CN116385476B - Iron tower quality analysis method based on visual detection - Google Patents
Iron tower quality analysis method based on visual detection Download PDFInfo
- Publication number
- CN116385476B CN116385476B CN202310650678.3A CN202310650678A CN116385476B CN 116385476 B CN116385476 B CN 116385476B CN 202310650678 A CN202310650678 A CN 202310650678A CN 116385476 B CN116385476 B CN 116385476B
- Authority
- CN
- China
- Prior art keywords
- pixel
- area
- growth
- welding seam
- weld
- 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
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 title claims abstract description 52
- 229910052742 iron Inorganic materials 0.000 title claims abstract description 26
- 238000001514 detection method Methods 0.000 title claims abstract description 24
- 238000004458 analytical method Methods 0.000 title claims abstract description 20
- 230000000007 visual effect Effects 0.000 title claims abstract description 13
- 238000003466 welding Methods 0.000 claims abstract description 189
- 238000000034 method Methods 0.000 claims abstract description 26
- 230000008569 process Effects 0.000 claims abstract description 16
- 230000007547 defect Effects 0.000 claims description 44
- 229910000679 solder Inorganic materials 0.000 claims description 17
- 239000002184 metal Substances 0.000 claims description 13
- 229910052751 metal Inorganic materials 0.000 claims description 13
- 230000005540 biological transmission Effects 0.000 claims description 11
- 230000002950 deficient Effects 0.000 claims description 9
- 238000011179 visual inspection Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000011282 treatment Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims 3
- 230000000694 effects Effects 0.000 abstract description 13
- 230000003044 adaptive effect Effects 0.000 abstract description 4
- 230000008859 change Effects 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 239000011324 bead Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004907 flux Effects 0.000 description 2
- 238000003754 machining Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 238000003801 milling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000005476 soldering 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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/10004—Still image; Photographic 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
- G06T2207/20112—Image segmentation details
- G06T2207/20156—Automatic seed setting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30136—Metal
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The application relates to the technical field of visual detection, in particular to a visual detection-based iron tower quality analysis method, which is characterized in that an area growth algorithm is improved, initial seed points are adaptively selected based on characteristics of a heat affected area, and the initial seed points are screened to effectively avoid the phenomenon that the growth is stopped when the growth times are smaller and an accurate heat affected area cannot be obtained, and simultaneously ensure the effect of adaptively adjusting a growth threshold value according to the characteristics of pixel points of the grown area. And the growth threshold is adaptively adjusted based on the pixel point characteristics of the grown region and the growth characteristics of the heat affected region, so that the local adaptive growth threshold is obtained, the acquisition accuracy of the heat affected region is improved, the acquisition accuracy of the welding seam region is further improved, the quality analysis of the welding seam is completed based on the texture characteristic difference of the welding seam region and the growth process difference during the acquisition of the welding seam region, and the quality detection accuracy of the welding seam region is improved.
Description
Technical Field
The application relates to the technical field of visual detection, in particular to a visual detection-based iron tower quality analysis method.
Background
Along with the continuous increase of the power demand, the high-capacity and high-voltage-class power transmission line is rapidly developed, so that the design load of the tower is larger and larger, and the structural design of the power transmission line tower tends to be large and complex. In the iron tower machining process, the welding workload accounts for 60% -65% of the whole foundation iron tower machining workload. Welding is a procedure in the processing of the power transmission iron tower, and the quality of the welding directly influences the processing quality of the whole-base iron tower. In order to ensure that the overall quality of the processing of the power transmission iron tower meets the design requirement, welding defects in the processing process of the iron tower are required to be detected, and corresponding repair welding and other treatments are required to be carried out.
Since the welding process is a high temperature process, in the welding process, the metal in the regions on both sides of the weld is subjected to a high temperature to change the surface thereof, and the affected metal region is called a heat affected zone. When the weld joint is subjected to defect detection, when a canny operator is used for extracting the weld joint region, the heat affected region and the weld joint region are often extracted together, so that the defect detection effect of the subsequent weld joint region is interfered.
When the region growing algorithm is used for extracting the weld joint region, the gray values of the metal surfaces at different positions are different due to the fact that the heat influence effects to the metal surfaces at different positions are different when the region growing algorithm is used for carrying out region growing, the gray non-uniformity phenomenon is extremely easy to occur, and when the preset fixed threshold value is used for carrying out the region growing, the growth effect is poor, the obtained heat affected region is inaccurate, and further the acquisition accuracy of the actual weld joint region is low.
Disclosure of Invention
Based on the above, it is necessary to extract the weld joint region by using a region growing algorithm, and when the region is grown, the gray values of the metal surfaces at different positions are different due to different heat influence effects received by the metal surfaces at different positions, so that the gray non-uniformity phenomenon is very easy to occur, and when the preset fixed threshold is used for growth, the growth effect is poor, namely the obtained heat affected region is inaccurate, and further, the acquisition accuracy of the actual weld joint region is not high.
The application provides a visual detection-based iron tower quality analysis method, which comprises the following steps:
acquiring a welding image of a tower body component of the power transmission tower, and preprocessing the welding image to obtain a welding gray level image of the tower body component;
processing the welding gray level image of the tower member to obtain a preliminary welding seam area, dividing the preliminary welding seam area into a plurality of welding seam pixel blocks, and using an area growth algorithm based on self-adaptive adjustment of a local growth threshold value for each welding seam pixel block to divide each welding seam pixel block into a heat affected area and an actual welding seam area;
and extracting a defect area from the actual weld area of each weld pixel block, and finishing the quality detection of the weld according to the characteristics of the defect area.
The application relates to a visual detection-based iron tower quality analysis method, which is characterized in that an area growth algorithm is improved, initial seed points are adaptively selected based on heat affected area characteristics, the initial seed points are screened, the phenomenon that growth is stopped when the number of growth times is small, an accurate heat affected area cannot be obtained is effectively avoided, and meanwhile, the effect of adaptively adjusting a growth threshold according to the characteristics of pixel points in a grown area can be ensured. And the growth threshold is adaptively adjusted based on the pixel point characteristics of the grown region and the growth characteristics of the heat affected region, so that the local adaptive growth threshold is obtained, the acquisition accuracy of the heat affected region is improved, the acquisition accuracy of the welding seam region is further improved, the quality analysis of the welding seam is completed based on the texture characteristic difference of the welding seam region and the growth process difference during the acquisition of the welding seam region, and the quality detection accuracy of the welding seam region is improved.
Drawings
Fig. 1 is a schematic flow chart of a method for analyzing quality of iron tower based on visual inspection according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a weld joint pixel block after being segmented in the visual inspection-based iron tower quality analysis method according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a minimum euclidean distance between a pixel point q and a pixel point at an edge of a preliminary welding seam region in a welding seam pixel block in an iron tower quality analysis method based on visual detection according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The application provides a visual detection-based iron tower quality analysis method. It should be noted that the iron tower quality analysis method based on visual detection provided by the application is applied to any kind of numerical control milling and drilling machine.
As shown in fig. 1, in an embodiment of the present application, the iron tower quality analysis method based on visual inspection includes the following steps S100 to S300:
s100, acquiring welding images of the tower body components of the power transmission tower, and preprocessing the welding images to obtain welding gray images of the tower body components.
And S200, processing the welding gray level image of the tower member to obtain a preliminary welding seam region, dividing the preliminary welding seam region into a plurality of welding seam pixel blocks, and using a region growing algorithm based on self-adaptive adjustment of a local growing threshold value for each welding seam pixel block to divide each welding seam pixel block into a heat affected region and an actual welding seam region.
S300, extracting a defect area from the actual welding seam area of each welding seam pixel block, and finishing welding seam quality detection according to the characteristics of the defect area.
Specifically, the pretreatment method of S100 may be various.
The application relates to a visual detection-based iron tower quality analysis method, which is characterized in that an area growth algorithm is improved, initial seed points are adaptively selected based on heat affected area characteristics, the initial seed points are screened, the phenomenon that growth is stopped when the number of growth times is small, an accurate heat affected area cannot be obtained is effectively avoided, and meanwhile, the effect of adaptively adjusting a growth threshold according to the characteristics of pixel points in a grown area can be ensured. And the growth threshold is adaptively adjusted based on the pixel point characteristics of the grown region and the growth characteristics of the heat affected region, so that the local adaptive growth threshold is obtained, the acquisition accuracy of the heat affected region is improved, the acquisition accuracy of the welding seam region is further improved, the quality analysis of the welding seam is completed based on the texture characteristic difference of the welding seam region and the growth process difference during the acquisition of the welding seam region, and the quality detection accuracy of the welding seam region is improved.
In an embodiment of the present application, the S100 includes the following S110 and S120:
s110, acquiring an image of a welded power transmission tower body component, and recording the image as a welding image.
And S120, carrying out graying treatment on the welding image by using a weighted graying method to obtain a tower member welding gray image.
Specifically, the acquired images of the welded power transmission tower body components are RGB images, and diffuse reflection light sources with uniform illumination are selected for ensuring imaging quality.
Graying means that in the RGB model, if r=g=b, the color represents a gray color, where the value of r=g=b is called a gray value, so that only one byte is needed for each pixel of the gray image to store the gray value (also called intensity value, brightness value), and the gray value ranges from 0 to 255.
In the embodiment, the acquisition of welding images of the iron tower components is completed through cooperation of the industrial camera and the light source, after the iron tower is welded, the light source is aligned to a welding position, a plurality of images are acquired through the industrial camera, the plurality of images are spliced into one welding image, and then the welding image is subjected to gray-scale treatment to obtain a welding gray-scale image.
In an embodiment of the present application, the S200 includes:
and S210, processing the welding gray level image by using canny operator detection to obtain a preliminary welding seam area.
S220, dividing the preliminary weld joint area into N weld joint pixel blocks with equal lengths.
S230, dividing each welding seam pixel block into a heat affected zone and an actual welding seam zone.
Specifically, the empirical value of N is 100.
As shown in FIG. 2, the length of the welding gray scale image is the length of the preliminary weld zoneThe width of the welding gray level image is the width of the preliminary welding seam area where the welding gray level image is located. S220 is to divide the preliminary bead region into N bead pixel blocks of equal length.
The preliminary weld joint area obtained by using the canny operator is not an accurate weld joint area, because the metals on the upper side and the lower side of the weld joint can be subjected to high temperature to generate a certain degree of color change in the welding process, and the welding flux can be splashed in the welding process. These phenomena can cause the gray value of the metal surface near the weld to change, and further cause gray differences with unaffected metal areas, so that errors occur in the areas detected by the canny operator, i.e., the detected areas include the heat affected area and the weld area. The heat affected zone may have a phenomenon of splashing of the solder, which is a defect of welding, and the worse the splashing phenomenon, the worse the welding quality of the place.
In this embodiment, a canny operator is used to detect a welding gray image to obtain a preliminary welding seam region, and if n=100 is taken, the preliminary welding seam region is divided into welding seam pixel blocksWeld pixel block->Weld pixel block>Weld pixel block->To the weld pixel block->The length of each welding seam pixel block is equal, and the width of each welding seam pixel block is the width of the preliminary welding seam area where the welding seam pixel block is positioned.
Weld joint pixel blockIs divided into a heat affected zone and an actual weld zone.
Weld joint pixel blockIs divided into a heat affected zone and an actual weld zone.
...
Weld joint pixel blockThe weld joint is divided into a heat affected zone and an actual weld joint zone, and thus the preliminary weld joint zone is divided.
In an embodiment of the present application, the S230 includes the following S2231 to S234:
s231, selecting a welding line pixel block.
S232, selecting an initial seed point in the selected welding line pixel block.
S233, performing region growth by taking the initial seed point as a starting point, and performing self-adaptive adjustment on a threshold value of the region growth based on the characteristics of the heat affected region in the process of the region growth to obtain the heat affected region and the actual weld region in the weld pixel block.
S234, returning to S231 until each welding line pixel block is selected.
Specifically, the purpose of dividing the welding gray image into preliminary welding seam areas is to better judge whether the welding seam areas in the welding seam pixel blocks have defects, and the welding seam areas may have the phenomenon of uneven welding flux part during welding, which also belongs to a representation of poor welding seam quality, and detection can be completed through the difference between different welding seam pixel blocks.
The logic is as follows: the weld joint area often has certain texture characteristics, when the weld joint area has defects, the texture characteristics are destroyed, so that the texture difference between different weld joint pixel blocks is larger, and the texture difference between the weld joint pixel blocks without defects is smaller.
In this embodiment, a weld pixel block is selectedIn the weld pixel block->Selecting initial seed point->Selecting->Is->Pixel in the neighborhood, pair +.>Performing region growth, performing adaptive adjustment on a region growth threshold after performing region growth for several times, and obtaining a welding line pixel block +.>A heat affected zone within and an actual weld zone.
In an embodiment of the present application, the S232 includes S232A and S232B:
S232A, calculating a pixel point optimal selection value Y of each pixel point in the welding line pixel block according to the formula 1.
Equation 1.
Wherein Y is a pixel point optimal selection value in the welding line pixel block, q is a pixel point corresponding to the pixel point optimal selection value Y,is pixel dot +.>Point->Gray variance of pixels in the neighborhood, +.>Is pixel dot +.>Minimum Euclidean distance between point and edge pixel point of preliminary weld joint region in weld joint pixel block, < ->Is pixel dot +.>Gray value of dot +.>Gray value of pixel point on metal surface not affected by heat +.>Is a very small positive number.
S232B, sorting the pixel point optimal selection values Y according to the sequence from large to small, and selecting the pixel point corresponding to the maximum pixel point optimal selection value Y as an initial seed point.
Specifically, as shown in fig. 3, the minimum euclidean distance is calculated from the q point, innumerable straight lines are made to the edge of the preliminary weld region, a plurality of line segments are formed, and the length of the shortest line segment is recorded as the minimum euclidean distance.
In this embodiment, if the weld pixel blockThe pixel point is included>、/>、/>、...、/>Z is the number of pixels contained in the weld pixel block.
Calculated according to equation 1Corresponding preference values ∈ ->。
Calculated according to equation 1Corresponding preference values ∈ ->。
Calculated according to equation 1Corresponding preference values ∈ ->。
...
Calculated according to equation 1Corresponding to a large preference value +.>。
Will be、/>、/>、...、/>Ordering of-></></><...</>Select +.>As an initial pixel point.
In an embodiment of the present application, the S233 includes the following S233-1 to S233-13:
s233-1, setting a local growth threshold T and setting the initial value of the area growth times w to 0.
S233-2, selecting initial seed pointsOne pixel point in the neighborhood +.>，/>Is the pixel sequence number.
S233-3, calculating initial seed points according to equation 2Intra-neighborhood +.>Gray scale difference values between the individual pixel points and the initial seed points.
Equation 2.
Wherein qw is the initial seed point used in the w-th growth,is +.>Intra-neighborhood +.>Gray difference value between individual pixel point and initial seed point,/and the like>Is +.>Intra-neighborhood +.>Gray value of each pixel, +.>Is the gray value of the initial seed point.
S233-4, judging initial seed pointIntra-neighborhood +.>Whether the gray difference value between each pixel point and the initial seed point is smaller than the local growth threshold value T.
S233-5, if the initial seed pointIn the neighborhood of->If the gray difference value between each pixel point and the initial seed point is smaller than the local growth threshold value T, carrying out regional growth on the initial seed point, and carrying out +.>Is used as an extension region generated after the w-th growth.
S233-6, adding 1 to the value of the region growing times w on the basis of the original value.
S223-7 at the initial seed PointSelecting a pixel point in the neighborhood of which the initial seed point is not defined as a new initial seed point.
S233-8, returning to S223-3 until the growth stopping condition is met, and stopping the growth.
S233-9, if the initial seed point is in the neighborhoodAnd if the gray difference value between each pixel point and the initial seed point is greater than or equal to the local growth threshold T, returning to the step S233-2.
S233-10, after stopping growing, all the expansion areas are communicated as growing areas.
S233-11, taking the growth area as a heat affected zone;
s233-12, acquiring a solder splash area;
and S233-13, taking other areas except the heat influence area and the solder splash area in the pixel block as actual welding seam areas.
Specifically, the local growth threshold t=10 is preset.
And the growth stopping condition is that no pixel points in the pixel block meet the threshold value judgment, namely after the growth is completed, the rest pixel points which do not undergo regional growth are smaller than the local growth threshold value T.
When the area growth is carried out, as the heat influence effects to the metal surfaces at different positions are different, the gray values of the metal surfaces at different positions are different, and the gray uneven phenomenon is very easy to occur, so that when the preset fixed threshold value is used for the growth, the growth effect is poor, and the obtained heat influence area is inaccurate. And further, the acquisition accuracy of the actual welding seam area is not high.
This practice isIn an embodiment, in a weld pixel blockInterior selection->As initial seed point +.>Is->Neighborhood pixel point, performing first region growth, and randomly selecting +.>Is->One pixel point in the neighborhood +.>Calculate->If-><T, pixel point +.>As an initial seed point.
SelectingIs->Neighborhood pixel point, performing second region growth, marking the second region growth range as a second expansion region, and randomly selecting +.>Is->Divide ∈>Any pixel outside +.>Calculation ofIf-><T, pixel point +.>As an initial seed point.
...
After stopping growing, the first expansion area is communicated with the z expansion area to serve as a growing area, the growing area is served as a heat affected area, an area which is contained in the heat affected area and is not the heat affected area is a solder splash area, and other areas in the pixel block are actual welding seam areas.
S233-8, if the welding seam pixel block contains G pixel points, respectivelyTo->The method comprises the steps of carrying out a first treatment on the surface of the Pixel dot +.>To->Sequencing to obtain:
</><...</><T</></></>then in the completion of pixel point +.>After the growth of the region of (2) the growth can be stopped.
In S233-12, the pixel block area a is composed of a heat affected area B, a solder splash area C and an actual weld area D, the area grown according to the above steps is the heat affected area B, and the solder splash tends to occur in the heat affected area B, i.e. the relationship between the heat affected area B and the solder splash area C is similar to a circular ring, wherein the heat affected area B is a large circle, and the solder splash area C is a small circle.
The region other than the heat influence region B and the solder splash region C in the pixel block region a is the actual weld region D.
In one embodiment of the present application, before S233-7, S233-a to S233-I are included:
S233-A, setting a threshold value omega of the growth times.
S233-B, judging whether the area growth times w are larger than or equal to a growth times threshold value omega.
S233-C, if the area growth times w are greater than or equal to the growth times threshold omega, connecting the expansion area generated after the first growth to the expansion area generated after the w th growth, and taking the expansion area as a growth area, wherein the growth area consists of P pixel point columns.
S233-D, selecting a pixel point column.
S233-E, obtaining the number of the pixels contained in the pixel column.
S233-F, judging whether the number of the pixel points contained in the pixel point row is larger than 1.
S233-G, if the number of the pixels contained in the pixel point row is greater than 1, marking the pixel point row as a feature row, and returning to the selection of one pixel point row until all the pixel point rows are selected, so as to obtain n feature rows.
S233-H, calculating the optimized local growth threshold according to the formula 3, the formula 4 and the formula 5。
Equation 3.
Equation 4.
Equation 5.
Wherein RA is a first regulation parameter, RB is a second regulation parameter, T is a local growth threshold before optimization,in order to optimize the local growth threshold, n is the number of feature columns in the growth region, U is the total number of pixel points in the ith feature column,for the j-th pixel point on the i-th characteristic column and +.>Is->Maximum gray scale difference of other pixels in the neighborhood, which are also located in the growth area, +.>On the ith feature columnGray values of j pixels, +.>For the gray value of the j+1th pixel point on the ith characteristic column, +.>Coordinate value of the jth pixel point on the ith characteristic column in plumb direction, +.>Is the coordinate value of the j+1th pixel point on the ith characteristic column in the plumb direction.
S233-I, replacing the original local growth threshold with the optimized local growth threshold, and executing the subsequent operation at the initial seed pointSelecting a pixel point in the neighborhood of which the initial seed point is not defined as a new initial seed point.
Specifically, the threshold number of growth times Ω may be 20.
The selection of the initial seed points is completed through the preferred values calculated in S233-A to S233-I, so that the phenomenon that the growth is stopped when the growth times are small and an accurate heat affected zone cannot be obtained can be effectively avoided, and meanwhile, the effect of self-adaptive adjustment of the growth threshold value according to the pixel point characteristics of the grown zone can be ensured.
In this embodiment, the growth number threshold Ω=20 is taken, and when the area growth number w=Ω=20, the expansion area generated after the first growth is communicated to the expansion area generated after the 20 th growth, and the growth area is composed of P pixel columns.
Randomly selecting pixel point columns from P pixel point columnsAcquiring pixel column +.>The number of pixels included, if the pixel column +.>The number of the pixel points is larger than 1, and the pixel point column is +.>As a characteristic line, if the pixel point line +.>And if the number of the contained pixels is less than or equal to 1, selecting the next pixel column.
Randomly selecting pixel point columns from P pixel point columnsAcquiring pixel column +.>The number of pixels included, if the pixel column +.>The number of the pixel points is larger than 1, and the pixel point column is +.>As a characteristic line, if the pixel point line +.>And if the number of the contained pixels is less than or equal to 1, selecting the next pixel column.
...
Selecting pixel point columns from P pixel point columnsAcquiring pixel column +.>The number of pixels included, if the pixel column +.>The number of the pixel points is larger than 1, and the pixel point column is +.>As a feature column.
If the number of pixel points contained in the P feature columns is greater than 1, P feature columns are obtained, and the optimized local growth threshold is calculated according to the formula 3, the formula 4 and the formula 5。
Let t=Executing the subsequent +.>Selecting a pixel point in the neighborhood of which the initial seed point is not defined as a new initial seed point.
In an embodiment of the present application, the S300 includes S310 to S360:
s310, selecting a welding seam pixel block, and acquiring an actual welding seam region of the welding seam pixel block.
S320, extracting texture features of an actual welding seam area by using the gray level co-occurrence matrix, and constructing the defect degree H by using a formula 6, a formula 7, a formula 8 and a formula 9 based on texture feature differences.
Equation 6.
Equation 7.
Equation 8.
Equation 9.
Wherein H is the defect degree,is->Energy of the actual weld area within the individual pixel block,/->Is->Contrast of the actual weld area within the individual pixel block, is->Is the energy of the actual weld area within pixel block Q, +.>For the contrast of the actual weld area within pixel block Q, +.>For the distance between pixel blocks, +.>For the maximum value of the distance between pixel block Q and the remaining pixel blocks, < >>For the final optimized growth threshold value of the region growth during the acquisition of the actual weld region in pixel block Q, ->Optimizing the number of times of the final optimized growth threshold value for the region growth in the process of acquiring the actual weld region in the pixel block Q,/for the region growth>For pixel block->Final optimized growth threshold for region growth during acquisition of the inner actual weld region, +.>For pixel block->Growth threshold optimization times in the process of acquiring the inner actual weld joint region, < ->Characterizing defectivity by texture feature differences of the actual weld area,/->Indicating differences in the region growing process.
S330, carrying out normalization processing on the defect degree H, and judging whether the defect degree H is smaller than a defect degree threshold value.
S340, if the defect degree H is smaller than the defect degree threshold value, no defect exists in the welding line area, and further extraction is not needed.
S350, if the defect degree H is greater than or equal to the defect degree threshold value, randomly selecting a pixel point with the largest gray frequency in the actual welding seam area as an initial seed point, re-using an area growth algorithm based on self-adaptive adjustment of a local growth threshold value for the actual welding seam area, stopping growth, taking a growth area in the actual welding seam area as a normal area, and taking other areas except the growth area in the actual welding seam area as defect areas.
S360, returning to S310 until all the welding line pixel blocks are selected.
Specifically, the defect level threshold may be set to 0.8.
Normalization is a way to simplify computation, i.e., an expression with dimensions is transformed into a dimensionless expression, which becomes a scalar.
The gray level co-occurrence matrix refers to a common method for describing textures by studying spatial correlation characteristics of gray levels. The texture features may be described using a gray level co-occurrence matrix. Since the texture is formed by repeatedly appearing the gray scale distribution at the spatial position, a certain gray scale relationship exists between two pixels at a certain distance in the image space, namely the spatial correlation characteristic of the gray scale in the image.
ASM is the sum of squares of each matrix element.
CON directly reflects the contrast of the brightness of a certain pixel value and its field pixel value.
In this embodiment, the defect degree H is calculated by equation 7, equation 9 and equation 9, and the defect degree H is normalized.
If the defect degree H is greater than or equal to the defect degree threshold value, a defect exists in the welding line area.
If the defect level H is less than the defect level threshold, no defect exists in the weld area.
In an embodiment of the present application, the S300 further includes S371 and S372:
s371, calculating uniformity of the solder during soldering according to the formula 10.
Equation 10.
Wherein R is the uniformity of the solder during welding, N is the total number of pixel blocks of the welding seam,represents the area of the actual weld area A within the weld pixel block Q, +.>For the serial number of the other weld pixel blocks than the weld pixel block Q, +.>Represents +.>The area of the actual weld area within each weld pixel block.
S372, calculating the weld quality according to the formula 11.
Equation 11.
Wherein ZL is the quality of the welding seam, N is the total number of the pixel blocks of the welding seam,representing the defectivity of the actual weld area in the kth weld pixel block, +.>Indicating the area of the defective area of the actual weld area in the kth weld pixel block, +.>Representing the uniformity of the actual weld area in the kth weld pixel block, +.>Is a very small positive number.
Specifically, since the defective region often occupies only a small portion of the region, the pixel corresponding to the maximum gray frequency is the solder pixel of the normal region.
The minimum positive number, the prevention denominator is 0, and the example value is 0.001.
In this embodiment, in S371,and->The smaller the difference, the smaller the difference of the solder at different positions during welding, so the larger the uniformity R, the larger the uniformity R value, and the better the quality of the welding seam.
In S372, the larger ZL, the better the quality of the weld, the better the welding effect.
The smaller ZL, the poorer the weld quality and the poorer the welding effect.
In an embodiment of the present application, the S300 includes the following S381 to S383:
s381, judging whether the weld quality ZL is smaller than a weld quality threshold.
S382, if the quality ZL of the welding line is smaller than the quality threshold value of the welding line, the quality of the welding line is not up to standard.
S383, if the quality ZL of the welding line is larger than or equal to the quality threshold value of the welding line, the quality of the welding line is up to the standard.
Specifically, the weld quality threshold may be 0.75.
In this embodiment, if ZL is less than 0.75, the quality of the weld is not up to standard, and repair welding or re-welding is required.
If ZL is greater than or equal to 0.75, the welding seam quality is better, and repair welding or re-welding is not needed.
The technical features of the above embodiments may be combined arbitrarily, and the steps of the method are not limited to the execution sequence, so that all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description of the present specification.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.
Claims (5)
1. The iron tower quality analysis method based on visual detection is characterized by comprising the following steps of:
acquiring a welding image of a tower body component of the power transmission tower, and preprocessing the welding image to obtain a welding gray level image of the tower body component;
processing the welding gray level image of the tower member to obtain a preliminary welding seam area, dividing the preliminary welding seam area into a plurality of welding seam pixel blocks, and using an area growth algorithm based on self-adaptive adjustment of a local growth threshold value for each welding seam pixel block to divide each welding seam pixel block into a heat affected area and an actual welding seam area;
extracting a defect area from an actual welding seam area of each welding seam pixel block, and finishing quality detection of the welding seam according to the characteristics of the defect area;
the method for processing the welding gray level image of the tower member to obtain a preliminary welding seam area, dividing the preliminary welding seam area into a plurality of welding seam pixel blocks, and using an area growth algorithm based on self-adaptive adjustment of a local growth threshold value for each welding seam pixel block to divide each welding seam pixel block into a heat affected area and an actual welding seam area, comprises the following steps:
processing the welding gray level image by using canny operator detection to obtain a preliminary welding seam area;
dividing the preliminary weld joint region into N weld joint pixel blocks with equal lengths;
dividing each welding seam pixel block into a heat affected zone and an actual welding seam zone;
the dividing each weld pixel block into a heat affected zone and an actual weld zone includes:
selecting a welding line pixel block;
selecting an initial seed point in the selected welding line pixel block;
performing region growth by taking an initial seed point as a starting point, and performing self-adaptive adjustment on a threshold value of the region growth based on the characteristics of the heat affected region in the process of the region growth to obtain the heat affected region and an actual weld joint region in a weld joint pixel block;
returning to the selected one welding seam pixel block until each welding seam pixel block is selected;
the selecting an initial seed point in the selected welding seam pixel block comprises the following steps:
calculating a pixel point optimal selection value Y of each pixel point in the welding line pixel block according to the formula 1;
equation 1;
wherein Y is a pixel point optimal selection value in a welding line pixel block, and q isA pixel corresponding to the pixel preference value Y,is pixel dot +.>Point->Gray variance of pixels in the neighborhood, +.>Is pixel dot +.>Minimum Euclidean distance between point and edge pixel point of preliminary weld joint region in weld joint pixel block, < ->Is pixel dot +.>Gray value of dot +.>Gray value of pixel point on metal surface not affected by heat +.>Is an extremely small positive number;
sorting the pixel point optimal selection values Y according to the sequence from large to small, and selecting a pixel point corresponding to the maximum pixel point optimal selection value Y as an initial seed point;
the method for performing region growth by taking the initial seed point as a starting point, and performing self-adaptive adjustment on a threshold value of the region growth based on the characteristics of the heat affected region in the process of the region growth to obtain the heat affected region and the actual weld region in the weld pixel block comprises the following steps:
setting a local growth threshold T and setting an initial value of the area growth times w to be 0;
selecting initial seed pointsOne pixel point in the neighborhood +.>，/>The pixel sequence number;
calculating an initial seed point according to equation 2Intra-neighborhood +.>Gray scale difference values between the individual pixel points and the initial seed points;
equation 2;
wherein qw is the initial seed point used in the w-th growth,is +.>Intra-neighborhood +.>Gray difference value between individual pixel point and initial seed point,/and the like>Is +.>Intra-neighborhood +.>Gray value of each pixel, +.>Gray values for the initial seed points;
judging initial seed pointIntra-neighborhood +.>Whether the gray difference value between each pixel point and the initial seed point is smaller than a local growth threshold value T or not;
if the initial seed pointIn the neighborhood of->If the gray difference value between each pixel point and the initial seed point is smaller than the local growth threshold value T, carrying out regional growth on the initial seed point, and carrying out +.>Is used as an expansion area generated after the w-th growth;
the value of the region growing times w is added with 1 on the basis of the original value;
at the initial seed pointSelecting a pixel point which is not defined with an initial seed point in the neighborhood of the initial seed point as a new initial seed point;
returning to said calculating an initial seed point according to equation 2Intra-neighborhood +.>Gray difference values between the individual pixel points and the initial seed points until the growth stopping condition is met, and stopping growth;
if the initial seed point is in the neighborhoodThe gray difference value between each pixel point and the initial seed point is larger than or equal to the local growth threshold value T, and the selected initial seed point is returned to +.>One pixel point in the neighborhood +.>；
After stopping growing, all the expansion areas are communicated to serve as growing areas;
the growth area is used as a heat affected zone;
acquiring a solder splash area;
taking other areas except the heat influence area and the solder splash area in the pixel block as actual welding seam areas;
at the initial seed pointBefore selecting a pixel point which is not defined with the initial seed point as a new initial seed point in the neighborhood of the initial seed point, the method further comprises:
setting a threshold omega of the growth times;
judging whether the growth times w of the region are larger than or equal to a threshold value omega of the growth times;
if the growth times w of the area are greater than or equal to the threshold omega of the growth times, the expansion area generated after the first growth is communicated with the expansion area generated after the w th growth, and the expansion area is used as a growth area, wherein the growth area consists of P pixel point columns;
selecting a pixel point column;
acquiring the number of the pixels contained in the pixel column;
judging whether the number of the pixel points contained in the pixel point row is larger than 1;
if the number of the pixel points contained in the pixel point row is greater than 1, marking the pixel point row as a feature row, and returning to the selection of one pixel point row until all the pixel point rows are selected, so as to obtain n feature rows;
calculating the optimized local growth threshold according to formulas 3, 4 and 5；
Equation 3;
equation 4;
equation 5;
wherein RA is a first regulation parameter, RB is a second regulation parameter, T is a local growth threshold before optimization,in order to optimize the local growth threshold, n is the number of feature columns in the growth region, U is the total number of pixel points in the ith feature column,for the j-th pixel point on the i-th characteristic column and +.>Is->Within the neighborhood also in the growth regionMaximum gray scale difference of other pixels in the pixel array, +.>For the gray value of the j-th pixel point on the i-th characteristic column,/>For the gray value of the j+1th pixel point on the ith characteristic column, +.>Coordinate value of the jth pixel point on the ith characteristic column in plumb direction, +.>The coordinate value of the (j+1) th pixel point on the ith characteristic column in the plumb direction;
replacing the original local growth threshold value with the optimized local growth threshold value, and executing the subsequent operation at the initial seed pointSelecting a pixel point in the neighborhood of which the initial seed point is not defined as a new initial seed point.
2. The iron tower quality analysis method based on visual inspection according to claim 1, wherein the step of acquiring welding images of the tower body members of the power transmission tower, and preprocessing the welding images to obtain welding gray images of the tower body members comprises the steps of:
acquiring an image of a welded power transmission tower body component, and recording the image as a welding image;
and carrying out graying treatment on the welding image by using a weighted graying method to obtain a tower member welding gray image.
3. The visual inspection-based iron tower quality analysis method according to claim 1, wherein the steps of extracting a defective area from the actual weld area of each weld pixel block and performing the quality inspection of the weld according to the characteristics of the defective area include:
selecting a welding seam pixel block, and acquiring an actual welding seam region of the welding seam pixel block;
extracting texture features of an actual welding seam region by using a gray level co-occurrence matrix, and constructing a defect degree H by using a formula 6, a formula 7, a formula 8 and a formula 9 based on texture feature differences;
equation 6;
equation 7;
equation 8;
equation 9;
wherein H is the defect degree,is->Energy of the actual weld area within the individual pixel block,/->Is->Contrast of the actual weld area within the individual pixel block, is->Is the energy of the actual weld area within pixel block Q, +.>For the contrast of the actual weld area within pixel block Q, +.>For the distance between pixel blocks, +.>For the maximum value of the distance between pixel block Q and the remaining pixel blocks, < >>For the final optimized growth threshold for region growth during the acquisition of the actual weld region within pixel block Q,the number of growth threshold optimizations after final optimization for region growth during acquisition of the actual weld region within pixel block Q,for pixel block->Final optimized growth threshold for region growth during acquisition of the inner actual weld region, +.>For pixel block->Growth threshold optimization times in the process of acquiring the inner actual weld joint region, < ->Characterizing defectivity by texture feature differences of the actual weld area,/->Indicating region growthProcess variation;
carrying out normalization processing on the defect degree H, and judging whether the defect degree H is smaller than a defect degree threshold value or not;
if the defect degree H is smaller than the defect degree threshold value, no defect exists in the welding line area, and further extraction is not needed;
if the defect degree H is greater than or equal to the defect degree threshold value, randomly selecting a pixel point with the largest gray frequency in an actual welding seam area as an initial seed point, re-using an area growth algorithm based on self-adaptive adjustment of a local growth threshold value for the actual welding seam area, taking a growth area in the actual welding seam area as a normal area after stopping growth, and taking other areas except the growth area in the actual welding seam area as defect areas;
and returning to the selected one welding seam pixel block, and obtaining the actual welding seam area of the welding seam pixel block until all the welding seam pixel blocks are selected.
4. A visual inspection-based iron tower quality analysis method according to claim 3, wherein the steps of extracting a defective area from the actual weld area of each weld pixel block and completing the quality inspection of the weld according to the characteristics of the defective area further comprise:
calculating the uniformity of the solder during welding according to formula 10;
equation 10;
wherein R is the uniformity of the solder during welding, N is the total number of pixel blocks of the welding seam,represents the area of the actual weld area A within the weld pixel block Q, +.>For the serial number of the other weld pixel blocks than the weld pixel block Q, +.>Represents +.>The areas of the actual weld areas in the individual weld pixel blocks;
calculating the weld quality according to formula 11;
equation 11;
wherein ZL is the quality of the welding seam, N is the total number of the pixel blocks of the welding seam,representing the defectivity of the actual weld area in the kth weld pixel block, +.>Indicating the area of the defective area of the actual weld area in the kth weld pixel block, +.>Representing the uniformity of the actual weld area in the kth weld pixel block, +.>Is a very small positive number.
5. The visual inspection-based iron tower quality analysis method according to claim 4, wherein the steps of extracting a defective area from the actual weld area of each weld pixel block and performing the quality inspection of the weld according to the characteristics of the defective area, further comprise:
judging whether the welding seam quality ZL is smaller than a welding seam quality threshold value or not;
if the quality ZL of the welding seam is smaller than the quality threshold value of the welding seam, the quality of the welding seam is not up to the standard;
and if the quality ZL of the welding seam is larger than or equal to the quality threshold value of the welding seam, the quality of the welding seam is up to the standard.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310650678.3A CN116385476B (en) | 2023-06-05 | 2023-06-05 | Iron tower quality analysis method based on visual detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310650678.3A CN116385476B (en) | 2023-06-05 | 2023-06-05 | Iron tower quality analysis method based on visual detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116385476A CN116385476A (en) | 2023-07-04 |
| CN116385476B true CN116385476B (en) | 2023-08-18 |
Family
ID=86980987
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310650678.3A Active CN116385476B (en) | 2023-06-05 | 2023-06-05 | Iron tower quality analysis method based on visual detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116385476B (en) |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014139273A1 (en) * | 2013-03-14 | 2014-09-18 | 苏州华源包装股份有限公司 | Weld seam defect detection method |
| WO2016172827A1 (en) * | 2015-04-27 | 2016-11-03 | 武汉武大卓越科技有限责任公司 | Stepwise-refinement pavement crack detection method |
| CN107093175A (en) * | 2017-04-21 | 2017-08-25 | 华南师范大学 | A kind of electrokinetic cell welding quality detection method |
| CN108537813A (en) * | 2017-03-03 | 2018-09-14 | 防城港市港口区思达电子科技有限公司 | Object detection method based on region growing |
| KR20190063726A (en) * | 2017-11-30 | 2019-06-10 | 오스템임플란트 주식회사 | Image processing method and system for dental ct imaging, computer readable recording medium and computer program |
| CN110400285A (en) * | 2019-05-30 | 2019-11-01 | 哈尔滨工业大学（深圳） | Wire bonding pad defect location and classification method |
| CN110751604A (en) * | 2019-09-30 | 2020-02-04 | 佛山科学技术学院 | Machine vision-based steel pipe weld defect online detection method |
| CN111667462A (en) * | 2020-05-12 | 2020-09-15 | 常熟理工学院 | Automatic detection method and system for small and weak targets in welding defects |
| CN112233133A (en) * | 2020-10-29 | 2021-01-15 | 上海电力大学 | Power plant high-temperature pipeline defect detection and segmentation method based on OTSU and region growing method |
| CN114240877A (en) * | 2021-12-15 | 2022-03-25 | 合肥欣奕华智能机器有限公司 | Method and device for detecting welding quality |
| CN114723700A (en) * | 2022-03-31 | 2022-07-08 | 江苏宝诺铸造有限公司 | Iron furnace surface crack growth prediction method and system based on computer vision |
| CN114743005A (en) * | 2022-04-12 | 2022-07-12 | 西安理工大学 | Infrared dim target detection method based on threshold region growth and gray level difference |
| CN115063404A (en) * | 2022-07-27 | 2022-09-16 | 建首(山东)钢材加工有限公司 | Weathering resistant steel weld joint quality detection method based on X-ray flaw detection |
| CN115082464A (en) * | 2022-08-22 | 2022-09-20 | 南通翡利达液压科技有限公司 | Method and system for identifying welding seam data in welding process of dust remover |
| CN115100171A (en) * | 2022-07-11 | 2022-09-23 | 常宝云 | Steel die welding defect detection method and system based on machine vision |
| CN115439481A (en) * | 2022-11-09 | 2022-12-06 | 青岛平电锅炉辅机有限公司 | Deaerator welding quality detection method based on image processing |
| CN115457035A (en) * | 2022-11-10 | 2022-12-09 | 山东鲁旺机械设备有限公司 | Machine vision-based construction hanging basket welding quality detection method |
| CN115719332A (en) * | 2022-10-18 | 2023-02-28 | 江苏思锐装备科技有限公司 | Welding quality detection method |
| CN115880302A (en) * | 2023-03-08 | 2023-03-31 | 杭州智源电子有限公司 | Instrument panel welding quality detection method based on image analysis |
| CN115984249A (en) * | 2023-02-10 | 2023-04-18 | 西南石油大学 | Radiographic image weld segmentation algorithm based on iteration threshold |
| CN116152214A (en) * | 2023-03-02 | 2023-05-23 | 山西春福科技有限公司 | Plastic powder weather resistance detection method for metal product coating |
-
2023
- 2023-06-05 CN CN202310650678.3A patent/CN116385476B/en active Active
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014139273A1 (en) * | 2013-03-14 | 2014-09-18 | 苏州华源包装股份有限公司 | Weld seam defect detection method |
| WO2016172827A1 (en) * | 2015-04-27 | 2016-11-03 | 武汉武大卓越科技有限责任公司 | Stepwise-refinement pavement crack detection method |
| CN108537813A (en) * | 2017-03-03 | 2018-09-14 | 防城港市港口区思达电子科技有限公司 | Object detection method based on region growing |
| CN107093175A (en) * | 2017-04-21 | 2017-08-25 | 华南师范大学 | A kind of electrokinetic cell welding quality detection method |
| KR20190063726A (en) * | 2017-11-30 | 2019-06-10 | 오스템임플란트 주식회사 | Image processing method and system for dental ct imaging, computer readable recording medium and computer program |
| CN110400285A (en) * | 2019-05-30 | 2019-11-01 | 哈尔滨工业大学（深圳） | Wire bonding pad defect location and classification method |
| CN110751604A (en) * | 2019-09-30 | 2020-02-04 | 佛山科学技术学院 | Machine vision-based steel pipe weld defect online detection method |
| CN111667462A (en) * | 2020-05-12 | 2020-09-15 | 常熟理工学院 | Automatic detection method and system for small and weak targets in welding defects |
| CN112233133A (en) * | 2020-10-29 | 2021-01-15 | 上海电力大学 | Power plant high-temperature pipeline defect detection and segmentation method based on OTSU and region growing method |
| CN114240877A (en) * | 2021-12-15 | 2022-03-25 | 合肥欣奕华智能机器有限公司 | Method and device for detecting welding quality |
| CN114723700A (en) * | 2022-03-31 | 2022-07-08 | 江苏宝诺铸造有限公司 | Iron furnace surface crack growth prediction method and system based on computer vision |
| CN114743005A (en) * | 2022-04-12 | 2022-07-12 | 西安理工大学 | Infrared dim target detection method based on threshold region growth and gray level difference |
| CN115100171A (en) * | 2022-07-11 | 2022-09-23 | 常宝云 | Steel die welding defect detection method and system based on machine vision |
| CN115063404A (en) * | 2022-07-27 | 2022-09-16 | 建首(山东)钢材加工有限公司 | Weathering resistant steel weld joint quality detection method based on X-ray flaw detection |
| CN115082464A (en) * | 2022-08-22 | 2022-09-20 | 南通翡利达液压科技有限公司 | Method and system for identifying welding seam data in welding process of dust remover |
| CN115719332A (en) * | 2022-10-18 | 2023-02-28 | 江苏思锐装备科技有限公司 | Welding quality detection method |
| CN115439481A (en) * | 2022-11-09 | 2022-12-06 | 青岛平电锅炉辅机有限公司 | Deaerator welding quality detection method based on image processing |
| CN115457035A (en) * | 2022-11-10 | 2022-12-09 | 山东鲁旺机械设备有限公司 | Machine vision-based construction hanging basket welding quality detection method |
| CN115984249A (en) * | 2023-02-10 | 2023-04-18 | 西南石油大学 | Radiographic image weld segmentation algorithm based on iteration threshold |
| CN116152214A (en) * | 2023-03-02 | 2023-05-23 | 山西春福科技有限公司 | Plastic powder weather resistance detection method for metal product coating |
| CN115880302A (en) * | 2023-03-08 | 2023-03-31 | 杭州智源电子有限公司 | Instrument panel welding quality detection method based on image analysis |
Non-Patent Citations (1)
| Title |
|---|
| 基于红外图像特征的建筑热工缺陷检测方法;鲁友存;《中国优秀硕士学位论文全文数据库 工程科技II辑》;第2023年卷(第2期);C038-652 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116385476A (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shah et al. | Butt welding joints recognition and location identification by using local thresholding | |
| CN117314912B (en) | Visual detection method and system for welding defects on surface of welded pipe | |
| CN109685760B (en) | MATLAB-based SLM powder bed powder laying image convex hull depression defect detection method | |
| CN115439481B (en) | Deaerator welding quality detection method based on image processing | |
| CN110363781B (en) | Molten pool contour detection method based on deep neural network | |
| KR20040026157A (en) | Method of setting reference images, method and apparatus using the setting method for inspecting patterns on a wafer | |
| CN103543394A (en) | Discharge ultraviolet imaging quantization parameter extraction method of high-voltage electric equipment | |
| CN114706358B (en) | Method and system for processing welding process information of straight welded pipe | |
| CN109829856B (en) | Bridge crack information fusion method | |
| TW201921139A (en) | Defect detection on transparent or translucent wafers | |
| CN114240845A (en) | Surface roughness measuring method by adopting light cutting method applied to cutting workpiece | |
| CN114881965A (en) | Wood board joint detection method based on artificial intelligence and image processing | |
| CN113538378A (en) | Bearing size online detection system based on deep learning | |
| TWI780234B (en) | Method, system and non-transitory computer readable medium for determining a critical dimension variation of a pattern | |
| CN110555385B (en) | Welding seam characteristic point solving method based on variable step length curvature filtering | |
| CN116385476B (en) | Iron tower quality analysis method based on visual detection | |
| WO2019114453A1 (en) | Highlight area self-adaptive dodging method for mirror-like object | |
| CN115326837A (en) | Stainless steel welding spot surface quality detection method and system based on machine vision | |
| CN114693681A (en) | Zipper production defect intelligent detection method and system based on optical means | |
| CN114663416A (en) | Computer vision-based copper bar cold shut defect detection method and system | |
| CN110211113B (en) | Detection algorithm and calculation equipment for groove abnormity | |
| JP2022150065A (en) | Inspection device, computer program, and inspection method | |
| CN111089865A (en) | F-RCNN-based defective cable detection method | |
| CN117115143B (en) | Method and device for detecting wafer core particle, computer equipment and storage medium | |
| CN117974657B (en) | Cable surface defect detection method based on computer vision |
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 |