CN113649672A - Adaptive extraction method for geometric characteristics of butt weld - Google Patents
Adaptive extraction method for geometric characteristics of butt weld Download PDFInfo
- Publication number
- CN113649672A CN113649672A CN202110899874.5A CN202110899874A CN113649672A CN 113649672 A CN113649672 A CN 113649672A CN 202110899874 A CN202110899874 A CN 202110899874A CN 113649672 A CN113649672 A CN 113649672A
- Authority
- CN
- China
- Prior art keywords
- welding
- roi
- image
- line
- point
- 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.)
- Pending
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 22
- 230000003044 adaptive effect Effects 0.000 title claims description 12
- 238000003466 welding Methods 0.000 claims abstract description 78
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 18
- 230000000007 visual effect Effects 0.000 claims abstract description 15
- 238000009499 grossing Methods 0.000 claims abstract description 3
- 238000007781 pre-processing Methods 0.000 claims abstract description 3
- 238000001914 filtration Methods 0.000 claims description 7
- 230000003287 optical effect Effects 0.000 claims description 7
- 238000012935 Averaging Methods 0.000 claims description 5
- 230000007547 defect Effects 0.000 claims description 5
- 235000002566 Capsicum Nutrition 0.000 claims description 4
- 239000006002 Pepper Substances 0.000 claims description 4
- 235000016761 Piper aduncum Nutrition 0.000 claims description 4
- 235000017804 Piper guineense Nutrition 0.000 claims description 4
- 235000008184 Piper nigrum Nutrition 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 150000003839 salts Chemical class 0.000 claims description 4
- 239000011347 resin Substances 0.000 claims description 3
- 229920005989 resin Polymers 0.000 claims description 3
- 244000203593 Piper nigrum Species 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 3
- 241000722363 Piper Species 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000010146 3D printing Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K9/00—Arc welding or cutting
- B23K9/12—Automatic feeding or moving of electrodes or work for spot or seam welding or cutting
- B23K9/127—Means for tracking lines during arc welding or cutting
- B23K9/1272—Geometry oriented, e.g. beam optical trading
- B23K9/1274—Using non-contact, optical means, e.g. laser means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—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
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- 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
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- 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
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- 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
- G06T2207/30152—Solder
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Optics & Photonics (AREA)
- Geometry (AREA)
- Plasma & Fusion (AREA)
- Mechanical Engineering (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Laser Beam Processing (AREA)
Abstract
The invention relates to a welding image processing technology, in particular to a self-adaptive extraction method of geometric characteristics of a butt weld, which utilizes a visual sensor to acquire a real-time image of the butt weld during welding; preprocessing the acquired real-time image of the butt weld, and extracting an ROI (region of interest) area during welding; performing image processing on the extracted ROI area during welding, removing interference and smoothing the image; and (4) extracting a laser center line from the image processed in the step (3) and calculating the position of the center point of the welding line. The method can effectively process the collected real-time welding image, accurately intercept the ROI area, and quickly and accurately extract the position information such as the central line of the light strip, the central point of the welding seam and the like, thereby realizing the tracking of the welding seam. The method has the characteristics of high processing speed, high accuracy, high image quality and the like.
Description
Technical Field
The invention belongs to the technical field of welding image processing, and particularly relates to a self-adaptive extraction method for geometric characteristics of a butt weld.
Background
Welding is an important processing technology in manufacturing industry, is used in aerospace, energy equipment, automobile manufacturing and ship manufacturing in large quantities, but has severe working conditions, large workload and higher requirements on quality. With the continuous development of industrial automation and intelligence, automation and intelligence of welding have become a necessary trend of the development of current welding technology.
With the popularization of a robot programming system, when a structural part is welded, a program is compiled according to a part three-dimensional digital analog in an off-line mode, the actual workpiece shape has certain deviation from a theoretical digital analog, the absolute positioning precision of the robot is low, and the error can reach 0.5-2 mm, so that welding defects and even interference collision are caused; in the welding process, the parts are heated and restrained to form obvious residual stress to cause the deformation of the parts, so that the actual welding seam position deviates from the theoretical track position, and the welding quality is poor.
Therefore, the real-time monitoring and adjustment of the welding track path in the welding process is very important for the butt welding seam of the structural part, and the realization of the rapid and accurate extraction of the welding seam is an important guarantee for guaranteeing the tracking quality of the welding seam. How to perform effective image processing and quickly and accurately extract the geometric characteristics and the central point of the welding seam becomes the key point of the tracking welding of the welding seam.
Disclosure of Invention
Aiming at the problems in the background art, the invention provides a self-adaptive extraction method for geometric characteristics of a butt weld.
In order to solve the technical problems, the invention adopts the following technical scheme: a self-adaptive extraction method for geometric characteristics of a butt weld comprises a visual sensor, wherein the visual sensor is integrally fixed on a welding gun through a clamp and moves along with the welding gun; the method comprises the following steps:
step 1, collecting a real-time image of a butt welding seam during welding by using a vision sensor;
and 4, extracting a laser center line from the image processed in the step 3 and calculating the position of the center point of the welding line.
In the above adaptive extraction method of geometric features of a butt weld, the implementation of step 1 includes: the visual sensor integrates a linear structured light generator, a CMOS industrial camera, an optical lens, a narrow-band filter, a dimmer and an arc baffle, and a shell of the visual sensor is a high-temperature-resistant resin shell for 3D printing; the CMOS industrial camera is arranged on one side of the shell, and the dimmer and the narrow-band filter are arranged below the optical lens; the linear structured light generator and the CMOS industrial camera form an included angle of 20-45 degrees, so that the laser stripes generate deformation information at welding seams, and the deformation view field of the laser stripes is positioned right below the industrial camera and used for collecting real-time images of welding.
In the adaptive extraction method of the geometric features of the butt weld, the ROI extraction method in the step 2 is as follows:
step 2.1, scanning the collected real-time image of the butt-joint weld in rows, and solving the pixel difference between two adjacent pixels for each row, namely the gradient of the pixels in the y direction:
preserving absolute value of gradientThe maximum pixel value f (x, y), and the pixel value on the ith column is denoted as fi(x,y);
Step 2.2, calculating the average value of the pixel values of all the columns as a boundary threshold value H:
step 2.3, performing column scanning on the collected real-time image of the butt-joint weld, finding out specific coordinates a and b with pixel values larger than H in the y direction, and obtaining the height position of the center line of a certain column of laser stripes:
wherein y isiRepresents the ordinate of the i point in the scanned column; b represents the ordinate of the first pixel value greater than the threshold H point scanned onto the column; a represents the ordinate of the last pixel value scanned onto the column above the threshold H point; (a-b +1) represents the width of the laser stripe in the column;
obtaining y value of each row after scanning, and averaging the y values to obtain a longitudinal coordinate value y of the central point of the ROI areaROI;
Step 2.4, scanning the central line, and recording the point abscissa x of which the pixel value is less than the threshold value H in each linejThe line scanning formula is as follows:
wherein xjThe abscissa representing the j point in the scanned line; d represents the abscissa scanned to the point on the row where the first pixel value is less than the threshold H; c represents the abscissa of the last pixel value scanned onto the line to be less than the threshold H; c-d +1 represents the width of the laser stripe defect in the row;
obtaining the x value of each line after scanning, and obtaining the abscissa position x of the central point of the ROI by averaging the x valuesROI;
Step 2.5, obtaining the coordinates (x) of the center point of the ROI area through the step 2.3 and the step 2.4ROI,yROI),A rectangular area of 300 pixels × 400 pixels centered on the central point coordinate position is extracted as an ROI area.
In the adaptive extraction method of the geometric features of the butt weld, the implementation of the step 3 comprises the following steps:
3.1, performing median filtering processing on the extracted ROI to remove salt and pepper noise and smooth laser stripes; performing morphological processing on the image subjected to the median filtering processing;
step 3.2, calculating the proportion of the pixel points larger than the threshold value H in each row to the total number of the pixel points in each row and recording the proportion as eta; when eta is larger than 80%, the laser stripe is located in the row, and pixel points of the row are reserved; and when the eta is less than 80 percent, removing the pixel points of the line and resetting the pixel points to be 0.
In the adaptive extraction method of the geometric features of the butt weld, the implementation of the step 4 comprises the following steps:
step 4.1, carrying out binarization and reverse color processing on the image processed in the step 3.2; extracting the laser stripes at the geometric center of the existing laser stripes as characteristic stripes of the welding seams;
step 4.2, scanning the characteristic stripes of the welding seams from left to right, and extracting left and right breakpoints of discontinuous positions at the center of the image;
step 4.2.1, judging conditions according to the left breakpoint:
obtain the left breakpoint coordinate (x)1,y1);
Step 4.2.2, judging conditions according to the right breakpoint:
get the right breakpoint coordinate (x)2,y2);
Step 4.3, utilizing the left breakpoint coordinate (x)1,y1) Right breakPoint coordinates (x)2,y2) Calculating the coordinates of the center point of the welding seam
Compared with the prior art, the method and the device have the advantages that the laser stripe boundary threshold is used as the judgment basis of column scanning and line scanning, and the central point position of the ROI can be simply and quickly positioned. The size of the obtained ROI area is far smaller than that of the original image, and the subsequent processing speed is greatly increased. The longitudinal interference except the laser stripes can be eliminated by utilizing the proportion of the laser stripes in the image lines, and the laser stripes are completely reserved. In the process of real-time tracking welding of the robot, the method can effectively process the collected real-time welding image, accurately intercept the ROI area, and quickly and accurately extract the position information of the central line of the light strip, the central point of the welding seam and the like, thereby realizing the tracking of the welding seam. The method has the characteristics of high processing speed, high accuracy, high image quality and the like.
Drawings
FIG. 1 is a flow chart of an adaptive extraction method for geometric features of a butt weld according to an embodiment of the present invention;
FIG. 2 is a schematic view of a vision sensor integration according to an embodiment of the present invention;
wherein, 1-welding plate, 2-line structure light generator, 3-welding gun, 4-CMOS industrial camera, 5-industrial robot arm end, 6-optical lens, 7-narrow band filter, 8-dimmer, 9-arc baffle;
FIG. 3 is a schematic diagram of a pixel matrix according to an embodiment of the invention;
FIG. 4 is a schematic diagram of ROI area extraction according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of image processing and feature extraction according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the following embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The present invention is further illustrated by the following examples, which are not to be construed as limiting the invention.
In order to solve the problem of adaptive extraction of the geometric features of the butt weld under the interference of arc light, splashing and dust fog, the method for extracting the geometric features of the butt weld provided by the embodiment can rapidly locate the position of a specific area by using a boundary threshold, completely reserve laser stripe information, rapidly and accurately extract the geometric features of the butt weld and the central point of the butt weld, and has important significance for the automation and intelligent development of the butt weld. In the embodiment, the visual sensor is used for acquiring the wired structure light-assisted real-time welding image, extracting the welding specific area, performing image processing on the acquired specific area, and extracting the geometric characteristics and the central point of the welding seam.
The embodiment is realized by the following technical scheme, as shown in fig. 1, a method for adaptively extracting geometric features of a butt weld comprises the following steps:
and S1, constructing a visual sensor containing the linear structured light generator, fixing the visual sensor with a welding gun, and acquiring a real-time image of a butt welding seam during welding by using the sensor.
S2, the image acquired in step S1 is preprocessed, and an ROI region (region of interest) during welding is extracted.
S3, the ROI extracted in step S2 is further image-processed to remove the interference and smooth the image.
And S4, extracting the laser center line of the image in the step S3 and calculating the position of the weld center point.
In step S1, the entire visual sensor is fixed to the welding gun by a specific jig, and moved together with the welding gun while maintaining a fixed positional relationship with the welding gun. The sensor integrates a line structured light generator, a CMOS industrial camera, an optical lens, a narrow-band filter, a dimmer and an arc baffle, and a shell of the sensor is a high-temperature resistant resin shell for 3D printing. The CMOS industrial camera is arranged on one side of the shell, and the dimmer and the narrow-band filter are arranged below the optical lens so as to reduce the arc intensity and the splash interference. The linear structured light generator and the CMOS industrial camera form an included angle of 20-45 degrees, so that the laser stripes generate deformation information at welding seams, and the deformation view field of the laser stripes is positioned right below the CMOS industrial camera, so as to collect real-time images of welding.
In step S2, on the basis of the real-time image of the butt weld, a pixel difference between two adjacent pixels, i.e., a gradient of the pixels in the y direction, is determined for each row by using a row scan:
preserving absolute value of gradientThe maximum pixel value f (x, y), and the pixel value on the ith column is denoted as fi(x,y)。
The average of the pixel values of all columns is calculated as the boundary threshold H:
then, scanning the image in a row, finding out specific coordinates a and b with the pixel value larger than H in the y direction, and obtaining the height position of the center line of a certain row of laser stripes:
wherein y isiRepresents the ordinate of the i point in the scanned column; b represents the ordinate of the first pixel value greater than the threshold H point scanned onto the column; a represents the ordinate of the last pixel value scanned onto the column above the threshold H point; (a-b +1) indicates that the laser stripe is at the sameThe width in the column;
after scanning, the y value of each row can be obtained, and the average value is obtained to obtain the longitudinal coordinate value y of the central point of the ROI areaROI。
Scanning the central line, and recording the point abscissa x with the pixel value smaller than the threshold value H in each linejThe line scanning formula is as follows:
wherein xjThe abscissa representing the j point in the scanned line; d represents the abscissa scanned to the point on the row where the first pixel value is less than the threshold H; c represents the abscissa of the last pixel value scanned onto the line to be less than the threshold H; (c-d +1) represents the width of the laser stripe defect in the row;
obtaining the x value of each line after scanning, and obtaining the abscissa position x of the central point of the ROI by averaging the x valuesROI。
Thereby obtaining the coordinates (x) of the center point of the ROI regionROI,yROI) A rectangular region of 300 pixels × 400 pixels centered on the center point coordinate position is extracted as the ROI region.
In step S3, the extracted ROI is subjected to median filtering to remove salt and pepper noise and smooth the laser stripe. The filtering method is suitable for processing randomly distributed noise, is beneficial to removing interference caused by splashing, and performs morphological processing on the filtered image.
And calculating the proportion of the pixel points larger than the threshold value H in each row to the total number of the pixel points in the row and recording as eta. When eta is larger than 80%, the laser stripe is regarded as the line where the laser stripe is located, and pixel points of the line are reserved; when eta is less than 80%, the line is regarded as the line where the interference item is located, and the pixel points of the line are removed and reset to be 0; the method can effectively remove longitudinal interference and completely reserve the laser stripes.
And then the processed image is subjected to binarization and reverse color processing. And extracting the laser stripes at the geometric center of the existing laser stripes to obtain the characteristic stripes of the welding seams.
Scanning the characteristic stripes of the welding seams from left to right to extract left and right break points of discontinuous positions at the center of the image, wherein the judgment conditions of the left and right break points are as follows:
the left breakpoint determination condition is:
left breakpoint coordinate is noted as (x)1,y1)
The right breakpoint judgment condition is:
the right breakpoint coordinate is noted as (x)2,y2)
The coordinates of the center point of the welding seam are calculated by utilizing the coordinates of the left breakpoint and the right breakpoint
In specific implementation, as shown in fig. 2, the device is a real-time image acquisition device, and comprises a welding plate 1, a vision sensor, a welding gun 3 and an industrial robot arm end 5. The visual sensor comprises a linear structured light generator 2, a CMOS industrial camera 4, an optical lens 6, a narrow-band filter 7, a dimmer 8, an arc baffle 9 and a shell. The line structure light generator 2 is installed on the visual sensor shell by adopting a laser generator, and the whole visual sensor is fixed with the welding gun 3 through a clamp. The line laser emitted by the laser generator 2 irradiates on the welding plate 1, so that the laser stripe deformation area is in the optimal shooting area of the CMOS industrial camera 4, and the image during welding is collected in real time.
As shown in fig. 4, on the basis of the welding image acquired by the vision sensor, according to the pixel matrix, as shown in fig. 3, the pixel difference between two adjacent pixels, i.e. the gradient of the pixel in the y direction, is first obtained for each column by using column scanning:
preserving absolute value of gradientThe maximum pixel value f (x, y), and the pixel value on the ith column is denoted as fi(x,y)。
The average of the pixel values of all columns is calculated as the boundary threshold H:
then, scanning the image in a row, finding out specific coordinates a and b with the pixel value larger than H in the y direction, and obtaining the height position of the center line of a certain row of laser stripes:
wherein y isiRepresents the ordinate of the i point in the scanned column; b represents the ordinate of the first pixel value greater than the threshold H point scanned onto the column; a represents the ordinate of the last pixel value scanned onto the column above the threshold H point; (a-b +1) represents the width of the laser stripe in the column;
after scanning, the y value of each row can be obtained, and the average value is obtained to obtain the longitudinal coordinate value y of the central point of the ROI areaROI。
Scanning the central line, and recording the point abscissa x with the pixel value smaller than the threshold value H in each linejThe line scanning formula is as follows:
wherein xjThe abscissa representing the j point in the scanned line; d represents the abscissa scanned to the point on the row where the first pixel value is less than the threshold H; c denotes the last pixel value scanned onto the lineThe abscissa of the point less than the threshold H; (c-d +1) represents the width of the laser stripe defect in the row;
obtaining the x value of each line after scanning, and obtaining the abscissa position x of the central point of the ROI by averaging the x valuesROI。
Thereby obtaining the coordinates (x) of the center point of the ROI regionROI,yROI) A rectangular region of 300 pixels × 400 pixels centered on the center point coordinate position is extracted as the ROI region.
As shown in fig. 5, the extracted ROI region is subjected to median filtering to remove salt and pepper noise and smooth the laser stripe. And performing morphological processing on the filtered image.
And calculating the proportion of the pixel points larger than the threshold value H in each row to the total number of the pixel points in the row and recording as eta. When eta is larger than 80%, the laser stripe is regarded as the line where the laser stripe is located, and pixel points of the line are reserved; when eta is less than 80%, the line is regarded as the line where the interference item is located, and the pixel points of the line are removed and reset to be 0; the method can effectively remove longitudinal interference and completely reserve the laser stripes.
And then the processed image is subjected to binarization and reverse color processing. And extracting the laser stripes at the geometric center of the existing laser stripes to obtain the characteristic stripes of the welding seams.
Scanning the characteristic stripes of the welding seams from left to right to extract left and right break points of discontinuous positions at the center of the image, wherein the judgment conditions of the left and right break points are as follows:
the left breakpoint determination condition is:
left breakpoint coordinate is noted as (x)1,y1)
The right breakpoint judgment condition is:
the right breakpoint coordinate is noted as (x)2,y2)
The coordinates of the center point of the welding seam are calculated by utilizing the coordinates of the left breakpoint and the right breakpoint
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (5)
1. A self-adaptive extraction method for geometric characteristics of a butt weld comprises a visual sensor, wherein the visual sensor is integrally fixed on a welding gun through a clamp and moves along with the welding gun; the method is characterized in that: the method comprises the following steps:
step 1, collecting a real-time image of a butt welding seam during welding by using a vision sensor;
step 2, preprocessing the acquired real-time image of the butt weld, and extracting an ROI (region of interest) area during welding;
step 3, processing the image of the ROI during the extracted welding, removing interference and smoothing the image;
and 4, extracting a laser center line from the image processed in the step 3 and calculating the position of the center point of the welding line.
2. The adaptive extraction method of the geometric features of the butt weld according to claim 1, characterized by comprising the following steps: the implementation of step 1 comprises: the visual sensor integrated line structure light generator, CMOS industrial camera, optical lens, narrow band filter, dimmer and arc baffle, its body adopts 3D printed high temperature resistant resin shell; the CMOS industrial camera is arranged on one side of the shell, and the dimmer and the narrow-band filter are arranged below the optical lens; the linear structured light generator and the CMOS industrial camera form an included angle of 20-45 degrees, so that the laser stripes generate deformation information at welding seams, and the deformation view field of the laser stripes is positioned right below the industrial camera and used for collecting real-time images of welding.
3. The adaptive extraction method of the geometric features of the butt weld according to claim 1, characterized by comprising the following steps: step 2, the ROI extraction method comprises the following steps:
step 2.1, scanning the collected real-time image of the butt-joint weld in rows, and solving the pixel difference between two adjacent pixels for each row, namely the gradient of the pixels in the y direction:
preserving absolute value of gradientThe maximum pixel value f (x, y), and the pixel value on the ith column is denoted as fi(x,y);
Step 2.2, calculating the average value of the pixel values of all the columns as a boundary threshold value H:
step 2.3, performing column scanning on the collected real-time image of the butt-joint weld, finding out specific coordinates a and b with pixel values larger than H in the y direction, and obtaining the height position of the center line of a certain column of laser stripes:
wherein y isiRepresents the ordinate of the i point in the scanned column; b represents the ordinate of the first pixel value greater than the threshold H point scanned onto the column; a represents the ordinate of the last pixel value scanned onto the column above the threshold H point; (a-b +1) represents the width of the laser stripe in the column;
get each after scanningThe y values of the rows are averaged to obtain a longitudinal coordinate value y of the center point of the ROI areaROI;
Step 2.4, scanning the central line, and recording the point abscissa x of which the pixel value is less than the threshold value H in each linejThe line scanning formula is as follows:
wherein xjThe abscissa representing the j point in the scanned line; d represents the abscissa scanned to the point on the row where the first pixel value is less than the threshold H; c represents the abscissa of the last pixel value scanned onto the line to be less than the threshold H; c-d +1 represents the width of the laser stripe defect in the row;
obtaining the x value of each line after scanning, and obtaining the abscissa position x of the central point of the ROI by averaging the x valuesROI;
Step 2.5, obtaining the coordinates (x) of the center point of the ROI area through the step 2.3 and the step 2.4ROI,yROI) A rectangular region of 300 pixels × 400 pixels centered on the center point coordinate position is extracted as the ROI region.
4. The adaptive extraction method of the geometric features of the butt weld according to claim 1, characterized by comprising the following steps: the implementation of step 3 comprises the following steps:
3.1, performing median filtering processing on the extracted ROI to remove salt and pepper noise and smooth laser stripes; performing morphological processing on the image subjected to the median filtering processing;
step 3.2, calculating the proportion of the pixel points larger than the threshold value H in each row to the total number of the pixel points in each row and recording the proportion as eta; when eta is larger than 80%, the laser stripe is located in the row, and pixel points of the row are reserved; and when the eta is less than 80 percent, removing the pixel points of the line and resetting the pixel points to be 0.
5. The adaptive extraction method of the geometric features of the butt weld according to claim 1, characterized by comprising the following steps: the implementation of step 4 comprises the following steps:
step 4.1, carrying out binarization and reverse color processing on the image processed in the step 3.2; extracting the laser stripes at the geometric center of the existing laser stripes as characteristic stripes of the welding seams;
step 4.2, scanning the characteristic stripes of the welding seams from left to right, and extracting left and right breakpoints of discontinuous positions at the center of the image;
step 4.2.1, judging conditions according to the left breakpoint:
obtain the left breakpoint coordinate (x)1,y1);
Step 4.2.2, judging conditions according to the right breakpoint:
get the right breakpoint coordinate (x)2,y2);
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110899874.5A CN113649672A (en) | 2021-08-06 | 2021-08-06 | Adaptive extraction method for geometric characteristics of butt weld |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110899874.5A CN113649672A (en) | 2021-08-06 | 2021-08-06 | Adaptive extraction method for geometric characteristics of butt weld |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113649672A true CN113649672A (en) | 2021-11-16 |
Family
ID=78490482
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110899874.5A Pending CN113649672A (en) | 2021-08-06 | 2021-08-06 | Adaptive extraction method for geometric characteristics of butt weld |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113649672A (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105571502A (en) * | 2015-12-29 | 2016-05-11 | 上海交通大学 | Measuring method of weld gap in friction-stir welding |
CN106971407A (en) * | 2017-02-16 | 2017-07-21 | 浙江工业大学 | A kind of weld seam three-dimensional rebuilding method based on two-dimensional wire structure light |
CN107798330A (en) * | 2017-11-10 | 2018-03-13 | 上海电力学院 | A kind of weld image characteristics information extraction method |
CN109226967A (en) * | 2018-07-25 | 2019-01-18 | 同高先进制造科技(太仓)有限公司 | A kind of steady seam tracking system of active laser vision for Laser-Arc Hybrid Welding |
CN110064819A (en) * | 2019-05-14 | 2019-07-30 | 苏州实创德光电科技有限公司 | The extraction of cylinder longitudinal seam characteristic area, welding seam tracking method and system based on structure light |
CN110681950A (en) * | 2019-08-29 | 2020-01-14 | 北京航空航天大学 | Robot weld joint tracking system based on structured light sensor and control method |
CN110930368A (en) * | 2019-10-31 | 2020-03-27 | 中船重工鹏力(南京)智能装备系统有限公司 | Method for extracting characteristics of real-time welding image of sheet lap weld |
CN111402239A (en) * | 2020-03-17 | 2020-07-10 | 山东奥太电气有限公司 | Laser welding seam tracking image processing method and system based on morphological feature filtering |
CN111402323A (en) * | 2020-03-31 | 2020-07-10 | 北京博清科技有限公司 | Butt weld joint identification method, device, equipment and storage medium |
CN113042863A (en) * | 2021-03-31 | 2021-06-29 | 山东齐星铁塔有限公司 | Weld joint real-time tracking method based on laser vision sensor |
-
2021
- 2021-08-06 CN CN202110899874.5A patent/CN113649672A/en active Pending
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105571502A (en) * | 2015-12-29 | 2016-05-11 | 上海交通大学 | Measuring method of weld gap in friction-stir welding |
CN106971407A (en) * | 2017-02-16 | 2017-07-21 | 浙江工业大学 | A kind of weld seam three-dimensional rebuilding method based on two-dimensional wire structure light |
CN107798330A (en) * | 2017-11-10 | 2018-03-13 | 上海电力学院 | A kind of weld image characteristics information extraction method |
CN109226967A (en) * | 2018-07-25 | 2019-01-18 | 同高先进制造科技(太仓)有限公司 | A kind of steady seam tracking system of active laser vision for Laser-Arc Hybrid Welding |
WO2020020113A1 (en) * | 2018-07-25 | 2020-01-30 | 同高先进制造科技(太仓)有限公司 | Active laser vision weld tracking system and weld position detecting method |
CN110064819A (en) * | 2019-05-14 | 2019-07-30 | 苏州实创德光电科技有限公司 | The extraction of cylinder longitudinal seam characteristic area, welding seam tracking method and system based on structure light |
CN110681950A (en) * | 2019-08-29 | 2020-01-14 | 北京航空航天大学 | Robot weld joint tracking system based on structured light sensor and control method |
CN110930368A (en) * | 2019-10-31 | 2020-03-27 | 中船重工鹏力(南京)智能装备系统有限公司 | Method for extracting characteristics of real-time welding image of sheet lap weld |
CN111402239A (en) * | 2020-03-17 | 2020-07-10 | 山东奥太电气有限公司 | Laser welding seam tracking image processing method and system based on morphological feature filtering |
CN111402323A (en) * | 2020-03-31 | 2020-07-10 | 北京博清科技有限公司 | Butt weld joint identification method, device, equipment and storage medium |
CN113042863A (en) * | 2021-03-31 | 2021-06-29 | 山东齐星铁塔有限公司 | Weld joint real-time tracking method based on laser vision sensor |
Non-Patent Citations (3)
Title |
---|
张万江等: "基于结构光视觉的激光拼焊焊缝质量检测方法研究", 《机械设计与制造》 * |
许敏等: "不等厚激光拼焊板焊缝质量检测图像处理方法", 《焊接技术》 * |
黄色吉等: "实时跟踪焊缝特征的感兴趣区域特征提取算法", 《上海交通大学学报》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN210046133U (en) | Welding seam visual tracking system based on laser structured light | |
CN109226967B (en) | Active laser vision steady weld joint tracking system for laser-arc hybrid welding | |
CN109272544B (en) | Structured light three-dimensional measurement model of pipeline all-position welding line and image processing method | |
CN108637435B (en) | Three-dimensional weld tracking system and method based on vision and arc pressure sensing | |
US4616121A (en) | Automatic welding with imaging of workpiece surfaces and of the junction between the surfaces | |
CN107424144B (en) | Laser vision-based weld joint tracking image processing method | |
Guo et al. | Weld deviation detection based on wide dynamic range vision sensor in MAG welding process | |
CN112529858A (en) | Welding seam image processing method based on machine vision | |
CN113798634B (en) | Method, system and equipment for teaching spatial circular weld and tracking weld | |
CN110930368B (en) | Method for extracting characteristics of real-time welding image of sheet lap weld | |
GB2131571A (en) | Automatically following a sensed path for welding | |
CN110717872A (en) | Method and system for extracting characteristic points of V-shaped welding seam image under laser-assisted positioning | |
CN113269762B (en) | Screen defect detection method, system and computer storage medium | |
CN112365439B (en) | Method for synchronously detecting forming characteristics of GMAW welding seam of galvanized steel and direction of welding gun in real time | |
Zhang et al. | Narrow-seam identification and deviation detection in keyhole deep-penetration TIG welding | |
US20230186516A1 (en) | Method and flat bed machine tool for detecting a fitting position of a supporting bar | |
CN108788467A (en) | A kind of Intelligent Laser welding system towards aerospace structural component | |
CN112862794A (en) | Fillet weld appearance detection method based on structured light | |
CN112191993A (en) | Argon arc welding seam tracking system and method | |
CN111402239B (en) | Laser welding seam tracking image processing method and system based on morphological feature filtering | |
CN110625644A (en) | Workpiece grabbing method based on machine vision | |
CN111397529A (en) | Complex surface shape detection method based on binocular vision structured light | |
Lin et al. | Intelligent seam tracking of an ultranarrow gap during K-TIG welding: a hybrid CNN and adaptive ROI operation algorithm | |
CN113551599A (en) | Welding seam position deviation visual tracking method based on structured light guidance | |
CN113649672A (en) | Adaptive extraction method for geometric characteristics of butt weld |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211116 |