CN117161582B - Laser cutting method based on computer vision - Google Patents
Laser cutting method based on computer vision Download PDFInfo
- Publication number
- CN117161582B CN117161582B CN202311457724.4A CN202311457724A CN117161582B CN 117161582 B CN117161582 B CN 117161582B CN 202311457724 A CN202311457724 A CN 202311457724A CN 117161582 B CN117161582 B CN 117161582B
- Authority
- CN
- China
- Prior art keywords
- cutting
- area
- boundary pixel
- flow
- subsequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 84
- 238000003698 laser cutting Methods 0.000 title claims abstract description 78
- 238000005520 cutting process Methods 0.000 claims abstract description 167
- 239000002893 slag Substances 0.000 claims abstract description 50
- 238000012544 monitoring process Methods 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims abstract description 5
- 230000011218 segmentation Effects 0.000 claims description 45
- 238000012216 screening Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 230000000877 morphologic effect Effects 0.000 claims description 5
- 239000000463 material Substances 0.000 description 9
- 239000007789 gas Substances 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 229910044991 metal oxide Inorganic materials 0.000 description 1
- 150000004706 metal oxides Chemical class 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
Landscapes
- Laser Beam Processing (AREA)
Abstract
The invention relates to the technical field of laser cutting, in particular to a laser cutting method based on computer vision, which comprises the following steps: acquiring laser cutting flow information, dividing a laser cutting process into a plurality of flow subsequences according to the flow information, and acquiring image data of each flow subsequence; performing preliminary analysis on a cutting path of the current flow subsequence through the image data to obtain a fuzzy cutting area of a laser cutting track rectangular gray scale map of the current flow subsequence; acquiring each boundary pixel point of the fuzzy cutting area and Euclidean distance between each boundary pixel point and the central pixel point, and dividing the fuzzy cutting area into an actual cutting area and an edge slag hanging area according to the two-dimensional coordinates and Euclidean distance of all boundary pixel points of the fuzzy cutting area; and (3) monitoring and controlling the laser cutting process in real time according to the actual cutting area and the edge slag hanging area in the current flow subsequence, so that the accuracy of laser cutting and the cutting quality of the plate are improved.
Description
Technical Field
The invention relates to the technical field of laser cutting, in particular to a laser cutting method based on computer vision.
Background
The laser cutting machine focuses the laser emitted from the laser into high-power density laser beam through an optical path system, the metal material is melted under the action of the high-energy density laser beam, and meanwhile, most of molten metal is removed under the action of high-pressure gas flow (part of gas participates in chemical reaction to generate metal oxide) coaxial with the laser beam, so that the surface tension of the metal and the adhesion tension generated due to the viscosity are overcome.
The comparison document CN114067121A 'a method for cutting a rusty plate based on computer vision' is characterized in that the laser cutting image and the plate gray level image are processed, the obtained structural part areas are scored through the edge rusting rate of each structural part and the comprehensive rusting rate of the structural part, the whole surface of the plate can be accurately and rapidly analyzed, the laser cutting is carried out according to the structural part area with the highest score, the influence caused by the rusting of the plate surface is reduced to the minimum, and the optimal laser cutting area of the rusty plate is obtained.
The comparative document CN115722805a "control method for laser cutting and related equipment" can determine the cutting process according to the specific situation of the sheet material during the laser cutting process. Specifically, in the cutting process, a cutting process can be determined for each contour to be cut, local film spraying processing of the plate is automatically realized, and meanwhile, the laser cutting efficiency is improved.
Before laser cutting, an operator firstly completes whole plate trepanning by using trepanning software on a computer, programs the rest plate trepanning and high-efficiency cutting trepanning, and sends the trepanning programs to a numerical control laser cutting machine, the numerical control laser cutting machine cuts a part to be cut according to a CNC cutting program provided by the trepanning software, the existing modern trepanning software maximally utilizes raw materials under the consideration of the shape, size, quantity of the part, size, thickness, cutting characteristics and other factors of the material, improves production efficiency, reduces cost, reduces waste generation as a target, arranges the position and direction of the part to be cut to enable the part to be most closely placed on the material, thereby reducing material waste, then produces a cutting path, and the laser cutting machine performs actual cutting operation according to set cutting parameters according to path instructions;
however, in the process of performing the laser cutting path, equipment cutting parameter abnormality, material quality failure standard, edge slag hanging and other defect problems may occur on the plate to be cut due to the failure of the laser cutting equipment and the material problem, so that the plate cutting quality of the same area of the subsequent laser cutting equipment is affected, and thus the production efficiency and the yield are affected.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a laser cutting method based on computer vision, which comprises the following steps:
step S1: acquiring laser cutting flow information, dividing a laser cutting process into a plurality of flow subsequences according to the flow information, and acquiring image data of each flow subsequence;
step S2: performing preliminary analysis on a cutting path of the current flow subsequence through the image data to obtain a fuzzy cutting area and a background area of a laser cutting track rectangular gray scale map of the current flow subsequence;
step S3: acquiring each boundary pixel point of the fuzzy cutting area and Euclidean distance between each boundary pixel point and the central pixel point, and dividing the fuzzy cutting area into an actual cutting area and an edge slag hanging area according to the two-dimensional coordinates and Euclidean distance of all boundary pixel points of the fuzzy cutting area;
step S4: and monitoring and controlling the laser cutting process in real time according to the actual cutting area and the edge slag hanging area in the current flow subsequence.
Further, the process of obtaining the laser cutting process information, dividing the laser cutting process into a plurality of process subsequences according to the process information, and obtaining the image data of each process subsequence comprises:
acquiring process flow characteristics of current laser cutting equipment according to whole plate trepanning programming information generated by trepanning software, extracting flow information according to the process flow characteristics, splitting a laser cutting flow according to the flow information, and dividing the laser cutting flow into a plurality of flow subsequences;
acquiring position features in each flow sub-sequence, and acquiring an image shooting angle and an image sampling frequency of each flow sub-sequence according to each process flow feature and the position features in each flow sub-sequence;
and acquiring image data of each flow sub-sequence according to the image shooting angle and the image sampling frequency.
Further, the process of primarily analyzing the cutting path of the current flow sub-sequence through the image data to obtain the fuzzy cutting area and the background area of the rectangular gray scale map of the laser cutting track of the current flow sub-sequence comprises the following steps:
acquiring a rectangular gray scale map of a laser pre-cutting track of a current flow subsequence according to whole plate jacking programming information, setting a pixel point at the lower left corner of the rectangular gray scale map as a coordinate origin, constructing a coordinate system, drawing coordinates of each pixel in the rectangular gray scale map into the coordinate system, mapping gray scale values of each pixel point in the rectangular gray scale map into the coordinate system as parameters, and acquiring a standard cutting area in the rectangular gray scale map;
obtaining a cut track rectangular gray scale map of a current flow subsequence, mapping the cut track rectangular gray scale map into the coordinate system, comparing the cut track rectangular gray scale map with the rectangular gray scale map, obtaining a gray scale difference value in a pixel point at a position corresponding to the laser cut track gray scale map and a background gray scale map, and converting the gray scale difference value into a pixel value of a binary image to generate the binary image;
setting two pixel point position traversing pointers and a preset pixel threshold value, and simultaneously starting traversing from the first pixel point position and the last pixel point position of the binary image area; marking a pixel point location area with a pixel value larger than a preset pixel threshold value in the binary image as a fuzzy cutting area; and marking a pixel point location area with a pixel value smaller than or equal to a preset pixel threshold value in the binary image as a background area.
Further, the process of dividing the fuzzy cutting area into an actual cutting area and an edge slag-adhering area according to the two-dimensional coordinates and Euclidean distance of all boundary pixel points of the fuzzy cutting area comprises the following steps:
acquiring a fuzzy cutting area of a binary image, setting a pixel point at the central part of the fuzzy cutting area as a central pixel, and establishing a two-dimensional coordinate system by taking the central pixel as an origin to acquire all boundary pixel points of the fuzzy cutting area and Euclidean distances between all boundary pixel points and the central pixel point;
constructing a unidirectional linked list, taking one boundary pixel point of the fuzzy cutting area edge as a starting point, and sequentially inputting the two-dimensional coordinates of all boundary pixel points of the fuzzy cutting area edge and Euclidean distances between the two-dimensional coordinates and a central pixel point into the unidirectional linked list clockwise;
acquiring the total number of boundary pixel points, setting the length of an identification interval of an actual cutting area, and dividing a single-direction linked list into a plurality of identification segments according to the total number of the boundary pixel points and the length of the identification interval;
obtaining an average Euclidean distance between boundary pixel points in an identification segmentation section and a central pixel point, obtaining boundary pixel points with the Euclidean distance in the identification segmentation section larger than the average Euclidean distance, dividing the boundary pixel points into a set to obtain a first set, setting a first detection multiple, obtaining a boundary pixel point comparison number according to the first detection multiple and the identification section length, screening the first set according to the boundary pixel point comparison number to obtain a second set, and then carrying out morphological standard screening on the boundary pixel points in the second set to obtain a third set;
obtaining boundary pixel points with the minimum Euclidean distance and boundary pixel points with the maximum Euclidean distance in a third set of each identification segmentation segment, obtaining auxiliary connecting lines of the boundary pixel points with the minimum Euclidean distance and the central part of the fuzzy cutting area in each identification segmentation segment, obtaining two segmentation line segments which are perpendicular to the auxiliary connecting lines and are respectively positioned at the boundary pixel points and the central part of the fuzzy cutting area, and taking the two segmentation line segments as boundaries respectively to obtain the actual cutting area of each identification segmentation segment;
similarly, an auxiliary connecting line of the boundary pixel point with the largest Euclidean distance and the central part of the fuzzy cutting area in the third set of each identification segmentation segment is obtained, a segmentation line segment which is perpendicular to the auxiliary connecting line and is positioned at the boundary pixel point with the largest Euclidean distance is obtained, meanwhile, a segmentation line segment which is positioned at the boundary pixel point with the smallest Euclidean distance is obtained, and the two segmentation line segments are respectively used as boundaries to obtain an edge slag hanging area of each identification segmentation segment;
and connecting an actual cutting area and an edge slag hanging area of each identification segmentation in the fuzzy cutting area to obtain the actual cutting area and the edge slag hanging area in the fuzzy cutting area.
Further, the process of screening the first set according to the comparison number of the boundary pixel points to obtain the second set includes:
obtaining the comparison number k and the average Euclidean distance of boundary pixel points in the current identification segmentation segment, obtaining the Euclidean distance of k/2 adjacent boundary pixel points adjacent to the left and right of the target boundary pixel point in the first set, and comparing the Euclidean distance of the adjacent boundary pixel points with the average Euclidean distance;
and if the Euclidean distances of the adjacent boundary pixels of the target boundary pixel point are all smaller than or equal to the average Euclidean distance, dividing the target boundary pixel point into a second set.
Further, the process of obtaining the third set includes:
constructing an approximate triangle by taking Euclidean distance of target boundary pixel points in a second set as the height of the triangle and taking the comparison number k of the boundary pixel points in the identification segmentation section of the target boundary pixel points as the base of the triangle, setting a vertex angle half-angle tangent threshold value of the approximate triangle, acquiring the vertex angle half-angle tangent value of the approximate triangle, and comparing the vertex angle half-angle tangent value with the vertex angle half-angle tangent threshold value;
and if the vertex angle half angle tangent value is larger than the vertex angle half angle tangent threshold value, dividing the target boundary pixel points into a third set.
Further, the process of monitoring and controlling the laser cutting process in real time according to the actual cutting area in the current flow subsequence comprises the following steps:
acquiring an actual region coordinate set of an actual cutting region of a current flow subsequence in the coordinate system, acquiring a standard region coordinate set of a standard cutting region of the rectangular gray scale image in the coordinate system, judging the number of coincident coordinates of the actual region coordinate set and the standard region coordinate set, presetting a coincident coordinate threshold, comparing the number of coincident coordinates with the coincident coordinate threshold, and marking the laser cutting equipment as a normal state if the number of coincident coordinates is greater than or equal to the coincident coordinate threshold and the cut plate of the current flow subsequence is marked as qualified;
and if the number of the coincident coordinates is smaller than a coincident coordinate threshold value, marking the laser cutting equipment as a fault state, and marking the plate cut by the current flow subsequence as unqualified.
Further, the process of monitoring and controlling the laser cutting process in real time according to the actual cutting area and the edge slag hanging area in the current flow subsequence comprises the following steps:
if the cut plate of the current flow subsequence is marked as qualified, acquiring a slag hanging region coordinate set of an edge slag hanging region of the current flow subsequence in the coordinate system, acquiring a standard region coordinate set of a standard cutting region of a next flow subsequence in the coordinate system, judging whether an intersection exists between the slag hanging region coordinate set and the standard region coordinate set, and if the intersection exists, feeding back the slag hanging region coordinate set of the edge slag hanging region of the current flow subsequence in the coordinate system to a nesting software, wherein the nesting software updates generated whole plate nesting programming information according to the slag hanging region coordinate set.
Compared with the prior art, the invention has the beneficial effects that:
1. the fuzzy cutting area is divided into an actual cutting area and an edge slag hanging area by using a computer vision method, an actual cutting area and an expected cutting path of a flow subsequence after cutting are obtained, cutting error information is obtained by comparing the actual cutting area with the expected cutting path, and quality evaluation is carried out on the actual cutting area, so that the accuracy of laser cutting and the cutting quality of the plate are improved.
2. The edge slag-hanging area of the current flow subsequence is monitored in real time through a computer vision system, the mutual influence and interaction of all cutting steps in all the flow subsequences are considered, the edge slag-hanging area of the current flow subsequence is compared with the standard cutting area of the next flow subsequence, the mutual influence factors in all the flow subsequences are utilized, closed-loop control is realized through a feedback control technology, and the laser cutting path and parameters are adjusted in real time, so that the cutting of the plate is more accurate and the cutting quality is higher.
Drawings
Fig. 1 is a schematic diagram of a laser cutting method based on computer vision according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application, taken in conjunction with the accompanying drawings, clearly and completely describes the technical solutions of the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
As shown in fig. 1, a laser cutting method based on computer vision includes the following steps:
step S1: acquiring laser cutting flow information, dividing a laser cutting process into a plurality of flow subsequences according to the flow information, and acquiring image data of each flow subsequence;
step S2: performing preliminary analysis on a cutting path of the current flow subsequence through the image data to obtain a fuzzy cutting area and a background area of a laser cutting track rectangular gray scale map of the current flow subsequence;
step S3: acquiring each boundary pixel point of the fuzzy cutting area and Euclidean distance between each boundary pixel point and the central pixel point, and dividing the fuzzy cutting area into an actual cutting area and an edge slag hanging area according to the two-dimensional coordinates and Euclidean distance of all boundary pixel points of the fuzzy cutting area;
step S4: and monitoring and controlling the laser cutting process in real time according to the actual cutting area and the edge slag hanging area in the current flow subsequence.
It should be further noted that, in the specific implementation process, the process of obtaining the laser cutting flow information, dividing the laser cutting process into a plurality of flow sub-sequences according to the flow information, and obtaining the image data of each flow sub-sequence includes:
acquiring process flow characteristics of current laser cutting equipment according to whole plate trepanning programming information generated by trepanning software, extracting flow information according to the process flow characteristics, splitting a laser cutting flow according to the flow information, and dividing the laser cutting flow into a plurality of flow subsequences;
acquiring position features in each flow sub-sequence, and acquiring an image shooting angle and an image sampling frequency of each flow sub-sequence according to each process flow feature and the position features in each flow sub-sequence;
and acquiring image data of each flow sub-sequence according to the image shooting angle and the image sampling frequency.
The laser cutting device comprises a bearing platform, a cutting platform arranged above the bearing platform, a plurality of laser cutting heads above the cutting platform and a high-definition camera arranged near the laser cutting heads, wherein the high-definition camera is used for acquiring images and acquiring real-time images of subsequences of each flow.
It should be further noted that, in the implementation process, the process of the whole board nesting programming information generated by the nesting software includes:
and importing the geometric data of a plurality of parts to be cut into a numerical control system. These geometric data may be CAD software generated files, common formats including DXF, DWG, etc.;
according to actual plate materials and specification requirements, parameters such as the size, the material quality and the like of the plate are defined in a numerical control system; the setting may be performed by manually inputting or selecting predefined parameters;
the numerical control system utilizes a jacking algorithm to perform layout optimization on the parts according to the imported geometric data and the board specifications; the stock-jacking algorithm generally considers the principle of maximum utilization rate and minimum waste amount, selects the optimal placement position and angle, and ensures the minimum gap between parts;
the numerical control system uses a path planning algorithm according to the optimized part layout, and takes factors such as part shape, cutting requirements and the like into consideration to generate the shortest cutting path so as to improve cutting efficiency;
according to the optimized cutting path, the numerical control system generates a corresponding Numerical Control (NC) code. The NC codes describe the motion track, speed and cutting parameters of the cutting machine tool, and can be stored as files or directly sent to a numerical control cutting machine for execution so as to realize automatic cutting.
It should be further noted that, in the specific implementation process, the process flow characteristics of the current laser cutting device represent determining, according to the whole plate set material programming information, cutting rates, laser powers, cutting gases, cutting paths, and time sequence and sequencing of cutting on a plurality of parts to be cut.
It should be further noted that, in the implementation process, the process of performing preliminary analysis on the cutting path of the current flow sub-sequence through the image data to obtain the fuzzy cutting area and the background area of the rectangular gray scale map of the laser cutting track of the current flow sub-sequence includes:
acquiring a rectangular gray scale map of a laser pre-cutting track of a current flow subsequence according to whole plate jacking programming information, setting a pixel point at the lower left corner of the rectangular gray scale map as a coordinate origin, constructing a coordinate system, drawing coordinates of each pixel in the rectangular gray scale map into the coordinate system, mapping gray scale values of each pixel point in the rectangular gray scale map into the coordinate system as parameters, and acquiring a standard cutting area in the rectangular gray scale map;
obtaining a cut track rectangular gray scale map of a current flow subsequence, mapping the cut track rectangular gray scale map into the coordinate system, comparing the cut track rectangular gray scale map with the rectangular gray scale map, obtaining a gray scale difference value in a pixel point at a position corresponding to the laser cut track gray scale map and a background gray scale map, and converting the gray scale difference value into a pixel value of a binary image to generate the binary image;
setting two pixel point position traversing pointers and a preset pixel threshold value, and simultaneously starting traversing from the first pixel point position and the last pixel point position of the binary image area; marking a pixel point location area with a pixel value larger than a preset pixel threshold value in the binary image as a fuzzy cutting area; and marking a pixel point location area with a pixel value smaller than or equal to a preset pixel threshold value in the binary image as a background area.
It should be further noted that, in the implementation process, the process of dividing the fuzzy cutting area into the actual cutting area and the edge slag-adhering area according to the two-dimensional coordinates and euclidean distance of all boundary pixel points of the fuzzy cutting area includes:
acquiring a fuzzy cutting area of a binary image, setting a pixel point at the central part of the fuzzy cutting area as a central pixel, and establishing a two-dimensional coordinate system by taking the central pixel as an origin to acquire all boundary pixel points of the fuzzy cutting area and Euclidean distances between all boundary pixel points and the central pixel point;
constructing a unidirectional linked list, taking one boundary pixel point of the fuzzy cutting area edge as a starting point, and sequentially inputting the two-dimensional coordinates of all boundary pixel points of the fuzzy cutting area edge and Euclidean distances between the two-dimensional coordinates and a central pixel point into the unidirectional linked list clockwise;
acquiring the total number of boundary pixel points, setting the length of an identification interval of an actual cutting area, and dividing a single-direction linked list into a plurality of identification segments according to the total number of the boundary pixel points and the length of the identification interval;
obtaining an average Euclidean distance between boundary pixel points in an identification segmentation section and a central pixel point, obtaining boundary pixel points with the Euclidean distance in the identification segmentation section larger than the average Euclidean distance, dividing the boundary pixel points into a set to obtain a first set, setting a first detection multiple, obtaining a boundary pixel point comparison number according to the first detection multiple and the identification section length, screening the first set according to the boundary pixel point comparison number to obtain a second set, and then carrying out morphological standard screening on the boundary pixel points in the second set to obtain a third set;
obtaining boundary pixel points with the minimum Euclidean distance and boundary pixel points with the maximum Euclidean distance in a third set of each identification segmentation segment, obtaining auxiliary connecting lines of the boundary pixel points with the minimum Euclidean distance and the central part of the fuzzy cutting area in each identification segmentation segment, obtaining two segmentation line segments which are perpendicular to the auxiliary connecting lines and are respectively positioned at the boundary pixel points and the central part of the fuzzy cutting area, and taking the two segmentation line segments as boundaries respectively to obtain the actual cutting area of each identification segmentation segment;
similarly, an auxiliary connecting line of the boundary pixel point with the largest Euclidean distance and the central part of the fuzzy cutting area in the third set of each identification segmentation segment is obtained, a segmentation line segment which is perpendicular to the auxiliary connecting line and is positioned at the boundary pixel point with the largest Euclidean distance is obtained, meanwhile, a segmentation line segment which is positioned at the boundary pixel point with the smallest Euclidean distance is obtained, and the two segmentation line segments are respectively used as boundaries to obtain an edge slag hanging area of each identification segmentation segment;
and connecting an actual cutting area and an edge slag hanging area of each identification segmentation in the fuzzy cutting area to obtain the actual cutting area and the edge slag hanging area in the fuzzy cutting area.
It should be further noted that, in the implementation process, the process of screening the first set according to the comparison number of the boundary pixel points to obtain the second set includes:
obtaining the comparison number k and the average Euclidean distance of boundary pixel points in the current identification segmentation segment, obtaining the Euclidean distance of k/2 adjacent boundary pixel points adjacent to the left and right of the target boundary pixel point in the first set, and comparing the Euclidean distance of the adjacent boundary pixel points with the average Euclidean distance;
if the Euclidean distance in the adjacent boundary pixel points of the target boundary pixel point is larger than the average Euclidean distance, the target boundary pixel point is not considered to be in accordance with the standard of being divided into the second set;
and if the Euclidean distances of the adjacent boundary pixels of the target boundary pixel point are all smaller than or equal to the average Euclidean distance, dividing the target boundary pixel point into a second set.
It should be further noted that, in the implementation process, the process of performing morphological criteria screening on the boundary pixel points in the second set to obtain the third set includes:
constructing an approximate triangle by taking Euclidean distance of target boundary pixel points in a second set as the height of the triangle and taking the comparison number k of the boundary pixel points in the identification segmentation section of the target boundary pixel points as the base of the triangle, setting a vertex angle half-angle tangent threshold value of the approximate triangle, acquiring the vertex angle half-angle tangent value of the approximate triangle, and comparing the vertex angle half-angle tangent value with the vertex angle half-angle tangent threshold value;
and if the vertex angle half angle tangent value is larger than the vertex angle half angle tangent threshold value, dividing the target boundary pixel points into a third set.
It should be further noted that, in the specific implementation process, the process of monitoring and controlling the laser cutting process in real time according to the actual cutting area in the current flow subsequence includes:
acquiring an actual region coordinate set of an actual cutting region of a current flow subsequence in the coordinate system, acquiring a standard region coordinate set of a standard cutting region of the rectangular gray scale image in the coordinate system, judging the number of coincident coordinates of the actual region coordinate set and the standard region coordinate set, presetting a coincident coordinate threshold, comparing the number of coincident coordinates with the coincident coordinate threshold, and marking the laser cutting equipment as a normal state if the number of coincident coordinates is greater than or equal to the coincident coordinate threshold and the cut plate of the current flow subsequence is marked as qualified;
and if the number of the coincident coordinates is smaller than a coincident coordinate threshold value, marking the laser cutting equipment as a fault state, and marking the plate cut by the current flow subsequence as unqualified.
After the cutting is completed, the quality of the cutting result is evaluated by using a computer vision method, and the accuracy of the cutting is evaluated by comparing the actual cutting area and the expected shape of the cut flow subsequence.
It should be further noted that, in the specific implementation process, the process of determining the number of coincident coordinates of the actual region coordinate set and the standard region coordinate set is illustrated:
four dividing lines of the current flow subsequence are obtained to form a closed actual cutting area AE, and edge coordinates on cutting lines of the actual cutting area AE of the current flow subsequence are as follows: { [ A, B ], [ B, C ], [ C, D ], [ D, E ], [ E, A ] }, i.e. two points of [ A, B ] and [ B, C ] form a straight line, and so on, form four straight lines, a closed quadrilateral area, i.e. an actual cutting area AE, is formed by the four straight lines, and meanwhile, a standard area and a standard area coordinate set of a current flow subsequence are obtained, a standard area UQ of the current flow subsequence is an irregular polygon, and the coordinates of the standard area UQ are drawn through the standard area coordinate set: { [ U1, Q1], [ U2, Q2], [ U3, Q3], … …, [ Un-1, qn-1], [ Un, qn ], [ U1, Q1] }. The number of coincident coordinates of the actual cutting area AE and the standard area UQ is calculated by analyzing the actual cutting area AE and the standard area UQ in the coordinate system, and the mathematical calculation principle is as follows: judging whether the X coordinate value of any point [ Ax, by ] of the actual cutting area AE is between the X coordinate values of any two points of the standard area UQ, namely Ux+m is less than or equal to Ax and less than or equal to Ux+n; judging whether the Y coordinate value of any point [ Ax, by ] in the actual cutting area AE is also between the Y coordinate values of the standard area UQ, namely Qy+r is less than or equal to By and less than or equal to Qy+g;
if the coordinates [ A, B ] meet Ux+m less than or equal to Ax less than or equal to Ux+n and Qy+r less than or equal to By less than or equal to Qy+g, marking the coordinates [ A, B ] as a combined coordinate;
if the coordinates [ A, B ] do not meet Ux+m.ltoreq.Ax.ltoreq.Ux+n and Qy+r.ltoreq.By.ltoreq.Qy+g, the coordinates [ A, B ] are marked as non-coincident coordinates.
It should be further noted that, in the specific implementation process, the process of monitoring and controlling the laser cutting process in real time according to the actual cutting area and the edge slag-hanging area in the current flow subsequence includes:
if the cut plate of the current flow subsequence is marked as qualified, acquiring a slag hanging region coordinate set of an edge slag hanging region of the current flow subsequence in the coordinate system, acquiring a standard region coordinate set of a standard cutting region of a next flow subsequence in the coordinate system, judging whether an intersection exists between the slag hanging region coordinate set and the standard region coordinate set, and if the intersection exists, feeding back the slag hanging region coordinate set of the edge slag hanging region of the current flow subsequence in the coordinate system to a nesting software, wherein the nesting software updates generated whole plate nesting programming information according to the slag hanging region coordinate set.
It should be further noted that, in the implementation process, the process of breaking whether the intersection exists between the slag hanging region coordinate set and the standard region coordinate set includes:
the method comprises the steps of obtaining a closed slag hanging area FJ formed by four dividing lines of a current flow subsequence, wherein edge coordinates on a cutting line of the slag hanging area FJ of the current flow subsequence are as follows: { [ F, G ], [ G, H ], [ H, I ], [ I, J ], [ J, F ] }, i.e. two points of [ F, G ] and [ G, H ] form a straight line, and so on, form four straight lines, a closed quadrilateral area, i.e. a slag hanging area FJ, is formed by the four straight lines, and meanwhile, a standard area and a standard area coordinate set of a next flow subsequence are obtained, the standard area OP of the current flow subsequence is an irregular polygon, and the coordinates of the standard area OP are drawn through the standard area coordinate set: { [ O1, P1], [ O2, P2], [ O3, P3], … …, [ On-1, pn-1], [ On, pn ], [ O1, P1] }. Through analyzing the slag adhering region FJ and the standard region OP in the coordinate system, the number of coincident coordinates of the slag adhering region FJ and the standard region OP is calculated, and the mathematical calculation principle is as follows: judging whether X coordinate values of any point [ Fx, gy ] of the slag hanging area FJ are between X coordinate values of any two points of the standard area OP, namely ox+m is less than or equal to Fx and less than or equal to Px+n; judging whether Y coordinate values of any point [ Fx, gy ] in the slag hanging area FJ are also between Y coordinate values of the standard area OP, namely Oy+r is less than or equal to Gy and less than or equal to Py+g; if the coordinates [ F, G ] meet that ox+m is less than or equal to Fx is less than or equal to ox+n and Py+r is less than or equal to Gy is less than or equal to Py+g, judging that the intersection exists between the coordinate set of the slag hanging region and the coordinate set of the standard region, feeding back the coordinate set of the slag hanging region of the edge slag hanging region of the current flow sub-sequence in the coordinate system to the jacking software, and updating the generated whole plate jacking programming information by the jacking software according to the coordinate set of the slag hanging region.
The invention monitors the image change in the cutting process in real time through the computer vision system, compares the image change with the expected cutting path, acquires the cutting error information, and simultaneously considers that each cutting step in each flow subsequence can be mutually influenced and interacted. And closed-loop control is realized by utilizing error information and mutual influence factors in each flow subsequence through a feedback control technology, and laser cutting paths and parameters are adjusted, so that cutting results are more accurate.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (1)
1. The laser cutting method based on computer vision is characterized by comprising the following steps of:
step S1: acquiring laser cutting flow information, dividing a laser cutting process into a plurality of flow subsequences according to the flow information, and acquiring image data of each flow subsequence;
the process for acquiring the laser cutting flow information, dividing the laser cutting process into a plurality of flow subsequences according to the flow information, and acquiring the image data of each flow subsequence comprises the following steps:
acquiring process flow characteristics of current laser cutting equipment according to whole plate trepanning programming information generated by trepanning software, extracting flow information according to the process flow characteristics, splitting a laser cutting flow according to the flow information, and dividing the laser cutting flow into a plurality of flow subsequences;
acquiring position features in each flow sub-sequence, and acquiring an image shooting angle and an image sampling frequency of each flow sub-sequence according to each process flow feature and the position features in each flow sub-sequence;
acquiring image data of each flow subsequence according to the image shooting angle and the image sampling frequency;
step S2: performing preliminary analysis on a cutting path of the current flow subsequence through the image data to obtain a fuzzy cutting area and a background area of a laser cutting track rectangular gray scale map of the current flow subsequence;
the process of primarily analyzing the cutting path of the current flow subsequence through the image data to obtain the fuzzy cutting area and the background area of the rectangular gray scale map of the laser cutting track of the current flow subsequence comprises the following steps:
acquiring a rectangular gray scale map of a laser pre-cutting track of a current flow subsequence according to whole plate jacking programming information, setting a pixel point at the lower left corner of the rectangular gray scale map as a coordinate origin, constructing a coordinate system, drawing coordinates of each pixel in the rectangular gray scale map into the coordinate system, mapping gray scale values of each pixel point in the rectangular gray scale map into the coordinate system as parameters, and acquiring a standard cutting area in the rectangular gray scale map;
obtaining a cut track rectangular gray scale map of a current flow subsequence, mapping the cut track rectangular gray scale map into the coordinate system, comparing the cut track rectangular gray scale map with the rectangular gray scale map, obtaining a gray scale difference value in a pixel point at a position corresponding to the laser cut track gray scale map and a background gray scale map, and converting the gray scale difference value into a pixel value of a binary image to generate the binary image;
setting two pixel point position traversing pointers and a preset pixel threshold value, and simultaneously starting traversing from the first pixel point position and the last pixel point position of the binary image area; marking a pixel point location area with a pixel value larger than a preset pixel threshold value in the binary image as a fuzzy cutting area; marking a pixel point location area with a pixel value smaller than or equal to a preset pixel threshold value in the binary image as a background area;
step S3: acquiring each boundary pixel point of the fuzzy cutting area and Euclidean distance between each boundary pixel point and the central pixel point, and dividing the fuzzy cutting area into an actual cutting area and an edge slag hanging area according to the two-dimensional coordinates and Euclidean distance of all boundary pixel points of the fuzzy cutting area;
the process of dividing the fuzzy cutting area into an actual cutting area and an edge slag-adhering area according to the two-dimensional coordinates and Euclidean distance of all boundary pixel points of the fuzzy cutting area comprises the following steps:
acquiring a fuzzy cutting area of a binary image, setting a pixel point at the central part of the fuzzy cutting area as a central pixel, and establishing a two-dimensional coordinate system by taking the central pixel as an origin to acquire all boundary pixel points of the fuzzy cutting area and Euclidean distances between all boundary pixel points and the central pixel point;
constructing a unidirectional linked list, taking one boundary pixel point of the fuzzy cutting area edge as a starting point, and sequentially inputting the two-dimensional coordinates of all boundary pixel points of the fuzzy cutting area edge and Euclidean distances of the central pixel points into the unidirectional linked list clockwise;
acquiring the total number of boundary pixel points, setting the length of an identification interval of an actual cutting area, and dividing a single-direction linked list into a plurality of identification segments according to the total number of the boundary pixel points and the length of the identification interval;
obtaining an average Euclidean distance between boundary pixel points in an identification segmentation section and a central pixel point, obtaining boundary pixel points with the Euclidean distance in the identification segmentation section larger than the average Euclidean distance, dividing the boundary pixel points into a set to obtain a first set, setting a first detection multiple, obtaining a boundary pixel point comparison number according to the first detection multiple and the identification section length, screening the first set according to the boundary pixel point comparison number to obtain a second set, and then carrying out morphological standard screening on the boundary pixel points in the second set to obtain a third set;
obtaining boundary pixel points with the minimum Euclidean distance and boundary pixel points with the maximum Euclidean distance in a third set of each identification segmentation segment, obtaining auxiliary connecting lines of the boundary pixel points with the minimum Euclidean distance and the central part of the fuzzy cutting area in each identification segmentation segment, obtaining two segmentation line segments which are perpendicular to the auxiliary connecting lines and are respectively positioned at the boundary pixel points and the central part of the fuzzy cutting area, and taking the two segmentation line segments as boundaries respectively to obtain the actual cutting area of each identification segmentation segment;
similarly, an auxiliary connecting line of the boundary pixel point with the largest Euclidean distance and the central part of the fuzzy cutting area in the third set of each identification segmentation segment is obtained, segmentation line segments which are perpendicular to the auxiliary connecting line and are positioned at the boundary pixel point with the largest Euclidean distance are obtained, segmentation line segments which are positioned at the boundary pixel point with the smallest Euclidean distance are obtained at the same time, and the two segmentation line segments are respectively used as boundaries to obtain an edge slag hanging area of each identification segmentation segment;
connecting an actual cutting area and an edge slag hanging area of each identification segmentation in the fuzzy cutting area to obtain the actual cutting area and the edge slag hanging area in the fuzzy cutting area;
screening the first set according to the comparison number of the boundary pixel points, wherein the process of obtaining the second set comprises the following steps:
obtaining the comparison number k and the average Euclidean distance of boundary pixel points in the current identification segmentation segment, obtaining the Euclidean distance of k/2 adjacent boundary pixel points adjacent to the left and right of the target boundary pixel point in the first set, and comparing the Euclidean distance of the adjacent boundary pixel points with the average Euclidean distance;
if the Euclidean distances of the adjacent boundary pixels of the target boundary pixel point are all smaller than or equal to the average Euclidean distance, dividing the target boundary pixel point into a second set;
and then carrying out morphological standard screening on the boundary pixel points in the second set, wherein the process for obtaining the third set comprises the following steps:
constructing an approximate triangle by taking Euclidean distance of target boundary pixel points in a second set as the height of the triangle and taking the comparison number k of the boundary pixel points in the identification segmentation section of the target boundary pixel points as the base of the triangle, setting a vertex angle half-angle tangent threshold value of the approximate triangle, acquiring the vertex angle half-angle tangent value of the approximate triangle, and comparing the vertex angle half-angle tangent value with the vertex angle half-angle tangent threshold value;
if the vertex angle half-angle tangent value is larger than the vertex angle half-angle tangent threshold value, dividing the target boundary pixel points into a third set;
step S4: the laser cutting process is monitored and controlled in real time according to the actual cutting area and the edge slag hanging area in the current flow subsequence;
the process for monitoring and controlling the laser cutting process in real time according to the actual cutting area in the current flow subsequence comprises the following steps:
acquiring an actual region coordinate set of an actual cutting region of a current flow subsequence in the coordinate system, acquiring a standard region coordinate set of a standard cutting region of the rectangular gray scale image in the coordinate system, judging the number of coincident coordinates of the actual region coordinate set and the standard region coordinate set, presetting a coincident coordinate threshold, comparing the number of coincident coordinates with the coincident coordinate threshold, and marking the laser cutting equipment as a normal state if the number of coincident coordinates is greater than or equal to the coincident coordinate threshold and the cut plate of the current flow subsequence is marked as qualified;
if the number of the coincident coordinates is smaller than a coincident coordinate threshold value, marking the laser cutting equipment as a fault state, and marking the plate cut by the current flow subsequence as unqualified;
the process for monitoring and controlling the laser cutting process in real time according to the actual cutting area and the edge slag-hanging area in the current flow subsequence comprises the following steps:
if the cut plate of the current flow subsequence is marked as qualified, acquiring a slag hanging region coordinate set of an edge slag hanging region of the current flow subsequence in the coordinate system, acquiring a standard region coordinate set of a standard cutting region of a next flow subsequence in the coordinate system, judging whether an intersection exists between the slag hanging region coordinate set and the standard region coordinate set, and if the intersection exists, feeding back the slag hanging region coordinate set of the edge slag hanging region of the current flow subsequence in the coordinate system to a nesting software, wherein the nesting software updates generated whole plate nesting programming information according to the slag hanging region coordinate set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311457724.4A CN117161582B (en) | 2023-11-03 | 2023-11-03 | Laser cutting method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311457724.4A CN117161582B (en) | 2023-11-03 | 2023-11-03 | Laser cutting method based on computer vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117161582A CN117161582A (en) | 2023-12-05 |
| CN117161582B true CN117161582B (en) | 2023-12-26 |
Family
ID=88941768
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311457724.4A Active CN117161582B (en) | 2023-11-03 | 2023-11-03 | Laser cutting method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117161582B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117745743B (en) * | 2023-12-26 | 2024-07-09 | 苏州市凯思泰克自动化设备有限公司 | Automatic generation method and device for cutting path of printed circuit board |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101885152A (en) * | 2010-05-06 | 2010-11-17 | 南京航空航天大学 | Automatically-aligned numerical control cutting method for pre-printed image plate |
| CN105014240A (en) * | 2014-04-29 | 2015-11-04 | 中国科学院福建物质结构研究所 | LED wafer laser cutting device and LED wafer laser cutting levelness adjustment method |
| CN105227322A (en) * | 2015-10-30 | 2016-01-06 | 刘荣花 | A kind of laser cutting device with security protection system |
| CN108875740A (en) * | 2018-06-15 | 2018-11-23 | 浙江大学 | A kind of machine vision cutting method applied to laser cutting machine |
| CN114659627A (en) * | 2022-05-16 | 2022-06-24 | 广东工业大学 | Laser spot monitoring method, system and equipment based on annular filtering |
| CN115330989A (en) * | 2022-10-13 | 2022-11-11 | 南通红运金属科技有限公司 | Control method for laser beam cutting |
-
2023
- 2023-11-03 CN CN202311457724.4A patent/CN117161582B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101885152A (en) * | 2010-05-06 | 2010-11-17 | 南京航空航天大学 | Automatically-aligned numerical control cutting method for pre-printed image plate |
| CN105014240A (en) * | 2014-04-29 | 2015-11-04 | 中国科学院福建物质结构研究所 | LED wafer laser cutting device and LED wafer laser cutting levelness adjustment method |
| CN105227322A (en) * | 2015-10-30 | 2016-01-06 | 刘荣花 | A kind of laser cutting device with security protection system |
| CN108875740A (en) * | 2018-06-15 | 2018-11-23 | 浙江大学 | A kind of machine vision cutting method applied to laser cutting machine |
| CN114659627A (en) * | 2022-05-16 | 2022-06-24 | 广东工业大学 | Laser spot monitoring method, system and equipment based on annular filtering |
| CN115330989A (en) * | 2022-10-13 | 2022-11-11 | 南通红运金属科技有限公司 | Control method for laser beam cutting |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117161582A (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110227876B (en) | Robot welding path autonomous planning method based on 3D point cloud data | |
| CN112712303B (en) | Digital planning method for machining process and software platform | |
| CN117161582B (en) | Laser cutting method based on computer vision | |
| CN113751824B (en) | Automatic generation method and device for groove cutting track, computer equipment and medium | |
| CN115993804B (en) | Cutter parameter adjustment method based on numerical control machine tool and related equipment | |
| CN117047286B (en) | Method for processing workpiece surface by laser, processing system, processor and storage medium | |
| TWI610749B (en) | Laser Cutting Power Modified System and Adjustment Method thereof | |
| CN115685876B (en) | Planar laser cutting control method and system based on track compensation | |
| CN116851930B (en) | Three-dimensional five-axis curved surface laser cutting track generation method and device | |
| US20220134464A1 (en) | Weld line data generation device, welding system, weld line data generation method, and computer readable medium | |
| CN116572535A (en) | Production optimization method and system for high polymer material product | |
| CN110153582B (en) | Welding scheme generation method and device and welding system | |
| CN116909211B (en) | Intelligent regulation and control method and system for high-precision numerical control machine tool | |
| KR101405039B1 (en) | Manufacturing process automation method and system | |
| CN116700149A (en) | Self-adaptive filtering optimization method for numerical control machining tool path | |
| CN116228766A (en) | Intelligent regulation and control method and system for plasma processing equipment | |
| CN115106539B (en) | Material increasing and decreasing integrated control method and system | |
| CN113927127B (en) | Automatic groove line cutting method and device, storage medium and computer equipment | |
| CN108705689A (en) | Heavy caliber precision optics plane of crystal microdefect reparation automatic tool preset process | |
| KR20200014019A (en) | Image analysis system and method for post processing of a workpiece | |
| CN113158503A (en) | Method for predicting cutting deformation of complex curved surface part | |
| CN114683283B (en) | Teaching-free welding method and device for welding robot | |
| JP7395954B2 (en) | NC data quality judgment device and processing device | |
| CN117270537B (en) | Automatic shaving moving path control system and control method | |
| JPS62221003A (en) | Method for producing locus of 2-dimensional tool |
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 |