CN114049351A - Door and window welding control method and system based on artificial intelligence - Google Patents
Door and window welding control method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN114049351A CN114049351A CN202210024106.XA CN202210024106A CN114049351A CN 114049351 A CN114049351 A CN 114049351A CN 202210024106 A CN202210024106 A CN 202210024106A CN 114049351 A CN114049351 A CN 114049351A
- Authority
- CN
- China
- Prior art keywords
- welding
- shadow
- welding seam
- weld
- edge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003466 welding Methods 0.000 title claims abstract description 247
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 23
- 238000009826 distribution Methods 0.000 claims abstract description 58
- 239000000463 material Substances 0.000 claims abstract description 38
- 230000011218 segmentation Effects 0.000 claims abstract description 8
- 238000010586 diagram Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000003287 optical effect Effects 0.000 claims description 8
- 239000000126 substance Substances 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004023 plastic welding Methods 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 3
- 230000000877 morphologic effect Effects 0.000 abstract description 10
- 230000000007 visual effect Effects 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 10
- 238000010606 normalization Methods 0.000 description 5
- 229920003023 plastic Polymers 0.000 description 5
- 239000004033 plastic Substances 0.000 description 5
- 239000011324 bead Substances 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 208000010392 Bone Fractures Diseases 0.000 description 1
- 206010017076 Fracture Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 229920011532 unplasticized polyvinyl chloride Polymers 0.000 description 1
Images
Classifications
-
- 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
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- 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
- 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
- G06T2207/30—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)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a door and window welding control method and system based on artificial intelligence. The method includes the steps that three cameras with different visual angles and light sources are arranged, and a shadow splicing image with a plurality of shadow areas is obtained. And obtaining a plurality of section shadow distribution maps according to the distribution of the camera and the distribution of the material shadow area, obtaining a curvature triangle representing the morphological characteristics of the edge of the welding seam through the distribution information of the shadow segmentation points, and obtaining the roughness of the edge of the welding seam according to the difference of the curvature triangles of all the section shadow distribution maps. And obtaining the longitudinal depth of the welding seam through the area ratio of the shadow area of the deepest welding seam. And judging the quality of the welding seam according to the roughness of the edge of the welding seam and the longitudinal depth of the welding seam, obtaining the adjusting direction when the quality requirement is not met, and controlling and adjusting the welding pressure. The invention obtains the morphological characteristics of the welding line through the shadow information of the welding line, quickly and accurately judges the quality of the welding line and provides reference for welding control.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a door and window welding control method and system based on artificial intelligence.
Background
Hot plate welding is one of the common processes in the manufacturing process of UPVC plastic doors and windows. In the hot plate welding process, the welding material can be accurately positioned in a machine through the positioning plate, pressure is applied to the plastic door and window material through the pressing clamp, the end surfaces of the two sections of materials close to the welding plate are melted under the action of the pressure, the welding plate is withdrawn subsequently, the materials are butted together until the materials are cooled, and welding is completed. The welding pressure is therefore the main influencing parameter of the welding quality. Welding quality can influence door and window's security performance, because the operation among the welding process is improper or machine parameter selection improper scheduling factor all can lead to plastics door and window welding seam intensity not enough, has the risk of fracture in the follow-up use, has the potential safety hazard to the user. Therefore, the welding seams of the plastic doors and windows need to be inspected to be qualified, machine parameters are adjusted in time, and potential safety hazards of the plastic doors and windows are avoided.
In the prior art, temperature information of each position of a welding seam can be obtained through an infrared detection method, the welding seam strength is represented through the temperature information, but infrared detection can be influenced by irrelevant temperature sources, the welding seam quality cannot be accurately judged, and false detection is easy to occur.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a door and window welding control method and system based on artificial intelligence, and the adopted technical scheme is as follows:
the invention provides a door and window welding control method based on artificial intelligence, which comprises the following steps:
arranging three cameras with respective light sources above the plastic welding line to respectively obtain overlooking images of the welding line; the welding seams are distributed along the width direction of an overlooking image of the welding seams; the cameras include a forward downward-looking camera, a right downward-looking camera, and a left downward-looking camera; the cameras are distributed at equal intervals at preset intervals on a preset height;
splicing the welding seam overlook images obtained by the three cameras to obtain a shadow splicing image; the shadow mosaic image comprises a plurality of shadow regions; the shadow area comprises a weld seam shadow area and a material shadow area;
constructing a plurality of section shadow distribution maps according to the distribution of the cameras and the distribution of the material shadow regions of the pixel points of each row of the shadow splicing image; end points on two sides of the material shadow area in the cross section shadow distribution diagram are material shadow dividing points; dividing the welding seam section shadow distribution map into a left side and a right side by using the optical axis of the forward overlooking visual angle camera; connecting the shadow division points of each side of the camera in the camera view angle direction to obtain three connecting lines; taking three intersection points of the three connecting lines on each side as a curvature triangle on each side; obtaining the roughness of the edge of the weld joint according to the difference of the curvature triangles on each side of all the section shadow distribution maps;
taking the shadow area of the welding seam with the darkest color as the shadow area at the top of the welding seam; taking the area ratio of the shadow area at the top of the welding seam and the shadow area of the welding seam as the longitudinal depth of the welding seam;
obtaining a weld quality score according to the weld edge roughness and the weld longitudinal depth; if the quality score of the welding seam is smaller than a preset quality reference threshold value, judging the current welding pressure adjusting direction according to the roughness of the edge of the welding seam and the longitudinal depth of the welding seam, and adjusting the welding pressure according to the welding pressure adjusting direction; and if the quality score of the welding seam is still smaller than the quality reference threshold value after the pressure adjustment for many times, early warning is carried out.
Further, the shadow mosaic image comprising a plurality of shadow regions comprises:
performing edge detection on the shadow splicing image to obtain a plurality of shadow dividing lines; the shadow segmentation line segments the shadow mosaic image into a plurality of shadow regions.
Further, said obtaining weld edge roughness from differences in the curvature triangles on each side of all of said cross-sectional shadow profiles comprises:
acquiring three vertex coordinates of the curvature triangle on each side of the cross-section shadow distribution map; constructing a welding seam edge description matrix of each side according to the vertex coordinates of each side of all the section shadow distribution maps; the length of the welding seam edge description matrix is three, and the width of the welding seam edge description matrix is the number of the section shadow distribution maps;
acquiring a column average coordinate value of each column of the welding seam edge description matrix; taking the average value of the distances between all the vertex coordinates and the corresponding row average coordinate values as the curvature triangle difference of each side of the section shadow distribution diagram;
and obtaining the roughness of the edge of the weld joint according to the difference of the curvature triangles on the two sides.
Further, the obtaining the weld edge roughness from the difference in the curvature triangles on the two sides includes: normalizing the difference of the curvature triangles on the two sides and then obtaining the roughness of the edge of the weld joint through a roughness calculation formula; the roughness calculation formula includes:
wherein the content of the first and second substances,
for the weld edge roughness,
to normalize the difference in the curvature triangle on one side of the weld,
and normalizing the difference of the curvature triangles on the other side of the welding seam.
Further, the obtaining a weld quality score based on the weld edge roughness and the weld longitudinal depth comprises: obtaining the weld quality score according to a quality score formula; the quality scoring formula includes:
wherein the content of the first and second substances,
the quality of the weld is scored and the quality of the weld is,
for the weld edge roughness,
is the longitudinal depth of the weld.
Further, the judging the current welding pressure adjusting direction according to the weld seam edge roughness and the weld seam longitudinal depth comprises:
constructing a quality scoring coordinate system by taking the roughness of the edge of the welding seam as a vertical coordinate and the longitudinal depth of the welding seam as a horizontal coordinate; setting a reference welding seam parameter straight line in the quality scoring coordinate system; acquiring a quality coordinate of the weld quality score in the quality score coordinate system; if the mass coordinate is above the reference welding seam parameter straight line, judging the pressure adjusting direction to reduce the welding pressure; otherwise, judging the pressure adjusting direction to increase the welding pressure.
Further, the adjusting the welding pressure according to the welding pressure adjusting direction comprises:
and continuously changing the adjustment step length in a pressure adjustment interval by a bisection method according to the welding pressure adjustment direction, and obtaining the welding seam quality score by adjusting the welding pressure by the adjustment step length every time until the welding seam quality score is not less than the quality reference threshold value.
Further, if the weld quality score is still smaller than a preset quality reference threshold value after the pressure adjustment for multiple times, performing early warning comprises:
and if the adjusting step length is changed for multiple times, the adjusting step length is smaller than the preset pressure minimum adjusting step length, and the corresponding welding seam quality score is smaller than the quality reference threshold value, early warning is carried out.
The invention also provides a door and window welding control system based on artificial intelligence, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor realizes any one of the steps of the door and window welding control method based on artificial intelligence when executing the computer program.
The invention has the following beneficial effects:
1. according to the embodiment of the invention, welding seam overlook images obtained by cameras with light sources at three different visual angles are spliced to obtain shadow spliced images with abundant shadow information. Because of the arrangement of multiple light sources, different brightness distributions can be generated on different areas of the welding seam and the surrounding materials, and different shadow information can be presented. Constructing a cross section shadow distribution diagram through the distribution of the material shadow area, and further obtaining the weld seam edge roughness of each cross section; the longitudinal depth of the welding seam is obtained through the shadow distribution of the shadow area of the welding seam. The morphological characteristics of the welding seam can be accurately obtained through the shadow information with obvious characteristics.
2. According to the embodiment of the invention, the quality score of the welding seam is evaluated through the roughness of the edge of the welding seam and the longitudinal depth of the welding seam, and further, if the quality score of the welding seam is smaller than a preset quality reference threshold value, the quality of the welding seam is unqualified, and the adjustment direction of the welding pressure can be judged through the roughness of the edge of the welding seam and the longitudinal depth of the welding seam. The welding pressure adjusting direction provides accurate reference in the process of adjusting the welding pressure, so that the subsequent welding pressure adjusting process can be accurately adjusted, and welding control is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a door/window welding control method based on artificial intelligence according to an embodiment of the present invention;
FIG. 2 is a schematic illustration of a distribution of shaded regions of material provided by an embodiment of the present invention;
FIG. 3 is a cross-sectional shadow profile provided by an embodiment of the present invention.
Fig. 4 is a schematic diagram of a quality score coordinate system according to an embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description will be given to a door and window welding control method and system based on artificial intelligence according to the present invention, and the specific implementation, structure, features and effects thereof with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of a door and window welding control method and system based on artificial intelligence in detail with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a door/window welding control method based on artificial intelligence according to an embodiment of the present invention is shown, where the method includes:
step S1: three cameras with respective light sources are arranged above the plastic welding seam, and overlooking images of the welding seam are respectively obtained.
In hot plate welding process, if welding pressure is too big, can make the welding seam extrude the face of weld too fast, lead to the face of weld atress inhomogeneous, the edge unevenness of welding seam left and right sides side produces the edge of coarse form. If the welding pressure is too low, a deep and unevenly distributed concave seam appears at the top of the welding seam, and the concave seam is called a middle seam in the embodiment of the invention because the concave seam is distributed at the center of the top of the welding seam. Because the characteristic information of the rough edge and the middle seam is obvious morphological characteristics, the morphological characteristics of the welding seam can be obtained through the processing of multiple welding seam images by using an image processing method.
Three gray cameras which are provided with respective light sources and distributed at equal intervals are arranged at preset heights above the plastic welding line to respectively acquire overlooking images of the welding line. The three cameras are positioned as a left downward-looking camera, a right downward-looking camera, and a forward-looking camera, respectively. Each camera is provided with a light source, the light source is shielded by the edge of the welding seam, so that the brightness of different areas of the welding seam and surrounding materials is different, and shadow information of each area can be obtained by overlooking images of the welding seam with different visual angles. In the bead top view image, the beads are distributed in the image width direction.
Because welding machine possess the locating plate, can be with accurate the fixing on the machine of welding material, so the optical axis of positive depression camera is perpendicular to the axis of welding seam. In the hot plate welding, after the welding plate is withdrawn, pressure is applied to the material to enable the material after the melting material is fused to be called secondary feeding amount, namely the secondary feeding amount influences the height of a welding seam, and in order to ensure that the height and the width of the welding seam cannot influence the imaging of a camera, in the embodiment of the invention, the height setting section of the camera is within a range from three times to four times of the secondary feeding amount, and the interval setting section of the camera is within a range from two times to three times of the secondary feeding amount.
Step S2: and splicing the welding seam overlook images obtained by the three cameras to obtain a shadow splicing image.
Because the three cameras are respectively provided with the light sources, the same area can present different shadow effects due to the illumination of the light sources with different viewing angles. And splicing the welding seam overlook images obtained by the three cameras to obtain a shadow splicing image. The gray scale of the shadow splicing image is different from that of other areas because the partial areas are irradiated by different light sources. Therefore, a plurality of shadow regions, namely, a shadow region of the welding seam and a shadow region of the material, exist in the shadow splicing image.
In order to effectively distinguish shadow areas, edge detection is carried out on the shadow splicing image to obtain a plurality of shadow dividing lines, and the shadow dividing lines divide the shadow splicing image into a plurality of shadow areas.
Step S3: constructing a cross-section shadow distribution map according to the distribution of the cameras and the material shadow region distribution of each row of shadow splicing image pixel points; obtaining a curvature triangle of each side in the cross-section shadow distribution graph; and obtaining the roughness of the edge of the welding seam according to the difference of the curved triangles on each side of all the section shadow distribution graphs.
Due to morphological characteristics of the edge of the welding line, the distribution of a material shadow region in a shadow splicing image also has regularity, for example, due to the shielding of the edge of the welding line on a light source near the welding line body, the nearby region is only irradiated by one light source of a right depression camera or a left depression camera, and the gray scale of the shadow region is larger; the farther away from the weld, the less grey the shade is present because there is no or less shading of the weld edge. Referring to fig. 2, a schematic distribution diagram of a material shadow region provided by an embodiment of the present invention is shown, where a middle region is a weld bead body, and the other regions are material shadow regions, numbers on the material shadow regions indicate the number of light sources irradiated, and the material shadow regions are distributed to be irradiated by 1 light source, 2 light sources, and 3 light sources with the weld bead body as a center. It should be noted that, the material shadow areas at the corresponding positions on the left and right sides in fig. 2 are the same in size for convenience of description, and in practical cases, the material shadow areas at the corresponding positions on the two sides may have different sizes due to the influence of the edge of the weld seam.
If the welding seam edge presents an irregular rough edge shape, the length of the shadow area near the welding seam varies along the distribution direction of the welding seam in the shadow splicing image due to the rough size, so that the distribution of the material shadow area in the shadow splicing image can be analyzed line by line. And constructing a plurality of section shadow distribution maps according to the distribution of the cameras and the distribution of the material shadow regions of the pixel points of each row of shadow splicing images. Referring to FIG. 3, a cross-sectional shadow map according to an embodiment of the present invention is shown. In the cross-section shadow distribution diagram, the end points on the two sides of the material shadow area are material shadow dividing points. The weld cross-section shadow distribution map is divided into left and right sides by the optical axis L of the forward looking down view angle camera, and it should be noted that, for convenience, the material shadow information distribution on the left and right sides in fig. 3 is the same, and in actual situations, the shadow distribution is different due to the difference of the two side edges, and it is necessary to analyze the two side information separately.
And connecting the shadow division points of each side camera in the camera view angle direction to obtain three connecting lines. With three intersections of three connecting lines on each sideAs a curved triangle on each side. Since the distribution of the shadow dividing points, i.e., the distribution of the material shadow region, is determined by the size of the weld edge irradiated by the light source, the curved triangle composed of the three connecting lines can represent the morphological characteristics of the weld edge on each side. Taking the left side in fig. 3 as an example, camera a is a left downward-looking camera, so camera a is connected to shadow segmentation point a; the camera B is a forward-downward-looking camera, so the camera is connected with the shadow dividing point B; camera C is a right downward-looking camera, so the camera is connected to the shadow segmentation point C. By geometric analysis of camera separation
Height of camera
Distance of shadow dividing point from optical axis of forward-downward-view camera
、
And
obtaining the intersection point of the three connecting lines
、
And
the coordinates of (c) in the embodiment of the present invention specifically include:
the distance from the intersection point to the optical axis of the positive depression camera is taken as the abscissa of the intersection point, and the longitudinal height of the intersection point is taken as the ordinate of the intersection point. The calculation formula of the coordinates of the three intersection points includes:
and obtaining the left curve triangle information through the three intersection point coordinate information.
Similarly, the distance between the shadow segmentation point and the optical axis of the forward-looking camera is determined according to the camera interval x, the camera height H and the distance between the shadow segmentation point and the optical axis of the forward-looking camera
、
And
obtaining the intersection point of the three connecting lines
、
And
the coordinates of (a). Obtaining the curvature III of the right side through the coordinate information of the three intersection pointsAngle information.
Because the sectional shadow profiles are analyzed line by line of the shadow-stitched image, i.e. the number of sectional shadow profiles is the width of the shadow-stitched image. Because the curve triangles represent morphological information of the weld edge, weld edge roughness can be obtained by the curve triangle difference on each side of all cross-sectional shadow profiles. The greater the roughness of the edge of the welding seam, the more uneven the overall edge shape of the welding seam is proved, namely the greater difference exists between the curvature triangles on the two sides of the shadow distribution diagram of all the sections. The obtaining of the weld edge roughness specifically comprises:
and acquiring three vertex coordinates of the curvature triangle on each side of the sectional shadow distribution map. And constructing a welding seam edge description matrix of each side according to the vertex coordinates of each side of all the section shadow distribution maps. The weld edge description matrix is three in length and three in width, which is the number of cross-sectional shadow profiles.
And acquiring column average coordinate values of the weld edge description matrix of each column. And taking the average value of the distances between all vertex coordinates and the corresponding column average coordinate values as the difference of the curvature triangles on each side of the section shadow distribution diagram.
And normalizing the difference of the curvature triangles at the two sides in order to facilitate the calculation of subsequent indexes and limit the value interval of each index. In the embodiment of the invention, the normalization process is realized by a normalization formula, and the normalization formula comprises the following steps:
wherein the content of the first and second substances,
is the difference of the curvature triangles on one side of the welding seam after normalization. And the curvature triangle difference on the other side is normalized according to the same formula.
And normalizing the difference of the two side curvature triangles and then obtaining the roughness of the edge of the weld joint through a roughness calculation formula. The roughness calculation formula includes:
wherein the content of the first and second substances,
in order to obtain the roughness of the edge of the weld seam,
is the difference of the curvature triangles on one side of the welding seam after normalization,
and (5) normalizing the difference of the curvature triangles on the other side of the welding seam.
Step S4: taking the shadow area of the welding seam with the darkest color as the shadow area at the top of the welding seam; and taking the area ratio of the shadow area at the top of the welding seam and the shadow area of the welding seam as the longitudinal depth of the welding seam.
Because of the seam characteristics at the top of the welding seam, the shadow area of the welding seam in the shadow splicing image has regular characteristics. If the middle seam exists at the top of the welding seam, the gray value of the shadow area of the middle area of the top of the early welding seam is larger, and a certain area exists, because the secondary feed amount is fixed, the larger the area of the middle area is, the larger the area of the middle seam is, and the smaller the area is, the shallower the middle seam is. Taking the shadow area of the welding seam with the darkest color as the shadow area at the top of the welding seam; and taking the area ratio of the shadow area at the top of the welding seam and the shadow area of the welding seam as the longitudinal depth of the welding seam.
Step S5: obtaining a weld quality score according to the roughness of the edge of the weld and the longitudinal depth of the weld; if the quality score of the welding seam is smaller than a preset quality reference threshold value, judging the current welding pressure adjusting direction according to the roughness of the edge of the welding seam and the longitudinal depth of the welding seam, and adjusting the welding pressure according to the welding pressure adjusting direction; and if the quality score of the welding seam is still smaller than the quality reference threshold value after the pressure adjustment for many times, early warning is carried out.
The greater the roughness of the edge of the welding seam is, the greater the longitudinal depth of the welding seam is, the smaller the quality of the welding seam is, and the quality scoring formula is obtained by fitting through a mathematical modeling method according to the inverse proportion relation. And obtaining the weld quality score according to a quality score formula. The quality scoring formula includes:
wherein the content of the first and second substances,
and the quality of the welding seam is graded,
in order to obtain the roughness of the edge of the weld seam,
is the longitudinal depth of the weld.
If the quality score of the welding seam is not less than the preset quality reference threshold value, the quality of the current welding seam is qualified, and the welding pressure does not need to be adjusted. If the quality score of the welding seam is smaller than the preset quality reference threshold value, the quality of the current welding seam is unqualified, the welding pressure needs to be adjusted and controlled, and more unqualified welding seams are avoided. Judging the current welding pressure adjusting direction according to the welding seam edge roughness and the welding seam longitudinal depth comprises the following steps:
and constructing a quality scoring coordinate system by taking the roughness of the edge of the welding seam as a vertical coordinate and the longitudinal depth of the welding seam as a horizontal coordinate. Referring to fig. 4, a diagram of a quality score coordinate system according to an embodiment of the invention is shown. And setting a reference welding seam parameter straight line F in the quality scoring coordinate system. And acquiring a quality coordinate Q of the weld quality score in a quality score coordinate system. And if the mass coordinate is above the reference welding seam parameter straight line, judging the pressure adjusting direction to reduce the welding pressure. Otherwise, judging the pressure adjusting direction to increase the welding pressure. Because the value intervals of the roughness of the edge of the welding seam and the longitudinal depth of the welding seam are both [0,1], in the embodiment of the invention, the slope of the reference welding seam parameter straight line is 1.
It should be noted that, if the mass coordinate is on the reference weld parameter straight line, the welding pressure is finely adjusted at random.
Adjusting the welding pressure according to the welding pressure adjustment direction comprises:
and continuously changing the adjusting step length in the pressure adjusting interval by a bisection method according to the adjusting direction of the welding pressure, and obtaining the quality score of the welding seam each time the welding pressure is adjusted by the adjusting step length until the quality score of the welding seam is not less than the quality reference threshold value. Dichotomy search is a conventional technical means, and is not described herein in detail, and only the corresponding process is briefly described:
an optimal pressure interval exists in the welding process, and if the current pressure adjusting direction is pressure reduction, namely the current pressure is overlarge, the pressure is distributed on the right side of the optimal pressure interval in the pressure interval. Taking half of the difference between the current pressure and the minimum pressure as an adjusting step length to carry out first adjustment; if the quality of the welding seam still does not meet the requirement after the adjustment and the welding pressure needs to be adjusted to be reduced continuously, taking half of the difference value from the pressure after the first adjustment to the minimum pressure as a new adjustment step length to perform second adjustment; and if the quality of the weld joint still does not meet the requirement after adjustment and the welding pressure needs to be increased, namely the second adjustment pressure is on the right side of the optimal pressure interval, taking half of the difference value between the second adjustment pressure and the first adjustment pressure as a new adjustment step length until the quality of the weld joint meets the requirement.
It should be noted that, because of the limitation of the welding machine, the welding machine has the minimum adjustment step length, and if the adjustment step length is smaller than the preset pressure minimum adjustment step length and the corresponding weld quality score is still smaller than the quality reference threshold value at this time, it is indicated that the optimal pressure interval is not within the controllable range, and other parameters of the welding machine are abnormal, an early warning is performed, and related personnel are notified to timely eliminate the abnormality.
In summary, in the embodiments of the present invention, three cameras with different viewing angles and light sources are arranged to obtain a shadow mosaic image having a plurality of shadow areas. And obtaining a plurality of section shadow distribution maps according to the distribution of the camera and the distribution of the material shadow area, obtaining a curvature triangle representing the morphological characteristics of the edge of the welding seam through distribution information of shadow segmentation points on the section shadow distribution maps, and obtaining the roughness of the edge of the welding seam according to the difference of the curvature triangles of all the section shadow distribution maps. And obtaining the longitudinal depth of the welding seam through the area ratio of the shadow area of the deepest welding seam. And judging the quality of the welding seam according to the roughness of the edge of the welding seam and the longitudinal depth of the welding seam, obtaining the adjusting direction when the quality requirement is not met, controlling and adjusting the welding pressure, and early warning when the adjustment fails. The embodiment of the invention obtains the morphological characteristics of the welding line through the shadow information of the welding line, quickly and accurately judges the quality of the welding line and provides reference for welding control.
The invention also provides a door and window welding control system based on artificial intelligence, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein when the processor executes the computer program, any one step of the door and window welding control method based on artificial intelligence is realized.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. A door and window welding control method based on artificial intelligence is characterized by comprising the following steps:
arranging three cameras with respective light sources above the plastic welding line to respectively obtain overlooking images of the welding line; the welding seams are distributed along the width direction of an overlooking image of the welding seams; the cameras include a forward downward-looking camera, a right downward-looking camera, and a left downward-looking camera; the cameras are distributed at equal intervals at preset intervals on a preset height;
splicing the welding seam overlook images obtained by the three cameras to obtain a shadow splicing image; the shadow mosaic image comprises a plurality of shadow regions; the shadow area comprises a weld seam shadow area and a material shadow area;
constructing a plurality of cross-section shadow distribution maps according to the distribution of the cameras and the distribution of the material shadow regions of the pixel points of the shadow splicing image in each line; end points on two sides of the material shadow area in the cross section shadow distribution diagram are material shadow dividing points; dividing the welding seam section shadow distribution map into a left side and a right side by using the optical axis of the positive depression camera; connecting the shadow division points of each side of the camera in the camera view angle direction to obtain three connecting lines; taking three intersection points of the three connecting lines on each side as a curvature triangle on each side; obtaining the roughness of the edge of the weld joint according to the difference of the curvature triangles on each side of all the section shadow distribution maps;
taking the shadow area of the welding seam with the darkest color as the shadow area at the top of the welding seam; taking the area ratio of the shadow area at the top of the welding seam and the shadow area of the welding seam as the longitudinal depth of the welding seam;
obtaining a weld quality score according to the weld edge roughness and the weld longitudinal depth; if the quality score of the welding seam is smaller than a preset quality reference threshold value, judging the current welding pressure adjusting direction according to the roughness of the edge of the welding seam and the longitudinal depth of the welding seam, and adjusting the welding pressure according to the welding pressure adjusting direction; and if the quality score of the welding seam is still smaller than the quality reference threshold value after the pressure adjustment for many times, early warning is carried out.
2. The artificial intelligence based door and window welding control method of claim 1, wherein the shadow mosaic image comprising a plurality of shadow areas comprises:
performing edge detection on the shadow splicing image to obtain a plurality of shadow dividing lines; the shadow segmentation line segments the shadow mosaic image into a plurality of shadow regions.
3. The artificial intelligence based door and window welding control method of claim 1, wherein the obtaining of the weld edge roughness from the differences in the curvature triangles on each side of all the cross-sectional shadow profiles comprises:
acquiring three vertex coordinates of the curvature triangle on each side of the cross-section shadow distribution map; constructing a welding seam edge description matrix of each side according to the vertex coordinates of each side of all the section shadow distribution maps; the length of the welding seam edge description matrix is three, and the width of the welding seam edge description matrix is the number of the section shadow distribution maps;
acquiring a column average coordinate value of each column of the welding seam edge description matrix; taking the average value of the distances between all the vertex coordinates and the corresponding row average coordinate values as the curvature triangle difference of each side of the section shadow distribution diagram;
and obtaining the roughness of the edge of the weld joint according to the difference of the curvature triangles on the two sides.
4. The artificial intelligence based door and window welding control method of claim 3, wherein the obtaining the weld edge roughness according to the difference of the curvature triangles on the two sides comprises: normalizing the difference of the curvature triangles on the two sides and then obtaining the roughness of the edge of the weld joint through a roughness calculation formula; the roughness calculation formula includes:
wherein the content of the first and second substances,
for the weld edge roughness,
to normalize the difference in the curvature triangle on one side of the weld,
and normalizing the difference of the curvature triangles on the other side of the welding seam.
5. The artificial intelligence based door and window welding control method of claim 1, wherein the obtaining of the weld quality score according to the weld edge roughness and the weld longitudinal depth comprises: obtaining the weld quality score according to a quality score formula; the quality scoring formula includes:
wherein the content of the first and second substances,
the quality of the weld is scored and the quality of the weld is,
for the weld edge roughness,
is the longitudinal depth of the weld.
6. The artificial intelligence based door and window welding control method of claim 1, wherein the determining a current welding pressure adjustment direction according to the weld seam edge roughness and the weld seam longitudinal depth comprises:
constructing a quality scoring coordinate system by taking the roughness of the edge of the welding seam as a vertical coordinate and the longitudinal depth of the welding seam as a horizontal coordinate; setting a reference welding seam parameter straight line in the quality scoring coordinate system; acquiring a quality coordinate of the weld quality score in the quality score coordinate system; if the mass coordinate is above the reference welding seam parameter straight line, judging the pressure adjusting direction to reduce the welding pressure; otherwise, judging the pressure adjusting direction to increase the welding pressure.
7. The door and window welding control method based on artificial intelligence of claim 1, wherein the adjusting the welding pressure according to the welding pressure adjusting direction comprises:
and continuously changing the adjustment step length in a pressure adjustment interval by a bisection method according to the welding pressure adjustment direction, and obtaining the welding seam quality score by adjusting the welding pressure by the adjustment step length every time until the welding seam quality score is not less than the quality reference threshold value.
8. The door and window welding control method based on artificial intelligence of claim 7, wherein if the quality score of the welding seam is still smaller than a preset quality reference threshold value after the pressure adjustment for a plurality of times, the early warning comprises:
and if the adjusting step length is changed for multiple times, the adjusting step length is smaller than the preset pressure minimum adjusting step length, and the corresponding welding seam quality score is smaller than the quality reference threshold value, early warning is carried out.
9. An artificial intelligence based door and window welding control system, comprising a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210024106.XA CN114049351B (en) | 2022-01-11 | 2022-01-11 | Door and window welding control method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210024106.XA CN114049351B (en) | 2022-01-11 | 2022-01-11 | Door and window welding control method and system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114049351A true CN114049351A (en) | 2022-02-15 |
| CN114049351B CN114049351B (en) | 2022-03-25 |
Family
ID=80213595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210024106.XA Active CN114049351B (en) | 2022-01-11 | 2022-01-11 | Door and window welding control method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114049351B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114612457A (en) * | 2022-03-21 | 2022-06-10 | 徐州华宝能源科技有限公司 | Automobile part laser welding adjusting method and system based on computer vision |
| CN114851569A (en) * | 2022-03-31 | 2022-08-05 | 启东亦大通自动化设备有限公司 | Intelligent control method and system for welding and forming ion exchange membrane |
| CN117073937A (en) * | 2023-10-18 | 2023-11-17 | 梁山华鲁专用汽车制造有限公司 | Tightness detection system based on powder material transport vehicle tank body |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6024273A (en) * | 1998-12-04 | 2000-02-15 | Caterpillar Inc. | Method and system for determining weld bead quality |
| CN101769728A (en) * | 2010-02-02 | 2010-07-07 | 浙江大学 | Method for detecting weld areas of hot-melting polythene pipe joints |
| CN101966617A (en) * | 2010-08-30 | 2011-02-09 | 东南大学 | Brief description method for view model for continuous motion of welding robot |
| US20130119040A1 (en) * | 2011-11-11 | 2013-05-16 | Lincoln Global, Inc. | System and method for adaptive fill welding using image capture |
| CN112053376A (en) * | 2020-09-07 | 2020-12-08 | 南京大学 | Workpiece weld joint identification method based on depth information |
| CN112184704A (en) * | 2020-10-28 | 2021-01-05 | 广西科技大学 | Welding seam processing method based on convolutional neural network |
| CN112306981A (en) * | 2020-11-03 | 2021-02-02 | 广州科泽云天智能科技有限公司 | Fault prediction method for fault log of high-performance computing system |
-
2022
- 2022-01-11 CN CN202210024106.XA patent/CN114049351B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6024273A (en) * | 1998-12-04 | 2000-02-15 | Caterpillar Inc. | Method and system for determining weld bead quality |
| CN101769728A (en) * | 2010-02-02 | 2010-07-07 | 浙江大学 | Method for detecting weld areas of hot-melting polythene pipe joints |
| CN101966617A (en) * | 2010-08-30 | 2011-02-09 | 东南大学 | Brief description method for view model for continuous motion of welding robot |
| US20130119040A1 (en) * | 2011-11-11 | 2013-05-16 | Lincoln Global, Inc. | System and method for adaptive fill welding using image capture |
| CN112053376A (en) * | 2020-09-07 | 2020-12-08 | 南京大学 | Workpiece weld joint identification method based on depth information |
| CN112184704A (en) * | 2020-10-28 | 2021-01-05 | 广西科技大学 | Welding seam processing method based on convolutional neural network |
| CN112306981A (en) * | 2020-11-03 | 2021-02-02 | 广州科泽云天智能科技有限公司 | Fault prediction method for fault log of high-performance computing system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114612457A (en) * | 2022-03-21 | 2022-06-10 | 徐州华宝能源科技有限公司 | Automobile part laser welding adjusting method and system based on computer vision |
| CN114851569A (en) * | 2022-03-31 | 2022-08-05 | 启东亦大通自动化设备有限公司 | Intelligent control method and system for welding and forming ion exchange membrane |
| CN114851569B (en) * | 2022-03-31 | 2024-01-05 | 中国电建集团吉林省电力勘测设计院有限公司 | Intelligent control method and system for ion exchange membrane welding forming |
| CN117073937A (en) * | 2023-10-18 | 2023-11-17 | 梁山华鲁专用汽车制造有限公司 | Tightness detection system based on powder material transport vehicle tank body |
| CN117073937B (en) * | 2023-10-18 | 2024-01-12 | 梁山华鲁专用汽车制造有限公司 | Tightness detection system based on powder material transport vehicle tank body |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114049351B (en) | 2022-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114049351B (en) | Door and window welding control method and system based on artificial intelligence | |
| EP3542211B1 (en) | Method and device and computer program for determining a representation of a spectacle glass rim | |
| CN103578430B (en) | A kind of method of brightness of automatic adjustable liquid crystal display screen and electronic equipment | |
| CN101256157B (en) | Method and apparatus for testing surface defect | |
| CN115082464A (en) | Method and system for identifying welding seam data in welding process of dust remover | |
| CN105678776B (en) | Weld image feature point extraction method based on laser vision sensor | |
| CN101082592B (en) | Uneven checking method and device | |
| DE4300103A1 (en) | ||
| CN117351019B (en) | Welding defect detection method | |
| CN113340909B (en) | Glue line defect detection method based on machine vision | |
| Fang et al. | Weld pool image segmentation of hump formation based on fuzzy C-means and Chan-Vese model | |
| CN111766245B (en) | Button cell negative electrode shell defect detection method based on machine vision | |
| CN116071363A (en) | Automatic change shaped steel intelligent production monitoring system | |
| CN112683166B (en) | Die-cutting product size detection method | |
| CN104408727A (en) | Image edge stain detection method and system | |
| CN116228798B (en) | Intelligent iron tower cutting detection method based on machine vision | |
| CN117456319A (en) | Industrial defect detection method combining rotation target detection and defect classification | |
| CN114663416B (en) | Copper bar cold-break defect detection method and system based on computer vision | |
| US20220076428A1 (en) | Product positioning method | |
| Zhang et al. | Extraction method of a nonuniform auxiliary laser stripe feature for three-dimensional reconstruction of large components | |
| Zhao et al. | Weld pool image acquisition and contour extraction based on arc spectrum and camera quantum efficiency | |
| CN112683902B (en) | Die-cutting product surface defect online detection method based on template | |
| CN114638847A (en) | Insulator hardware trimming method and system based on image processing | |
| CN115393290A (en) | Edge defect detection method, device and equipment | |
| US10257405B2 (en) | Automatic focusing method and automatic focusing system |
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 |