CN108344693B - Automatic welding-oriented visual measurement method for misalignment of sheet welding seam - Google Patents
Automatic welding-oriented visual measurement method for misalignment of sheet welding seam Download PDFInfo
- Publication number
- CN108344693B CN108344693B CN201810151824.7A CN201810151824A CN108344693B CN 108344693 B CN108344693 B CN 108344693B CN 201810151824 A CN201810151824 A CN 201810151824A CN 108344693 B CN108344693 B CN 108344693B
- Authority
- CN
- China
- Prior art keywords
- welding
- image
- value
- misalignment
- gradient
- 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.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/9515—Objects of complex shape, e.g. examined with use of a surface follower device
Abstract
The invention provides a visual measurement device and method for misalignment of a welding line of a thin plate for automatic welding, wherein the device comprises: the welding gun comprises a supporting structure body, an image acquisition unit for acquiring image information at a butt joint of welding seams, a backlight source device for providing a stable optical environment for the image acquisition unit, a clamping device for clamping a welding workpiece, an installation mechanism for controlling the swinging of a welding gun, an image processing platform for receiving and processing the image information and a fixing mechanism; the fixing mechanism, the image acquisition unit and the backlight source device are sequentially arranged from top to bottom and are arranged on one side of the supporting structure body; the mounting mechanism and the clamping device are sequentially arranged from top to bottom and are mounted on the other side of the supporting structure body; the image acquisition unit is connected with the image processing platform. The vision measuring device provided by the invention has the advantages that the image acquisition unit is used for acquiring the image at the welding seam, the image is processed in real time through the misalignment amount extraction algorithm, and the high-precision and accurate measurement of the misalignment amount of the welding seam is realized.
Description
Technical Field
The invention belongs to the field of vision measurement, and particularly relates to a vision measurement method for misalignment of a welding line of a thin plate for automatic welding.
Background
With the development of automatic welding technology and image processing technology, automatic welding systems based on visual perception have been widely applied to a plurality of industrial fields such as automobile manufacturing, aerospace, power electronics and the like. At present, in the field of manufacturing of transformer oil conservators, metal corrugated oil conservators have occupied an important market share in power equipment in China, but the metal corrugated oil conservators are hollow inside and have annular wave-shaped designs outside, so that a plurality of technical problems exist in realizing automatic welding of thin plates of corrugated pipe parts, and large-scale automatic welding industrial production of the corrugated oil conservators is not realized in the field. In fact, the scheme of manual welding has high technical requirements on welding workers, the labor cost is high, the construction difficulty is high, large-scale industrial production is not facilitated, and a system capable of realizing high-precision automatic welding is urgently needed in the field.
In the traditional method, the welding of the sheets of the corrugated pipe parts of the oil conservator needs to be carried out by spot welding in a manual welding mode, but small gaps and butt joint misalignment errors of millimeter-scale errors still remain at intervals of welding spots, so that the operation difficulty of manual welding is greatly increased, and in addition, the method for welding by workers through respective experience has certain uncertainty and welding defects are easy to occur. According to the technical requirements of welding the outer body of the industrial corrugated oil conservator, the automatic welding system needs to measure accurate misalignment before welding, and the measurement accuracy needs to be controlled below 1.0mm, so that the automatic welding system can accurately control the deflection angle and the deflection direction of a welding gun according to the misalignment of the welding line at the butt joint of the oil conservator, and the automatic welding of the metal corrugated oil conservator is realized.
Disclosure of Invention
Technical problem to be solved
The invention provides a visual measurement method for the misalignment amount of a thin plate welding seam facing automatic welding, aiming at solving the problem of inaccurate measurement of the misalignment amount of the welding seam in the industrial application field.
(II) technical scheme
In order to achieve the purpose, the invention adopts the main technical scheme that:
a visual measurement device for misalignment of thin plate welding seams facing automatic welding comprises: the welding gun comprises a supporting structure body, an image acquisition unit for acquiring image information at a butt joint of welding seams, a backlight source device for providing a stable optical environment for the image acquisition unit, a clamping device for clamping a welding workpiece, an installation mechanism for controlling the swinging of a welding gun, an image processing platform for receiving and processing the image information and a fixing mechanism for being connected with other equipment; the fixing mechanism, the image acquisition unit and the backlight source device are sequentially arranged from top to bottom and are arranged on one side of the supporting structure body; the mounting mechanism and the clamping device are sequentially arranged on the other side of the supporting structure body from top to bottom; the image acquisition unit is connected with the image processing platform.
Further, the image acquisition unit comprises a first image acquisition device and a second image acquisition device; the first image acquisition device and the second image acquisition device both comprise mounting bases and endoscope cameras, the mounting bases are connected with the supporting structure body, the mounting bases are provided with cylindrical through holes, the endoscope cameras pass through the cylindrical through holes and are mounted in the mounting bases, the endoscope cameras pass through circular openings of the cylindrical through holes and acquire image information of butt joints of welding seams, and the image information area is circular.
Further, the first image acquisition device and the second image acquisition device are symmetrical left and right about a center line of the measuring device; the first image acquisition device is used for acquiring image information of one side of a butt joint of the welding seam; the second image acquisition device is used for acquiring image information of the other side of the butt joint of the welding seam; the adjusting range of the included angle between the endoscope cameras of the first image acquisition device and the second image acquisition device and the horizontal line is more than or equal to 10 degrees and less than or equal to 20 degrees.
Furthermore, the image processing platform is used for receiving and processing the image information of the welding seam butt joint transmitted by the image acquisition unit and extracting the misalignment measurement value of the welding seam butt joint.
Further, the clamping device is used for clamping the welding workpiece, and comprises a compressor and a clamping piece.
Further, the mounting mechanism comprises a motor mounting seat, a transmission middle shaft and a swinging component which are sequentially arranged from top to bottom; the swinging component is used for mounting a welding gun.
A method of the automatic welding-oriented visual measurement device for the misalignment of the sheet welding seam comprises the following steps:
s01: adjusting the position of the image acquisition unit away from the center of the welding seam, adjusting the distance and the inclination angle of the first image acquisition device away from the butt joint of the welding seam, and adjusting the distance and the inclination angle of the second image acquisition device away from the butt joint of the welding seam;
s02: starting the clamping device, enabling the clamping piece of the clamping device to clamp the welding seam, starting the first image acquisition device and the second image acquisition device, and acquiring image information of two sides of the butt joint of the welding seam;
s03: and extracting the misalignment amount of the welding seam from the image information by utilizing a welding seam misalignment amount extraction algorithm in the image processing platform.
Further, the weld joint misregistration extraction algorithm in S03 includes an initialization program module, a modified Canny edge detection algorithm, and an image column scanning algorithm; the initialization program module specifically includes:
l01: setting an ROI (region of interest) region for the initialized image information to obtain setting parameters of the ROI region;
l02: determining a measuring point of the weld joint misalignment in the ROI, aligning the measuring point, and acquiring a positioning parameter of the measuring point;
l03: calculating a height difference conversion coefficient through the measured number of pixels of the misalignment amount at the measuring point and the physical measurement value of the misalignment amount at the measuring point measured by the vernier caliper;
l04: and adjusting the threshold value of a Canny operator to obtain two threshold value parameters of the Canny operator.
Further, the improved Canny edge detection algorithm specifically comprises:
m01: preprocessing the image acquired by the image acquisition unit by adopting Gaussian filtering and median filtering to obtain a binary gray image suitable for Canny operator processing;
m02: processing the binary gray level image by using a Canny edge detection operator, and calculating to obtain the horizontal gradient G of each pixel in the binary gray level imagexVertical gradient GyAnd a gradient direction;
m03: according to the horizontal gradient GxVertical gradient GyCalculating gradient values, comparing the gradient values at two sides of a butt joint of the welding seam when the calculated gradient value is larger than a set first pixel threshold value, and keeping the gradient value of the current measuring point unchanged if the gradient value of the current measuring point is larger than the gradient values of other measuring points in the same direction; if the gradient value of the current measuring point is not larger than the gradient values of other measuring points in the same direction, setting the gradient value of the current measuring point to be 0, which is non-maximum suppression;
m04: and judging the gradient value according to the set upper limit threshold and lower limit threshold.
Further, the image column scanning algorithm specifically includes:
n01: scanning the measuring points from top to bottom and from bottom to top respectively based on the image column where the ROI measuring points are located, determining and recording coordinates of an upper edge measuring point and a lower edge measuring point on one side of the welding line, and obtaining coordinates of the upper edge measuring point and the lower edge measuring point on the other side of the welding line based on the same method;
n02: calculating the pixel height difference of the measuring points according to the obtained coordinates of the measuring points on the two sides of the welding line of the thin plate, then calculating the misalignment value of the measuring points according to the height difference conversion coefficient, judging whether the misalignment value is larger than the historical record value or not, updating the extreme value of the misalignment value if the misalignment value is larger than the historical record value, and storing the new extreme value.
(III) advantageous effects
The invention has the beneficial effects that:
the invention provides a visual measurement device and a visual measurement method for misalignment of a welding line of a thin plate for automatic welding, wherein the visual measurement device comprises: the fixing mechanism, the image acquisition unit and the backlight source device are arranged on one side of the supporting structure body in sequence from top to bottom; and the mounting mechanism and the clamping device are sequentially arranged from top to bottom and are mounted on the other side of the supporting structure body. The measuring device adopts the endoscope camera with adjustable inclination angle to acquire the image of the welding seam area of the corrugated oil conservator with a plurality of grooves with the distance of about 2.0cm near the splicing position of the corrugated pipe of the corrugated oil conservator, and the welding seam image meeting the precision requirement can be obtained within the tolerable error range. And a clamping device is adopted to eliminate narrow gaps between welding spots at the butt joint, so that the extraction precision of the misalignment of the welding line is further ensured.
The image acquisition unit is used for acquiring the image of the welding seam, and then the image is processed through a welding seam misalignment amount extraction algorithm in the image processing platform, so that a stable processing effect and a high-precision welding seam misalignment amount measurement value can be obtained, and the measurement precision can be controlled below 1.0 mm. The measuring device disclosed by the invention is simple and convenient to operate, has good stability of the welding seam misalignment extraction algorithm, can meet the technical requirements of actual scenes, has the advantages of high detection speed, strong stability, high measurement precision, low production cost and the like, fills up the technical blank in the field, and provides possibility for automatic and efficient welding and large-scale production of the corrugated oil conservator.
Meanwhile, the measuring device can be connected with an automatic welding system through the mounting mechanism, and the automatic welding system accurately controls the deflection degree and the direction of the welding gun according to the measurement data of the measuring device, so that accurate welding is realized.
Drawings
FIG. 1 is a front view of an apparatus for visually measuring misalignment of a welding seam of a thin plate facing automatic welding according to an embodiment of the present invention;
FIG. 2 is a structural side view of an apparatus for visually measuring misalignment of a welding seam of a thin plate facing automatic welding according to an embodiment of the present invention;
FIG. 3 is a rear view of a structure of an automatic welding-oriented visual measurement device for measuring misalignment of a thin plate weld joint according to an embodiment of the present invention;
fig. 4 is a flowchart of an misalignment amount extraction algorithm of an automatic welding-oriented visual measurement method for a welding seam misalignment amount of a thin plate according to an embodiment of the present invention.
[ description of reference ]
1: a support structure body; 2: a fixing mechanism; 31a-31 b: an endoscope camera; 32a-32b, a mounting base; 4; the clamping device 5: a backlight device; 6: a motor mounting seat; 7: a welding gun; 8: a transmission middle shaft; 9: a swinging member.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
Example 1
An apparatus for visually measuring misalignment of a weld of a thin plate facing automatic welding, as shown in fig. 1-3, comprises: the welding line butt joint welding device comprises a supporting structure body 1, an image acquisition unit for acquiring image information of a butt joint of a welding line, a backlight source device 5 for providing a stable optical environment for the image acquisition unit, a clamping device 4 for clamping a welding workpiece, a mounting mechanism for controlling the swinging of a welding gun 7, and an image processing platform for receiving and processing the image information, wherein the image processing platform is positioned in a server (not shown in the drawing) and a fixing mechanism 2 for being connected with other equipment; the supporting structure body 1 is a fixed parent body of other components and parts, and the supporting structure body 1 in the embodiment is formed by combining and linking a plurality of strip-shaped rigid members. The fixing mechanism 2, the image acquisition unit and the backlight device 5 are sequentially arranged from top to bottom and are arranged on one side of the supporting structure body 1; the mounting mechanism and the clamping device 4 are sequentially arranged from top to bottom and mounted on the other side of the supporting structure body 1; the image acquisition unit is connected with the image processing platform.
The image acquisition unit comprises a first image acquisition device and a second image acquisition device; the first image acquisition device and the second image acquisition device respectively comprise mounting bases 32a and 32b and endoscope cameras 31a and 31b, the mounting bases 32a and 32b are movably connected with the supporting structure body 1, the inclination angles of the mounting bases 32a and 32b and the supporting structure body 1 can be adjusted, the mounting bases 32a and 32b are provided with cylindrical through holes, the endoscope cameras 31a and 31b are mounted inside the mounting bases 32a and 32b through the cylindrical through holes, the endoscope cameras 31a and 31b acquire image information of butt joints of welding seams through circular openings of the cylindrical through holes, and the image information area is circular.
The first image acquisition device and the second image acquisition device are bilaterally symmetrical about the center line of the measuring device; the first image acquisition device is used for acquiring image information of one side of a butt joint of the welding seam; the second image acquisition device is used for acquiring image information of the other side of the butt joint of the welding seam; the adjusting range of the included angles between the endoscope cameras 31a and 31b of the first image acquisition device and the second image acquisition device and the horizontal line is more than or equal to 10 degrees and less than or equal to 20 degrees, the height distance between the endoscope cameras 31a and 31b and the welding line is kept within the range of 2.8-3cm, the distance between the endoscope cameras and the central axis of the supporting structure body 1 is 6cm, meanwhile, in order to ensure the quality of video images acquired by the endoscope cameras 31a and 31b, a backlight source device 5 is arranged right above an image acquisition unit and used for providing stable optical environment for the endoscope cameras 31a and 31b to acquire the video images, the backlight source device 5 is fixedly connected with the center axis of the supporting structure body 1, in the embodiment, the backlight source device 5 is an output adjustable single-path strip-shaped backlight source, the output voltage of the light source is adjustable, so that the illumination brightness can be adjusted, a reasonable welding line processing, 10cm above 31 b.
The image processing platform is used for receiving and processing the image information of the butt joint of the welding seam transmitted by the image acquisition unit and extracting the misalignment measurement value of the butt joint of the welding seam.
The clamping device 4 is used for clamping the welding workpiece, and the clamping device 4 includes a compressor (not shown in the figure) and a clamping piece, in this embodiment, the clamping device 4 is located at the lowest part of the supporting structure body 1, and the compressor located at the lowest part of the clamping device 4 is a pneumatic high-pressure air compressor.
The mounting mechanism comprises a motor mounting seat 6, a transmission middle shaft 8 and a swinging component 9 which are sequentially arranged from top to bottom, wherein the swinging component 9 is used for mounting a welding gun 7.
A visual measurement method for misalignment of a thin plate welding seam facing automatic welding comprises the following steps:
s01: the position of the image acquisition unit of the well-adjusted vision measuring device from the center of the welding seam comprises: adjusting the distance and the inclination angle between the first image acquisition device and the butt joint of the welding seam, and adjusting the distance and the inclination angle between the second image acquisition device and the butt joint of the welding seam; the height of two endoscope cameras 31a and 31b from the horizontal position of a welding seam is 1cm, and the inclination angles of the endoscope cameras 31a and 31b and the horizontal plane are 10 degrees;
s02: starting a clamping device of the vision measuring device, opening a switch of a pneumatic type high-pressure air compressor, opening a switch of a control pneumatic valve after the high-pressure air pressure reaches a standard value, enabling clamping pieces of the clamping device to generate pressure to clamp a thin wall of a welding line, enabling a gap between welding points to be closed, starting endoscope cameras 31a and 31b of a first image acquisition device and a second image acquisition device, and acquiring image information of two sides of a butt joint of the welding line;
s03: and extracting the misalignment amount of the welding seam from the image information by using a welding seam misalignment amount extraction algorithm in the image processing platform.
The weld joint misalignment amount extraction algorithm in the S03 includes an initialization program module, a modified Canny edge detection algorithm and an image column scanning algorithm, and as shown in fig. 4, a flowchart of the misalignment amount extraction algorithm is shown.
The software environment of this embodiment is a WINDOWS 7 system, and an OpenCV (Open Source computer vision Library, abbreviated as OpenCV) computer vision Library is used for image processing.
The initialization operation controlled by the initialization program module specifically includes:
l01: setting a region of interest (ROI) region for the initialized image information to obtain ROI region setting parameters, wherein the original resolution of the initialized image information is 640 × 480, the ROI region is a strip-shaped region which is 640 × 280 and has a fixed size of the acquired initialized image information, and the position of the ROI region can be adjusted by an initialization program module. Because the outer body surface of the corrugated oil conservator is provided with a plurality of grooves with the distance of 2.0cm, and the outer body surface is very smooth, the reflected light of the outer body surface is strong under the irradiation of a backlight source, the non-welding seam area of the collected image has strip-shaped light spots, and the arrangement of the ROI area can eliminate the interference of the light spots and improve the processing efficiency of an image processing algorithm;
l02: determining a measuring point of the weld joint misalignment in the ROI area, aligning the measuring point, namely corresponding the positions of the weld joint misalignment detecting points in the endoscope cameras 31a and 31b to belong to the same position on two sides of a weld joint, and acquiring positioning parameters of the measuring point;
l03: calculating a height difference conversion coefficient through the measured number of pixels of the misalignment amount at the measuring point and the physical measurement value of the misalignment amount at the measuring point measured by the vernier caliper;
l04: and (4) carrying out Canny operator threshold value adjustment to obtain two threshold value parameters of the Canny operator.
The method comprises the following steps of performing weld edge detection by adopting an improved Canny edge detection algorithm, and extracting the weld edge in a weld image ROI region by combining set double-threshold parameters, wherein the method specifically comprises the following steps:
m01: carrying out image graying and image filtering enhancement operation processing on image information acquired by the endoscope cameras 31a and 31b, wherein the image filtering enhancement operation comprises mean filtering and median filtering, firstly, filtering processing is carried out on images by respectively adopting convolution check of 5 × 5 and 3 × 3 of the mean filtering, then filtering processing is carried out on the image information by adopting convolution check of 7 × 7 of the median filtering, and finally, smoothing processing is carried out on the processed images by adopting a Gaussian filter;
m02: the horizontal gradient G of each pixel in the binary gray level image is obtained through the calculation of a Canny edge detection operatorxVertical gradient GyThe optimal edge detection effect is obtained by combining the gradient direction and the double-threshold adjustment of a Canny operator, and the double-threshold parameters at the moment are recorded; the horizontal gradient G is calculatedxAnd a vertical gradient GyThen, the gradient direction can be calculated by using an inverse trigonometric function, and the gradient direction can be calculatedThe formula is as follows:
θ＝arct(Gy,Gx)
the gradient angle θ ranges from-pi to pi, and then is projected in four directions respectively representing the horizontal direction, the vertical direction, and two diagonal directions, i.e., (0 °,45 °,90 °,135 °), where non-maximum suppression is performed, and the gradient direction of each pixel is projected to the four directions according to the approximation, and further can be divided into (i ═ 1,3,5,7) four regions, the gradient angle of each region being given a specific value representing the four directions;
m03: according to a horizontal gradient GxVertical gradient GyCalculating the gradient value of each pixel, wherein the maximum value suppression is a method for edge refinement, when the calculated gradient value is greater than a set first pixel threshold value, the gradient values at two sides of a butt joint of a welding seam are compared, and if the gradient value of the current measuring point is greater than the gradient values of other measuring points in the same direction, the gradient value of the current measuring point is kept unchanged; if the gradient value of the current measuring point is not larger than the gradient values of other measuring points in the same direction, setting the gradient value of the current measuring point to be 0, which is non-maximum value suppression;
m04: judging the gradient value according to the set upper limit threshold value and the lower limit threshold value, and finally obtaining a binary image with a single-pixel weld joint edge;
the Canny operator adopts a double-threshold method to set an upper threshold and a lower threshold, the upper threshold and the lower threshold are adjustable parameters, the edge connection is controlled through a small threshold of the two thresholds, and a large threshold is used for controlling the initial segmentation of the strong edge. It is generally considered that a weak edge point and a strong edge point caused by a real edge are connected to each other, and a weak edge point caused by noise does not exist. For example, if the pixel gradient value is less than the lower threshold, it is discarded; if the pixel gradient is greater than the upper threshold, then it is considered an edge pixel; the pixel point between the upper and lower limits is called a candidate (also called a weak boundary), and we only keep when this pixel point is connected to the pixel point higher than the upper limit value, otherwise, delete. Thus, the algorithm detects 8 connected domain pixels with weak edge points. If there are strong edge points, the weak edge points are considered as true edges and are preserved. The threshold adjustment strategy can be used for guiding double-threshold adjustment in the welding line extraction process.
The image column scanning algorithm specifically comprises:
n01: scanning the measuring points from top to bottom and from bottom to top respectively based on the image column where the measuring points of the ROI are located, determining and recording the coordinates of the upper edge measuring point and the lower edge measuring point on one side of the welding line, and obtaining the coordinates of the upper edge measuring point and the lower edge measuring point on the other side of the welding line based on the same method;
n02: and calculating the pixel height difference of the measuring points according to the obtained coordinates of the measuring points on the two side surfaces of the welding line, then calculating the misalignment value of the measuring points according to the height difference conversion coefficient, storing the obtained misalignment value, judging whether the newly obtained misalignment value is greater than the historical record value, updating the extreme value of the misalignment value if the misalignment value is greater than the historical record value, and storing the new extreme value.
The measurement precision of the weld joint misalignment measured by the measurement device meets the actual application requirement, and the highest measurement precision can reach 1.0 mm.
Example 2
The vision measuring device for the misalignment of the welding line of the automatically-welded thin plate can be matched with an automatic welding system for use, and is connected with a welding gun 7 of the automatic welding system through an installation mechanism and an image processing platform of the vision measuring device.
The mounting mechanism comprises a motor mounting seat 6, a transmission middle shaft 8 and a swinging component 9 which are sequentially arranged from top to bottom, wherein the swinging component 9 is used for mounting the welding gun 7, and in the embodiment, the motor mounting seat 6 is provided with a direct-current servo motor.
In practical application, the image processing platform of the vision measuring device transmits the obtained measured values of the misalignment quantities on the two sides of the weld joint to the servo control system of the automatic welding system in real time, the control system sends out a control signal to drive the transmission middle shaft 8 to work so as to realize the left-right deflection motion of the welding gun 7, for example, when the measured value of the misalignment quantity in the visual field of the endoscope camera 31a is greater than that in the visual field of the endoscope camera 31b, the welding position of the welding gun 7 deflects to the left, and vice versa. Such a welding process can result in high quality welds with greater efficiency than manual welding methods.
In practical application, the visual measurement device enters a real-time cyclic processing state by controlling the movement of the welding seam of the corrugated oil conservator sheet, the measurement value of the misalignment amount of the whole welding seam can be measured, meanwhile, the image processing platform transmits the measurement value to the center of a servo control system of an automatic welding system in real time, and the deflection direction and the deflection amount of a welding gun are controlled by comparing the misalignment amount of two sides of the welding seam and outputting a control signal, so that the accurate control in the welding process of automatic detection and tracking of the welding seam is realized. In addition, the error amount extreme value is updated by judging whether the error amount measured value is larger than the historical value.
It should be understood that the above description of specific embodiments of the present invention is only for the purpose of illustrating the technical lines and features of the present invention, and is intended to enable those skilled in the art to understand the contents of the present invention and to implement the present invention, but the present invention is not limited to the above specific embodiments. It is intended that all such changes and modifications as fall within the scope of the appended claims be embraced therein.
Claims (2)
1. A visual measurement method for misalignment of a thin plate welding seam facing automatic welding is characterized by comprising the following steps:
s01: adjusting the position of the image acquisition unit away from the center of the welding seam, adjusting the distance and the inclination angle of the first image acquisition device away from the butt joint of the welding seam, and adjusting the distance and the inclination angle of the second image acquisition device away from the butt joint of the welding seam;
s02: starting a clamping device, enabling a clamping piece of the clamping device to clamp a welding seam, starting the first image acquisition device and the second image acquisition device, and acquiring image information of two sides of a butt joint of the welding seam;
s03: extracting the misalignment amount of the welding seam from the image information by utilizing a welding seam misalignment amount extraction algorithm in an image processing platform;
the weld joint false edge extraction algorithm in the S03 comprises an initialization program module, a modified Canny edge detection algorithm and an image column scanning algorithm;
the initialization program module specifically includes:
l01: setting an ROI (region of interest) region for the image information to obtain setting parameters of the ROI region;
l02: determining a measuring point of the weld joint misalignment in the ROI, aligning the measuring point, and acquiring a positioning parameter of the measuring point;
l03: calculating a height difference conversion coefficient through the measured number of pixels of the misalignment amount at the measuring point and the physical measurement value of the misalignment amount at the measuring point measured by the vernier caliper:
l04: carrying out Canny operator threshold value adjustment to obtain two threshold value parameters of the Canny operator;
the improved Canny edge detection algorithm specifically comprises the following steps:
m01: preprocessing the image acquired by the image acquisition unit by adopting Gaussian filtering and median filtering to obtain a binary gray image suitable for Canny operator processing;
m02: processing the binary gray level image by using a Canny edge detection operator, and calculating to obtain the horizontal gradient G of each pixel in the binary gray level imagexVertical gradient GyAnd a gradient direction;
m03: according to the horizontal gradient GxVertical gradient GyCalculating gradient values, comparing the gradient values at two sides of a butt joint of the welding seam when the calculated gradient value is larger than a set first pixel threshold value, and keeping the gradient value of the current measuring point unchanged if the gradient value of the current measuring point is larger than the gradient values of other measuring points in the same direction; if the gradient value of the current measuring point is not larger than the gradient values of other measuring points in the same direction, setting the gradient value of the current measuring point to be 0, which is non-maximum suppression;
m04: and judging the gradient value according to the set upper limit threshold and lower limit threshold.
2. The method according to claim 1, wherein the image column scanning algorithm comprises in particular:
n01: scanning the measuring points from top to bottom and from bottom to top respectively based on the image column where the ROI measuring points are located, determining and recording coordinates of an upper edge measuring point and a lower edge measuring point on one side of the welding line, and obtaining coordinates of the upper edge measuring point and the lower edge measuring point on the other side of the welding line based on the same method;
n02: calculating the pixel height difference of the measuring points according to the obtained coordinates of the measuring points on the two sides of the welding line of the thin plate, then calculating the misalignment value of the measuring points according to the height difference conversion coefficient, judging whether the misalignment value is larger than the historical record value or not, updating the extreme value of the misalignment value if the misalignment value is larger than the historical record value, and storing the new extreme value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810151824.7A CN108344693B (en) | 2018-02-14 | 2018-02-14 | Automatic welding-oriented visual measurement method for misalignment of sheet welding seam |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810151824.7A CN108344693B (en) | 2018-02-14 | 2018-02-14 | Automatic welding-oriented visual measurement method for misalignment of sheet welding seam |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108344693A CN108344693A (en) | 2018-07-31 |
| CN108344693B true CN108344693B (en) | 2020-05-05 |
Family
ID=62959955
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810151824.7A Expired - Fee Related CN108344693B (en) | 2018-02-14 | 2018-02-14 | Automatic welding-oriented visual measurement method for misalignment of sheet welding seam |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108344693B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108680647B (en) * | 2018-08-06 | 2020-10-30 | 中国计量大学 | Ultrasonic flaw detection robot for large spherical tank steel container |
| CN110276296B (en) * | 2019-06-21 | 2023-04-28 | 新疆大学 | Structured light weld joint identification method based on image processing |
| CN111208061B (en) * | 2020-02-16 | 2022-09-13 | 常熟理工学院 | Workpiece welding seam automatic visual detection device based on memristor neural network technology |
| CN111730180A (en) * | 2020-06-10 | 2020-10-02 | 北京工业大学 | Dynamic assembling and adjusting method for thin plate during welding and pressing |
| CN113996931A (en) * | 2021-11-12 | 2022-02-01 | 中国航空制造技术研究院 | Automatic adjusting device and method for laser welding butt joint and edge staggering |
| CN116021119B (en) * | 2023-03-29 | 2023-06-13 | 中建安装集团有限公司 | Petrochemical process pipeline robot welding assembly misalignment detection system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102458746A (en) * | 2009-06-10 | 2012-05-16 | 西门子公司 | Method and device for testing a weld joint for a shaft by means of a detection device introduced through a passage of the shaft |
| CN104741778A (en) * | 2015-03-30 | 2015-07-01 | 中国石油天然气集团公司 | Welding seam tracking and controlling system for combined welding |
| CN105571502A (en) * | 2015-12-29 | 2016-05-11 | 上海交通大学 | Measuring method of weld gap in friction-stir welding |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201432161Y (en) * | 2009-06-08 | 2010-03-31 | 广州粤研液压机电有限公司 | Automatic double-gun girth welding machine |
| CN104266615A (en) * | 2014-10-14 | 2015-01-07 | 上海电气集团股份有限公司 | Visual detection device and method for pipeline inner wall |
| CN105241384B (en) * | 2015-11-20 | 2017-09-29 | 南昌航空大学 | A kind of steel rail welding line misalignment measuring method based on computer vision |
| CN206764099U (en) * | 2016-12-07 | 2017-12-19 | 淮海工学院 | Optical, mechanical and electronic integration welder with CCD camera monitoring |
| CN107030435A (en) * | 2017-05-22 | 2017-08-11 | 杨煜海 | One kind compression machining capping weld jig |
| CN107020442B (en) * | 2017-05-26 | 2019-08-06 | 株洲日望精工有限公司 | Application of the arc pulling type stud welding technology in terms of welding tie pole |
| CN107378324B (en) * | 2017-08-18 | 2023-06-16 | 华南理工大学 | Welding system and method for special-shaped workpiece based on visual extraction |
-
2018
- 2018-02-14 CN CN201810151824.7A patent/CN108344693B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102458746A (en) * | 2009-06-10 | 2012-05-16 | 西门子公司 | Method and device for testing a weld joint for a shaft by means of a detection device introduced through a passage of the shaft |
| CN104741778A (en) * | 2015-03-30 | 2015-07-01 | 中国石油天然气集团公司 | Welding seam tracking and controlling system for combined welding |
| CN105571502A (en) * | 2015-12-29 | 2016-05-11 | 上海交通大学 | Measuring method of weld gap in friction-stir welding |
Non-Patent Citations (2)
| Title |
|---|
| "基于"准双目"立体视觉的球罐多层焊缝坡口轮廓检测";王军波等;《机械工程学报》;20051031;第41卷(第01期);第85-87页 * |
| "激光切焊一体化夹具设计与错边量对焊缝性能影响的研究";蔡兴;《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》;20130215(第02期);摘要、正文第45页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108344693A (en) | 2018-07-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108344693B (en) | Automatic welding-oriented visual measurement method for misalignment of sheet welding seam | |
| CN107414253B (en) | Welding seam tracking control device and method based on cross laser | |
| CN106181162B (en) | A kind of real-time weld joint tracking detection method based on machine vision | |
| CN102699534A (en) | Scanning type laser vision sensing-based narrow-gap deep-groove automatic laser multilayer welding method for thick plate | |
| CN108127217B (en) | Welding device and method for automatic guided welding of end face weld joint of spiral plate type heat exchanger | |
| CN112629441B (en) | 3D curved surface glass contour scanning detection method and system | |
| CN110220481B (en) | Handheld visual detection equipment and pose detection method thereof | |
| CN111192307A (en) | Self-adaptive deviation rectifying method based on laser cutting of three-dimensional part | |
| CN113674345B (en) | Two-dimensional pixel-level three-dimensional positioning system and positioning method | |
| CN108907526A (en) | A kind of weld image characteristic recognition method with high robust | |
| CN110966956A (en) | Binocular vision-based three-dimensional detection device and method | |
| CN115880296B (en) | Machine vision-based prefabricated part quality detection method and device | |
| CN111879264A (en) | Flatness measurement and evaluation system based on line structured light | |
| CN114434036B (en) | Three-dimensional vision system for gantry robot welding of large ship structural member and operation method | |
| CN114964007A (en) | Visual measurement and surface defect detection method for weld size | |
| CN109128540B (en) | Method for determining laser welding focal track of T-shaped joint | |
| CN111397529A (en) | Complex surface shape detection method based on binocular vision structured light | |
| CN109636859A (en) | A kind of scaling method of the 3D vision detection based on one camera | |
| CN116560062B (en) | Microscope focusing anti-collision control method | |
| CN116393982B (en) | Screw locking method and device based on machine vision | |
| CN110657750B (en) | Detection system and method for passivation of cutting edge of cutter | |
| Ye et al. | Weld seam tracking based on laser imaging binary image preprocessing | |
| CN206583440U (en) | A kind of projected image sighting distance detecting system | |
| CN112834505B (en) | Three-dimensional visual detection positioning device and method for pasted welding line of pipeline workpiece | |
| CN114111578A (en) | Automatic pose determination method for large-diameter element |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200505 Termination date: 20220214 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |