of fit, the number of corner points, the crack width and the first number of clusters comprises the following steps:
respectively acquiring a crack width average value, a maximum crack width and a minimum crack width based on all the crack widths; calculating a first difference absolute value between the maximum crack width and the minimum crack width; obtaining the absolute value of the difference between each crack width and the average value of the crack widths, and obtaining the average absolute value of the difference according to the absolute value of the difference corresponding to each crack width;
taking the product of the goodness of fit, the number of corner points, the first absolute difference value, the average absolute difference value and the first number of clusters as the thermal cracking tortuosity of the flaw to-be-detected area.
Further, the method for obtaining the thermal cracking significance by combining the azimuth gloss characteristic value and the thermal cracking tortuosity of the flaw to-be-detected area comprises the following steps:
and taking the product of the azimuth gloss characteristic value of the flaw to-be-detected area and the hot crack tortuosity as the hot crack significance.
Further, the method for detecting the surface flaws of the corresponding component according to the thermal cracking significance of each flaw to-be-detected area comprises the following steps:
normalizing the hot crack significance of each flaw to-be-detected area to obtain a corresponding hot crack significance normalization value, setting a normalization threshold value, and confirming that the corresponding flaw to-be-detected area is hot crack when the hot crack significance normalization value is greater than or equal to the normalization threshold value;
and setting a quantity threshold, counting the quantity of the flaw to-be-detected areas confirmed as hot cracks, and confirming that the hot crack flaws exist on the surface of the corresponding part when the quantity is larger than or equal to the quantity threshold.
In a second aspect, another embodiment of the present invention provides a surface flaw detection system in transformer production, the system comprising: a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of any one of the methods described above when executing the computer program.
The invention has the following beneficial effects: the method comprises the steps of acquiring surface images of all parts of a transformer bushing, and taking an area corresponding to a closed edge line as a flaw to-be-detected area through the edge line in the surface images based on the closed edge characteristics of the area corresponding to the hot cracks; because the flaw to-be-detected area is a position where cracks possibly affect the air tightness, in order to confirm whether the flaw to-be-detected area is hot cracks, the position where the flaw to-be-detected area is easily generated at the edge of a casting is known based on the reason that the hot cracks are generated, and then the edge distance from each pixel point in the flaw to the image edge corresponding to the surface image is combined, and the azimuth gloss characteristic value of the flaw to-be-detected area is obtained by combining the values and the gray values corresponding to three RGB channels of each pixel point; in order to further accurately confirm whether the flaw to-be-detected area is hot-cracked or not, based on characteristics of strip shape, trend bending distortion, uneven internal shape thickness and the like generated by the hot-cracked, linear fitting and thinning are carried out on the flaw to-be-detected area to obtain fitting goodness and skeleton lines, further, the first quantity of crack widths and clustered clusters is obtained according to the skeleton lines, the quantity of corner points of the flaw to-be-detected area is detected at the same time and used for representing bending phenomena, the hot-cracked bending degree of the flaw to-be-detected area is obtained by combining the fitting goodness, the quantity of corner points, the crack widths and the first quantity of clusters, the hot-cracked bending degree measures the bending degree and the uneven width of the hot-cracked corresponding to the flaw to-be-detected area, and the more the bending distortion and the uneven width are more likely to be the hot-cracked; the azimuth gloss characteristic value and the hot crack tortuosity of the flaw to-be-detected area are used as two characteristic indexes for judging whether hot cracks exist or not, so that the obtained hot crack significance is more rigorous and effective, the accuracy and convenience for detecting the surface flaws of the corresponding parts according to the hot crack significance are improved, and meanwhile, the loss of manpower and resources is reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart illustrating a method for detecting surface defects in transformer production according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a skeleton line of a flaw detection area in an embodiment of the present invention.
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 a method and a system for detecting surface flaws in transformer production according to the invention, which are described in detail below with reference to the accompanying drawings and preferred embodiments. 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 specific scene aimed by the invention is as follows: in the production process of each part of the transformer bushing, image acquisition is carried out on the forged part corresponding to each manufactured part, the image is processed and analyzed, and the crack part is identified so as to ensure the air tightness of the whole bushing and ensure that the inside of the bushing can be kept in vacuum in the working process.
The following specifically describes a specific scheme of a surface flaw detection method and a system in transformer production.
Referring to fig. 1, a flowchart of a method for detecting surface defects in transformer production according to an embodiment of the invention is shown, where the method includes:
s001, collecting surface images of all components of the transformer bushing, wherein the surface images are RGB images, and obtaining surface gray images corresponding to the RGB images; and carrying out edge detection on the surface gray level image to obtain an area corresponding to the closed edge line as a flaw to-be-detected area.
Specifically, a CCD camera is fixed above and to the side of the completed position of each forged part of the transformer bushing, an image of the surface of each part completed in production is acquired using the CCD camera, the acquired image is recorded as a surface image, and the surface image is an RGB image.
Noise inevitably occurs on the acquired surface image due to factors such as environment, slight vibration of internal elements of the camera, and in order to avoid the influence of the noise on the subsequent analysis result in the surface image, each channel in the surface image is respectively convolved by using Gaussian filtering, so that the surface image is denoised, and the accuracy and quality of the surface image are improved. The gaussian filtering denoising is a well-known technique, and the specific process is not described in detail.
The surface image is converted into a gray image, which is noted as a surface gray image. And carrying out edge detection on the surface gray level image by using a canny detection operator to obtain an edge image, wherein the edge image is a binary image, and the edge image is marked as a surface edge image. The canny detection operator is a known technology, and this scheme is not described in detail.
According to the edge lines in the surface edge image, the areas marked by each closed edge line are used as flaw to-be-detected areas, and the areas correspond to positions in the surface image, which are possibly crack to influence the air tightness, so that all flaw to-be-detected areas in the surface image are obtained according to all the closed edge lines.
Step S002, the edge distance between any one pixel point in the flaw detection area and the image edge corresponding to the surface image is obtained, and the azimuth gloss characteristic value of the flaw detection area is obtained by combining the value, the gray value and the corresponding edge distance of each pixel point in the flaw detection area corresponding to three RGB channels.
Specifically, the cause of occurrence of thermal cracks is known to be a characteristic that the thermal cracks easily occur at the edge of the casting and are oxidized and have no metallic luster. The hot cracking is a crack caused by the blocked solid shrinkage of the casting in the state that the strength and the plasticity of the casting are very low at the final stage of solidification or after solidification, and is often generated at the corner of the casting, the rapid change of the section thickness or the local solidification slow position, so that a flaw to-be-detected area corresponding to the hot cracking is also positioned at a position close to the edge of the casting.
Taking a flaw to-be-detected area as an example based on the position characteristics of the flaws, taking any pixel point in the flaw to-be-detected area as a target pixel point, respectively calculating the difference value between the corresponding Manhattan distance and the Chebyshev distance between the target pixel point and any pixel point on the image edge corresponding to the surface image based on the coordinates of the target pixel point, and taking the minimum difference value as the edge distance from the target pixel point to the image edge corresponding to the surface image.
As an example, any one pixel point A in the flaw detection area is taken, and the coordinates of the pixel point A are recorded as
Wherein, the method comprises the steps of, wherein,
is the abscissa of the pixel point,
Is the ordinate of the pixel point; and taking the distance difference between the Manhattan distance and the Chebyshev distance corresponding to each pixel point A on the edge of the surface image. The edge of the surface image refers to the peripheral boundary of the surface image, not the edge of each region in the surface image, that is, the image edge of the surface image, and the shortest distance difference among these distance differences is taken as the edge distance corresponding to the pixel point A
The method comprises the steps of carrying out a first treatment on the surface of the Each pixel point in the flaw detection area can obtain a corresponding edge distance
。
The surface of the thermal crack is oxidized, for example, the surface of the crack of the steel casting is approximately black, the aluminum alloy is dark gray and has no metallic luster, so that the gray value corresponding to each pixel point in the flaw detection area and the corresponding value under three channels of RGB are respectively marked as H, R, G and B based on the color characteristic. Because the gray value corresponding to the pixel point in the area corresponding to the thermal crack and the corresponding value under the three channels of RGB are smaller, the gray value corresponding to each pixel point in the flaw to-be-detected area, the corresponding value under the three channels of RGB and the corresponding edge distance are combined to obtain the azimuth gloss characteristic value of the flaw to-be-detected area, and the method for obtaining the azimuth gloss characteristic value of the flaw to-be-detected area comprises the following steps: arranging the edge distances of each pixel point in the flaw detection area in a sequence from small to large to obtain a sequence, and acquiring the pixel points corresponding to the first M edge distances in the sequence as first pixel points; m is a positive integer; calculating an average edge distance according to the edge distance of the first pixel point; the corresponding values of the three RGB channels are R value, B value and G value, for each first pixel point, a first difference value between the maximum value and the R value, a second difference value between the maximum value and the B value and a third difference value between the maximum value and the G value are calculated, the addition results of the first difference value, the second difference value and the third difference value are calculated, the addition results are molecules, the gray values of the corresponding first pixel points are denominators to obtain corresponding ratio values, and the ratio values of all the first pixel points are added to obtain ratio value summation; and taking the product of the inverse of the average edge distance and the sum of the ratios as an azimuth gloss characteristic value of the flaw to-be-detected area.
As an example, the embodiment of the invention arranges the edge distance of each pixel point in the flaw to-be-detected area in order from small to large to obtain a sequence, and obtains the pixel point corresponding to the first 50% of the edge distance in the sequence as the first pixel point, so as to construct a calculation formula of the azimuth gloss characteristic value of the flaw to-be-detected area according to the gray value corresponding to the first pixel point, the corresponding values under three channels of RGB and the corresponding edge distance:
wherein, the liquid crystal display device comprises a liquid crystal display device,
the azimuth gloss characteristic value of the flaw to-be-detected area;
the edge distance of the ith first pixel point;
the number of the first pixel points;
the value of the ith first pixel point in the R channel is the value of the ith first pixel point in the R channel;
the ith first pixel point is in G-passThe value of the track;
the value of the ith first pixel point in the B channel;
the gray value of the ith first pixel point; 255 is a maximum.
It should be noted that, the edge distance corresponding to each pixel point in the flaw to-be-detected area is smaller because the flaw to-be-detected area is located at a position closer to the edge of the casting, when the crack grows larger, the partial area of the crack may extend to the middle of the casting, in order to avoid the influence of the larger edge distance corresponding to the pixel points contained in the partial areas on the calculated value, the smaller edge distance in the edge distance corresponding to each pixel point in the flaw to-be-detected area is selected to calculate, the average value of the smaller edge distance is utilized to represent the position of the flaw corresponding to the flaw to-be-detected area, the larger the average value is, the less likely the flaw to be the heat crack, and the azimuth gloss characteristic value corresponding to the flaw to-be-detected area
The smaller; when the color corresponding to each pixel point in the flaw detection area is darker, the corresponding gray value is smaller, and the pixel point is more likely to be hot-cracked, so that the azimuth gloss characteristic value of the flaw detection area
The larger the defect to be detected is, the more likely the defect to correspond to the hot crack position.
Step S003, performing straight line fitting on all pixel points in the flaw to-be-detected area to obtain a fitting goodness; performing corner detection on the surface gray level image, and counting the number of corner points in the flaw to-be-detected area; refining the flaw to-be-detected area to obtain a corresponding skeleton line, making a straight line for each pixel point on the skeleton line, and obtaining the crack width of the corresponding pixel point according to the intersection point of the straight line and the flaw to-be-detected area; and clustering the pixel points based on the crack width to obtain a first number of clusters, and obtaining the thermal crack tortuosity of the flaw to-be-detected area by combining the fitting goodness, the number of corner points, the crack width and the first number of clusters.
Specifically, since the thermal cracking is entirely stripe-shaped, and is initiated at and spread along the grain boundary, the trend is tortuous and discontinuous, and the internal shape is uneven in thickness, the flaw to-be-detected area is analyzed according to the characteristics.
Performing straight line fitting on all pixel points contained in the flaw to-be-detected area by using a least square method to obtain corresponding fitting goodness
The goodness of fit refers to the degree of fit between the position distribution of all the pixel points contained in the flaw to-be-detected area and the fitting straight line, and the larger the degree of fit is, the more similar the flaw to-be-detected area is to the strip shape.
Because the trend of the thermal cracks is curved, the edge of the corresponding flaw to-be-detected area is curved, and the number of corner points is more, the corner point detection is carried out on the surface gray level image, and the number of the corner points contained in the flaw to-be-detected area is counted as
。
Image refinement is performed on the flaw to-be-detected area by using a Hilditch algorithm, so as to obtain a skeleton line of the flaw to-be-detected area, as shown in fig. 2, wherein the Hilditch algorithm is a known technology, and the scheme is not repeated. Numbering each pixel point on the skeleton line from one section to the other end of the skeleton line in sequence, wherein for the ith pixel point, i=1, 2, 3 … N-1, N is the number of the pixel points on the skeleton line, and N is a positive integer; connecting the ith pixel point and the (i+1) th pixel point to obtain a straight line serving as the ith straight line, passing the ith pixel point to make a target straight line perpendicular to the ith straight line, obtaining two intersection points of the target straight line in a flaw to-be-detected area, and taking the Euclidean distance between the two intersection points as the crack width of the ith pixel point.
As an example, from one section to the other end of the skeleton line, numbering each pixel point on the skeleton line one by one from the number 1, connecting the pixel points corresponding to the number 1 and the number 2 into a straight line as a first straight line corresponding to the pixel point of the number 1, and passing through the image of the number 1The element point is a straight line perpendicular to the first straight line and is marked as a straight line
The method comprises the steps of carrying out a first treatment on the surface of the Straight line
Two intersection points are arranged between the pixel point and the flaw to-be-detected area, and the Euclidean distance between the two intersection points is calculated to be used as the crack width C corresponding to the pixel point with the number of 1.
It should be noted that, since the crack width of each pixel point in the flaw detection area is obtained based on the straight line between the adjacent pixel points, the crack width does not exist for the last pixel point on the skeleton line, that is, the crack width of the last pixel point on the skeleton line is not calculated in the invention.
Because the trend of the hot cracks is tortuous and twisted, the widths of adjacent and nearer cracks corresponding to small parts with the same trend on the hot cracks are relatively similar and correspond to the same cluster, the widths of adjacent and nearer cracks at the position where the direction of the hot cracks changes correspond to the new cluster, so that the number of clusters of a flaw to be detected area corresponding to the hot cracks is relatively large, a plane rectangular coordinate system is constructed by taking the crack width of the pixel point corresponding to the serial number as the abscissa and taking the crack width of the pixel point corresponding to the serial number as the ordinate, and all the pixel points are clustered by using a DBSCAN algorithm according to each coordinate point in the plane rectangular coordinate system to obtain a first number of clusters, and the method comprises the following steps: obtaining a coordinate point of each pixel point on the plane rectangular coordinate system by taking the serial number of the pixel point on the skeleton line as an abscissa and the crack width corresponding to the pixel point as an ordinate
Using DBSCAN algorithm to obtain each pixel point corresponding to the coordinate points
And clustering, wherein the DBSCAN algorithm uses 5 as the minimum point number and uses 2 as the neighborhood radius.
Combining the fitting goodness R and the corner number
First number of clusters, crack width C
The hot crack tortuosity of the flaw to-be-detected area is obtained, specifically: respectively acquiring a crack width average value, a maximum crack width and a minimum crack width based on all the crack widths; calculating a first difference absolute value between the maximum crack width and the minimum crack width; obtaining the absolute value of the difference between each crack width and the average value of the crack widths, and obtaining the average absolute value of the difference according to the absolute value of the difference corresponding to each crack width; taking the product of the goodness of fit, the number of corner points, the first absolute difference value, the average absolute difference value and the first number of clusters as the thermal cracking tortuosity of the flaw to-be-detected area.
As an example, the thermal crack tortuosity of the flaw to-be-detected area is calculated as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,
is thermal crack tortuosity;
is the number of crack widths;
is the ith crack width;
is the average value of the width of the crack;
is the maximum crack width;
is the minimum crack width;
the number of corner points in the flaw to-be-detected area is the number of corner points in the flaw to-be-detected area;
a first number of clusters;
fitting goodness of a flaw to-be-detected area;
is the mean difference absolute value.
It should be noted that, the internal shape of the thermal crack is uneven, and the crack width difference corresponding to each pixel point on the skeleton line of the to-be-detected area corresponding to the flaw is larger, so the difference between the maximum crack width and the minimum crack width is larger
The larger the more likely it is to be a thermal crack, the greater the corresponding thermal crack tortuosity; difference between crack width and crack width mean
The larger the average difference
The larger the crack width is, the more disordered the crack width is, the more the crack to-be-detected area is possibly hot cracking, and the larger the corresponding hot cracking tortuosity is; number of corner points in flaw detection area
The more the edges of the flaw to-be-detected area are, the more the edges are bent, the more the edges are likely to be hot cracks, and the larger the bending degree of the corresponding hot cracks is; first number of clusters
The more the defects are, the more the trend bending torsion in the to-be-detected area is described, the more the defects are likely to be hot cracks, and the larger the corresponding hot crack bending degree is; as the thermal cracks are in the shape of a strip, different cracks are discontinuous, the larger the fitting goodness R of the corresponding flaw to-be-detected area is, the more similar the flaw to-be-detected area is to the strip, and the larger the corresponding thermal crack tortuosity is. The tortuosity of the hot cracks measures the tortuosity and the width of the hot cracks corresponding to the area to be detected of the defectsThe non-uniformity is that the more the bending and twisting are performed, the more the width is non-uniform, and the corresponding hot crack bending degree is larger, so that the hot crack bending degree is in positive correlation with the fitting goodness of a flaw to-be-detected area, the number of corner points of the flaw to-be-detected area, the absolute value of the first difference value of the maximum crack width and the minimum crack width, the absolute value of the average difference value of the crack widths and the first number of clusters respectively.
Step S004, combining the azimuth gloss characteristic value of the flaw to-be-detected area and the hot cracking tortuosity to obtain the hot cracking significance; and detecting the surface flaws of the corresponding component according to the hot crack significance of each flaw to-be-detected area.
Specifically, since thermal cracking is liable to occur at the position close to the edge of the casting and is characterized by being oxidized and free of metallic luster, and the thermal cracking is generally characterized by a meandering and uneven width, thermal cracking significance is obtained by combining the azimuth luster characteristic value and thermal cracking tortuosity of the flaw to-be-detected area, and each flaw to-be-detected area is evaluated by using the thermal cracking significance.
As an example, the azimuth gloss characteristic value of the flaw to-be-detected area
And thermal crack tortuosity
Is the hot crack significance, then
The calculation formula of (2) is as follows:
。
it should be noted that, the larger the azimuth gloss characteristic value is, the more likely the flaw to be detected is a thermal crack, and the larger the thermal crack tortuosity is, the more likely the flaw to be detected is a thermal crack, so that when the azimuth gloss characteristic value corresponding to the flaw to be detected is
And thermal crack tortuosity
The larger the thermal crack, the more likely it is, its corresponding thermal crack significance
The larger the azimuth gloss characteristic value and the hot crack tortuosity are, the positive correlation is formed between the azimuth gloss characteristic value and the hot crack tortuosity and the hot crack saliency.
The method comprises the steps of obtaining the hot crack significance of each flaw to-be-detected area by using a hot crack significance obtaining method, normalizing the hot crack significance of each flaw to-be-detected area to obtain a corresponding hot crack significance normalization value, setting a normalization threshold, and confirming that the corresponding flaw to-be-detected area is hot crack when the hot crack significance normalization value is greater than or equal to the normalization threshold; and setting a quantity threshold, counting the quantity of the flaw to-be-detected areas confirmed as hot cracks, and confirming that the hot crack flaws exist on the surface of the corresponding part when the quantity is larger than or equal to the quantity threshold.
As an example, a normalization threshold is set to 0.6, and when the thermal cracking significance normalization value corresponding to the flaw to-be-detected area is greater than or equal to 0.6, the flaw to-be-detected area is confirmed to be thermal cracking, otherwise, the flaw to-be-detected area is not thermal cracking; and (3) carrying out hot crack confirmation on each flaw to-be-detected area, counting the number of flaw to-be-detected areas confirmed to be hot cracks, setting the number threshold value to be 1, and confirming that the surface of the corresponding part has hot crack defects when one flaw to-be-detected area confirms to be hot cracks, so that the air tightness is influenced by the existence of the hot crack defects of the part.
In summary, the embodiment of the invention acquires the surface images of each component of the transformer bushing to obtain the corresponding surface gray level image; performing edge detection on the surface gray level image to obtain an area corresponding to the closed edge line as a flaw to-be-detected area; obtaining the edge distance between any pixel point in the flaw to-be-detected area and the image edge corresponding to the surface image, and obtaining the azimuth gloss characteristic value of the flaw to-be-detected area by combining the color characteristic of each pixel point in the flaw to-be-detected area and the corresponding edge distance; performing straight line fitting on all pixel points in the flaw to-be-detected area to obtain a fitting goodness; performing corner detection on the surface gray level image, and counting the number of corner points in a flaw to-be-detected area; refining the flaw to-be-detected area to obtain a corresponding skeleton line, making a straight line for each pixel point on the skeleton line, and obtaining the crack width of the corresponding pixel point according to the intersection point of the straight line and the flaw to-be-detected area; clustering pixel points based on the crack width to obtain a first number of clusters, and obtaining thermal crack tortuosity of a flaw to-be-detected area by combining the fitting goodness, the number of corner points, the crack width and the first number of clusters; combining the azimuth gloss characteristic value of the flaw to-be-detected area and the hot cracking tortuosity to obtain the hot cracking significance; and detecting the surface flaws of the corresponding component according to the hot crack significance of each flaw to-be-detected area. The invention improves the accuracy and convenience of detecting the surface flaws of the corresponding parts according to the hot crack significance, and reduces the loss of manpower and resources.
Based on the same inventive concept as the above method embodiment, the embodiment of the present invention further provides a surface flaw detection system in transformer production, including: the method comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps in the embodiment of the surface flaw detection method in the transformer production, for example, the steps shown in fig. 1. The method for detecting surface flaws in the production of the transformer is described in detail in the above embodiments, and will not be described again.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (6)
1. The surface flaw detection method in transformer production is characterized by comprising the following steps of:
collecting surface images of all parts of the transformer bushing, wherein the surface images are RGB images, and obtaining surface gray images corresponding to the RGB images; performing edge detection on the surface gray level image to obtain an area corresponding to the closed edge line as a flaw to-be-detected area;
obtaining the edge distance between any pixel point in the flaw to-be-detected area and the image edge corresponding to the surface image, and obtaining the azimuth gloss characteristic value of the flaw to-be-detected area by combining the value and gray value corresponding to each pixel point in the flaw to-be-detected area in three RGB channels and the corresponding edge distance;
performing straight line fitting on all pixel points in the flaw to-be-detected area to obtain a fitting goodness; performing corner detection on the surface gray level image, and counting the number of corner points in a flaw to-be-detected area; refining the flaw to-be-detected area to obtain a corresponding skeleton line, making a straight line for each pixel point on the skeleton line, and obtaining the crack width of the corresponding pixel point according to the intersection point of the straight line and the flaw to-be-detected area; clustering pixel points based on the crack width to obtain a first number of clusters, and obtaining thermal crack tortuosity of a flaw to-be-detected area by combining the fitting goodness, the number of corner points, the crack width and the first number of clusters;
combining the azimuth gloss characteristic value of the flaw to-be-detected area and the hot cracking tortuosity to obtain the hot cracking significance; detecting surface flaws of the corresponding component according to the hot crack significance of each flaw to-be-detected area;
the method for acquiring the azimuth gloss characteristic value of the flaw to-be-detected area by combining the values, gray values and corresponding edge distances corresponding to the RGB three channels of each pixel point in the flaw to-be-detected area comprises the following steps:
arranging the edge distances of each pixel point in the flaw detection area in a sequence from small to large to obtain a sequence, and acquiring the pixel points corresponding to the first M edge distances in the sequence as first pixel points; m is a positive integer;
calculating an average edge distance according to the edge distance of the first pixel point; the corresponding values of the three RGB channels are R value, B value and G value, for each first pixel point, a first difference value between the maximum value and the R value, a second difference value between the maximum value and the B value and a third difference value between the maximum value and the G value are calculated, the addition results of the first difference value, the second difference value and the third difference value are calculated, the addition results are molecules, the gray values of the corresponding first pixel points are denominators to obtain corresponding ratio values, and the ratio values of all the first pixel points are added to obtain ratio value summation;
taking the product of the reciprocal of the average edge distance and the sum of the ratios as an azimuth gloss characteristic value of the flaw to-be-detected area;
the method for obtaining the thermal cracking tortuosity of the flaw to-be-detected area by combining the goodness of fit, the number of corner points, the crack width and the first number of clusters comprises the following steps:
respectively acquiring a crack width average value, a maximum crack width and a minimum crack width based on all the crack widths; calculating a first difference absolute value between the maximum crack width and the minimum crack width; obtaining the absolute value of the difference between each crack width and the average value of the crack widths, and obtaining the average absolute value of the difference according to the absolute value of the difference corresponding to each crack width;
taking the product of the goodness of fit, the number of corner points, the absolute value of the first difference value, the absolute value of the average difference value and the first number of clusters as the thermal cracking tortuosity of the flaw to-be-detected area;
the method for obtaining the crack width of the corresponding pixel point according to the intersection point of the straight line and the flaw to-be-detected area comprises the following steps:
numbering each pixel point on the skeleton line in sequence, wherein for the ith pixel point, i=1, 2 and 3 … N-1, N is the number of the pixel points on the skeleton line, and N is a positive integer; connecting the ith pixel point and the (i+1) th pixel point to obtain a straight line serving as the ith straight line, passing the ith pixel point to make a target straight line perpendicular to the ith straight line, obtaining two intersection points of the target straight line in a flaw to-be-detected area, and taking the Euclidean distance between the two intersection points as the crack width of the ith pixel point.
2. The method for detecting surface flaws in transformer production according to claim 1, wherein the method for obtaining the edge distance from any one pixel point in the flaw detection area to the image edge corresponding to the surface image comprises the following steps:
and (3) acquiring any pixel point in the flaw to-be-detected area as a target pixel point, respectively calculating the difference value of the corresponding Manhattan distance and the Chebyshev distance between the target pixel point and any pixel point on the image edge corresponding to the surface image based on the coordinates of the target pixel point, and taking the minimum difference value as the edge distance from the target pixel point to the image edge corresponding to the surface image.
3. The method for detecting surface flaws in transformer production as claimed in claim 1, wherein the clustering of pixels based on the crack width is performed to obtain a first number of clusters, comprising:
and constructing a plane rectangular coordinate system by taking the crack width of the pixel points corresponding to the numbers with the numbers as the abscissa and taking the crack width of the pixel points corresponding to the numbers as the ordinate, and clustering all the pixel points by using a DBSCAN algorithm according to each coordinate point in the plane rectangular coordinate system to obtain a first number of clusters.
4. The method for detecting surface flaws in transformer production according to claim 1, wherein the method for obtaining thermal cracking significance by combining azimuth gloss characteristic values and thermal cracking tortuosity of flaw to-be-detected areas comprises the following steps:
and taking the product of the azimuth gloss characteristic value of the flaw to-be-detected area and the hot crack tortuosity as the hot crack significance.
5. The method for detecting surface flaws in transformer production as claimed in claim 1, wherein the method for detecting surface flaws of a corresponding component based on thermal cracking significance level of each flaw to be detected comprises:
normalizing the hot crack significance of each flaw to-be-detected area to obtain a corresponding hot crack significance normalization value, setting a normalization threshold value, and confirming that the corresponding flaw to-be-detected area is hot crack when the hot crack significance normalization value is greater than or equal to the normalization threshold value;
and setting a quantity threshold, counting the quantity of the flaw to-be-detected areas confirmed as hot cracks, and confirming that the hot crack flaws exist on the surface of the corresponding part when the quantity is larger than or equal to the quantity threshold.
6. A surface flaw detection system in transformer production, the system comprising: a memory, a processor and a computer program stored in said memory and executable on the processor, the processor implementing the steps of a method for detecting surface flaws in the production of a transformer according to any one of the preceding claims 1-5 when said computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310251600.4A CN115965623B (en) | 2023-03-16 | 2023-03-16 | Surface flaw detection method and system in transformer production |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310251600.4A CN115965623B (en) | 2023-03-16 | 2023-03-16 | Surface flaw detection method and system in transformer production |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115965623A CN115965623A (en) | 2023-04-14 |
| CN115965623B true CN115965623B (en) | 2023-05-30 |
Family
ID=85899689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310251600.4A Active CN115965623B (en) | 2023-03-16 | 2023-03-16 | Surface flaw detection method and system in transformer production |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115965623B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116309550B (en) * | 2023-05-11 | 2023-08-04 | 聊城市飓风工业设计有限公司 | Integrated circuit patch abnormality identification method based on image processing |
| CN116258713B (en) * | 2023-05-11 | 2023-07-21 | 青岛穗禾信达金属制品有限公司 | Welding processing detection method for metal cabinet |
| CN116385434B (en) * | 2023-06-02 | 2023-08-08 | 同济检测(济宁)有限公司 | Intelligent detection method for precast beam cracks |
| CN117237646B (en) * | 2023-11-15 | 2024-01-30 | 深圳市润海电子有限公司 | PET high-temperature flame-retardant adhesive tape flaw extraction method and system based on image segmentation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105957088A (en) * | 2016-05-12 | 2016-09-21 | 广东电网有限责任公司云浮供电局 | Method and system for monitoring transformer composite insulator sleeve based on computer vision |
| CN115063423A (en) * | 2022-08-18 | 2022-09-16 | 启东市嘉信精密机械有限公司 | Mechanical casting cold and hot crack self-adaptive identification method based on computer vision |
| CN115222733A (en) * | 2022-09-19 | 2022-10-21 | 山东金源不锈钢制品有限公司 | Metal component surface defect detection method based on image recognition |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102011076631B4 (en) * | 2011-05-27 | 2016-01-07 | Airbus Operations Gmbh | METHOD FOR DETECTING ANY FAILURE OF ERROR IN A WHIRLPOOL WELD SEAM |
-
2023
- 2023-03-16 CN CN202310251600.4A patent/CN115965623B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105957088A (en) * | 2016-05-12 | 2016-09-21 | 广东电网有限责任公司云浮供电局 | Method and system for monitoring transformer composite insulator sleeve based on computer vision |
| CN115063423A (en) * | 2022-08-18 | 2022-09-16 | 启东市嘉信精密机械有限公司 | Mechanical casting cold and hot crack self-adaptive identification method based on computer vision |
| CN115222733A (en) * | 2022-09-19 | 2022-10-21 | 山东金源不锈钢制品有限公司 | Metal component surface defect detection method based on image recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115965623A (en) | 2023-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115965623B (en) | Surface flaw detection method and system in transformer production | |
| CN106404793B (en) | Bearing sealing element defect detection method based on vision | |
| CN102156996B (en) | Image edge detection method | |
| CN114882044B (en) | Metal pipe surface quality detection method | |
| CN115330645B (en) | Welding image enhancement method | |
| CN116109644A (en) | Surface defect detection method for copper-aluminum transfer bar | |
| CN115841434A (en) | Infrared image enhancement method for gas concentration analysis | |
| CN116721107B (en) | Intelligent monitoring system for cable production quality | |
| CN115063430B (en) | Electric pipeline crack detection method based on image processing | |
| CN115937216A (en) | Magnetic rotor appearance quality detection method for new energy automobile | |
| CN116168037B (en) | Method and system for calculating bending degree of wire crimping based on image processing | |
| CN115457063A (en) | Method, device and equipment for extracting edge of circular hole of PCB (printed Circuit Board) and storage medium | |
| CN114332081B (en) | Textile surface abnormity determination method based on image processing | |
| CN115272336A (en) | Metal part defect accurate detection method based on gradient vector | |
| CN107230212B (en) | Vision-based mobile phone size measuring method and system | |
| CN115330646A (en) | Metal component inspection method for fitness equipment paint spraying process | |
| CN115274486A (en) | Semiconductor surface defect identification method | |
| CN111638218A (en) | Method for detecting surface defects of coating | |
| CN111489389A (en) | Light spot center detection method | |
| WO2024016686A1 (en) | Corner detection method and apparatus | |
| CN116612112A (en) | Visual inspection method for surface defects of bucket | |
| CN115830027B (en) | Machine vision-based automobile wire harness cladding defect detection method | |
| CN114565563B (en) | Color steel plate surface abnormity detection method based on artificial intelligence | |
| CN114219802B (en) | Skin connecting hole position detection method based on image processing | |
| CN115797353B (en) | Intelligent detection system and method for quality of cold-rolled strip steel |
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 | ||
| CP03 | Change of name, title or address |
Address after: 518000 room 1101-1105, building 6, Hengda fashion Huigu building (East District), Fulong Road, Henglang community, Dalang street, Longhua District, Shenzhen, Guangdong Patentee after: Cenke Technology (Shenzhen) Group Co.,Ltd. Address before: Room 1101-1105, Building 6, Evergrande Fashion Huigu Building (East Area), Fulong Road, Henglang Community, Dalang Street, Longhua District, Shenzhen, Guangdong 518110 Patentee before: SHENZHEN CENKER ENTERPRISE Ltd. |
|
| CP03 | Change of name, title or address |