CN118015015A - Laser cleaning intelligent coverage path planning method and system - Google Patents
Laser cleaning intelligent coverage path planning method and system Download PDFInfo
- Publication number
- CN118015015A CN118015015A CN202410155735.5A CN202410155735A CN118015015A CN 118015015 A CN118015015 A CN 118015015A CN 202410155735 A CN202410155735 A CN 202410155735A CN 118015015 A CN118015015 A CN 118015015A
- Authority
- CN
- China
- Prior art keywords
- image data
- path
- planning
- point
- boundary image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004140 cleaning Methods 0.000 title claims abstract description 72
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000000605 extraction Methods 0.000 claims abstract description 21
- 239000011159 matrix material Substances 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 9
- 239000002699 waste material Substances 0.000 abstract description 5
- 230000008569 process Effects 0.000 description 20
- 230000008901 benefit Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 4
- 230000002829 reductive effect Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000005520 cutting process Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 229910000838 Al alloy Inorganic materials 0.000 description 1
- 229910000975 Carbon steel Inorganic materials 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 229910001069 Ti alloy Inorganic materials 0.000 description 1
- 238000002679 ablation Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000010962 carbon steel Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 239000003016 pheromone Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000010926 purge Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Image Processing (AREA)
Abstract
The invention provides a laser cleaning intelligent coverage path planning method and a system, which relate to the technical field of laser cleaning, and the method comprises the following steps: acquiring original binarized image data corresponding to a target area; and carrying out boundary extraction on the original binary image data based on an improved raster heuristic algorithm to obtain effective boundary image data, carrying out path planning on the current effective boundary image data according to a preset neighborhood scanning rule, obtaining next effective boundary image data adjacent to the current effective boundary image data, and carrying out path planning on the next effective boundary image data until the planning of the original binary image data is completed, so as to obtain a complete coverage path planning. The invention utilizes the improved grating heuristic algorithm to carry out boundary extraction and path planning, can reduce the waste of energy and time, improves the cleaning quality, and can effectively identify and cover the path of the randomly generated target area to be cleaned.
Description
Technical Field
The invention relates to the technical field of laser cleaning, in particular to a laser cleaning intelligent coverage path planning method and system.
Background
The laser cleaning is a green nondestructive surface cleaning method for removing attachments on the surface of a substrate by utilizing the interaction of laser and substances, and the efficient and high-quality cleaning is a main development direction of the laser cleaning and is also a key for realizing large-scale industrial application of the laser cleaning. The cleaning efficiency is one of the main factors which restrict the wide application of the laser cleaning at present, and is related to the aspects of cleaning a laser light source, a cleaning process, a cleaning method and the like.
However, at present, from the perspective of laser beam scanning control, there is the long time-consuming condition of planning, can lead to operation and maintenance cycle to increase, and in the processing procedure of laser cleaning technique, generally adopt manual calibration and full coverage scanning scheme for the precision that the route covered is too low, and whole coverage process is too complicated, leads to inefficiency, and then leads to the washing cost to increase.
Disclosure of Invention
The invention aims to solve the problem of improving the efficiency of laser cleaning in the process of planning a laser cleaning coverage path.
In order to solve the above problems, the present invention provides a method for planning an intelligent coverage path for laser cleaning, comprising:
acquiring original binarized image data corresponding to a target area;
And carrying out boundary extraction on the original binary image data based on an improved raster heuristic algorithm to obtain effective boundary image data, carrying out path planning on the current effective boundary image data according to a preset neighborhood scanning rule, obtaining next effective boundary image data adjacent to the current effective boundary image data, and carrying out path planning on the next effective boundary image data until the planning of the original binary image data is completed, so as to obtain a complete coverage path planning.
Optionally, the performing boundary extraction on the original binarized image data to obtain valid boundary image data includes:
expanding the original binarized image data to obtain target image data;
And extracting the boundary of the target image data to obtain the effective boundary image data.
Optionally, the performing path planning on the current valid boundary image data according to a preset neighborhood scanning rule, and obtaining next valid boundary image data adjacent to the current valid boundary image data, and performing path planning on the next valid boundary image data until the planning of the original binarized image data is completed, to obtain a complete coverage path plan, including:
Carrying out path planning on the effective boundary image data according to the preset neighborhood scanning rule to obtain boundary coverage path planning data;
deleting the boundary coverage path planning in the target image data, extracting the boundary of the processed target image number to obtain the next effective boundary image data, planning the path of the next effective boundary image data based on the preset neighborhood scanning rule until the path planning of the target area is completed, and obtaining the complete coverage path planning according to all the boundary coverage path planning data.
Optionally, the performing path planning on the effective boundary image data according to a preset neighborhood scanning rule to obtain boundary coverage path planning data includes:
And determining corresponding initial path points based on the effective boundary image data, starting from the initial path points, sequentially determining the path points in the effective boundary image data according to the preset neighborhood scanning rule until reaching a target path point, and obtaining the boundary coverage path planning data.
Optionally, starting from the initial path point, sequentially determining path points in the effective boundary image data according to the preset neighborhood scanning rule until reaching a target path point, and obtaining the boundary coverage path planning data, including:
Determining a first path point in the effective boundary image data based on the initial path point and the preset neighborhood scanning rule;
And determining a second path point in the rest effective boundary image data according to the first path point and a preset neighborhood scanning rule until reaching the target path point, and obtaining the boundary coverage path planning data.
Optionally, the preset neighborhood scanning rule includes an initial neighborhood point selection condition and a preset scanning direction; the determining a first path point in the effective boundary image data based on the initial path point and the preset neighborhood scanning rule includes:
Judging whether a neighborhood point exists in the effective boundary image data according to the preset scanning direction according to the current initial path point and the initial neighborhood point selection condition;
If yes, the neighborhood point is used as the first path point;
And if the effective boundary image data does not exist, acquiring the next effective boundary image data again, determining the initial path point of the next effective boundary image data according to the next effective boundary image data and the current initial path point, and taking the initial path point of the next effective boundary image data as the first path point.
Optionally, the valid boundary image data includes a plurality of sub-path points; said determining said initial waypoint for the next said valid boundary image data from the next said valid boundary image data and said initial waypoint comprising:
When the neighborhood point of the current initial path point does not exist in the effective boundary image data, generating the distance between the current initial path point and each sub path point in the next effective boundary image data to obtain a plurality of distance data;
And selecting the sub-path point corresponding to the minimum value from all the distance data as the initial path point of the next effective boundary image data.
Optionally, the laser cleaning intelligent coverage path planning method further includes:
obtaining the total distance of the complete coverage path according to the complete coverage path planning;
the total distance is:
Where L is the total distance, b tz is an information matrix for determining whether to go to the path point t and the path point z, dtz is a distance matrix for the path point t and the path point z, and D is the total number of the path points.
Compared with the prior art, the laser cleaning intelligent coverage path planning method has the advantages that: the method comprises the steps of obtaining original binarized image data corresponding to a workpiece (target area) to be cleaned, extracting boundaries of the image data by utilizing an improved grating heuristic algorithm to obtain effective boundary image data so as to more accurately identify boundary conditions (such as shape, size and the like) of the target area, planning a path of the effective boundary image data based on a preset neighborhood scanning rule, obtaining new effective boundary image data again, namely, carrying out path planning on the next effective boundary image data, repeating the operation until coverage planning of the target area is completed, namely, carrying out path planning on the target area by utilizing the improved grating heuristic algorithm and combining the preset neighborhood scanning rule, thereby obtaining complete coverage path planning. Therefore, the invention can more accurately plan the laser cleaning path by continuously planning the path of the effective boundary image data and acquiring the effective boundary image data of the next step, thereby improving the cleaning efficiency and coverage area. The boundary extraction and path planning are carried out by utilizing the improved grating heuristic algorithm, so that the waste of energy and time can be reduced, the cleaning quality is improved, meanwhile, the method can effectively identify and cover the path of the randomly generated target area to be cleaned, is simple to operate, saves labor and time cost, can greatly reduce the maintenance cost and resource consumption of a laser cleaning system, and improves the cleaning efficiency and benefit.
In order to solve the technical problem, the invention also provides a laser cleaning intelligent coverage path planning system, which comprises:
The acquisition unit is used for acquiring original binarized image data corresponding to the target area;
And the processing unit is used for extracting the boundary of the original binarized image data based on an improved grating heuristic algorithm to obtain effective boundary image data, planning a path of the current effective boundary image data according to a preset neighborhood scanning rule, obtaining the next effective boundary image data adjacent to the current effective boundary image data, and planning the path of the next effective boundary image data until the planning of the original binarized image data is completed, so as to obtain a complete coverage path plan.
The laser cleaning intelligent coverage path planning system and the laser cleaning intelligent coverage path planning method have the same advantages as those of the laser cleaning intelligent coverage path planning method in the prior art, and are not described in detail herein.
Drawings
FIG. 1 is a flow chart of a method for planning a laser cleaning intelligent coverage path in an embodiment of the invention;
FIG. 2 is a schematic diagram of a target image corresponding to a workpiece to be cleaned according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an effective boundary image of a target image according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a planning process for a current effective boundary in an embodiment of the present invention;
FIG. 5 is a schematic diagram of a complete coverage path planning in an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a laser cleaning intelligent coverage path planning device in an embodiment of the invention.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings.
As shown in fig. 1, in one embodiment, a method for planning a laser cleaning intelligent coverage path is provided, which specifically includes the following steps:
Step S1, original binarized image data corresponding to a target area is obtained.
Specifically, the original binarized image data corresponding to the workpiece to be cleaned (target area) is obtained, and digital imaging acquisition, coordinate matching and image processing functions can be performed on the surface of the workpiece to be cleaned in different cleaning stages, for example, the image acquisition device is used for acquiring the image of the workpiece to be cleaned, and calibration cutting is performed on the image to generate an RGB color image (RGB is the color representing three channels of red, green and blue) covering the surface of the workpiece to be cleaned.
After the color image is obtained, the color morphology image is converted into binary characteristic image data (comprising a binary characteristic image (shown in fig. 2) and corresponding data) by adopting an image processing method such as a color gamut space separation method, a gray threshold segmentation method, a closed-operation algorithm and the like, and the obtained binary characteristic image data is converted into a corresponding binary characteristic matrix (original binary image data) with target area space coordinate information by taking pixels as discrete units. The workpiece to be cleaned is often a material matrix with rust or adhesive paint and other dirt attached to the surface, and the material comprises steel, aluminum alloy, titanium alloy, composite glass and the like.
And S2, carrying out boundary extraction on the original binarized image data based on an improved grating heuristic algorithm to obtain effective boundary image data, carrying out path planning on the current effective boundary image data according to a preset neighborhood scanning rule, obtaining next effective boundary image data adjacent to the current effective boundary image data, and carrying out path planning on the next effective boundary image data until the planning of the original binarized image data is completed, so as to obtain a complete coverage path planning.
Specifically, a modified grating heuristic algorithm is utilized to conduct boundary extraction on the binarization feature matrix to obtain effective boundary image data, route planning is conducted on the current effective boundary image data according to a preset neighborhood scanning rule, and the effective boundary image data of the next step are obtained; and the effective boundary image data of the next step is subjected to path planning until the planning of the original binarized image data is completed, so that a complete coverage path planning is obtained, and the path of laser cleaning can be planned more accurately by continuously carrying out path planning and updating on the effective boundary image data.
According to the laser cleaning intelligent coverage path planning method, original binarization image data corresponding to a workpiece (target area) to be cleaned are obtained, and an improved grating heuristic algorithm is utilized to conduct boundary extraction on the image data, so that effective boundary image data are obtained; the method comprises the steps of identifying boundary conditions (such as shape, size and the like) of a target area more accurately, planning a path of effective boundary image data based on a preset neighborhood scanning rule, acquiring new effective boundary image data again, namely effective boundary image data of the next step, planning the path of the effective boundary image data, repeating the operation until coverage planning of the target area is completed, namely effectively planning the path of the target area by utilizing an improved grating heuristic algorithm and combining the preset neighborhood scanning rule, and thus obtaining complete coverage path planning. Therefore, the invention can more accurately plan the laser cleaning path by continuously planning the path of the effective boundary image data and acquiring the effective boundary image data of the next step, thereby improving the cleaning efficiency and coverage area. The boundary extraction and path planning are carried out by utilizing the improved grating heuristic algorithm, so that the waste of energy and time can be reduced, the cleaning quality is improved, meanwhile, the method can effectively identify and cover the path of the randomly generated target area to be cleaned, is simple to operate, saves labor and time cost, can greatly reduce the maintenance cost and resource consumption of a laser cleaning system, and improves the cleaning efficiency and benefit.
In some embodiments, step S1, performing boundary extraction on the original binarized image data to obtain valid boundary image data, includes:
expanding the original binarized image data to obtain target image data;
And extracting the boundary of the target image data to obtain the effective boundary image data.
Specifically, the background expansion is performed on the binary characteristic image to obtain target image data (target image and corresponding path point data), and for a sample to be cleaned which is identified randomly, the path points to be cleaned may exist on the boundary of the initial binary characteristic image, at the moment, the effective extraction and graphic display process of the initial edge of the target area are affected to a certain extent, so that the background expansion is performed on the original binary image, namely, a circle of invalid background is added around the original binary image, at the moment, the coordinate set of the binary characteristic image to be cleaned is equivalent to the vector bias of an [1,1] on the updated [0,0] point, and therefore, the reverse bias compensation is performed on the transformed coordinate point set, so that the process does not affect the subsequent global path planning process. The purpose of expanding the original binarized image is to ensure that the original boundary of the target area part is cleaned in the whole extracted binarized image.
The process for expanding the original binarized image comprises the following steps: the original binarized image can be subjected to background expansion by utilizing padarray (filling array) functions to obtain target image data, the target image data is an image after background boundary expansion and comprises a binary matrix corresponding to the target image data, the coordinates of the matrix after expansion are subjected to offset processing and are subjected to coordinate matching alignment with an original target area, then, the first boundary extraction is performed by utilizing bwperim functions to obtain effective boundary image data, namely an effective boundary image (the effective boundary image of the target image shown in fig. 3) and comprises corresponding boundary data (such as coordinate data of each path), wherein padarray functions are filling array functions, and the filling amount of each dimension is manually specified. The padarray function fills the numeric or logical image with a value of 0 and fills the classified image with a category < undefined >. By default paddarray adds padding before the first element and after the last element of each dimension, bwprim functions are functions that find the boundaries of objects in the binary image. A binary image is returned, which image contains only boundary pixels of the object in the input image BW (binary image). A pixel is part of a boundary if it is non-zero and is in communication with at least one zero value pixel.
By expanding the original binarized image data, boundary extraction can be more comprehensive and accurate, and a more accurate data basis is provided for subsequent path planning. Therefore, the efficiency and coverage range of laser cleaning can be improved, and the waste of energy and time is reduced, so that a better cleaning effect is achieved.
In some embodiments, in step S2, the performing path planning on the valid boundary image data according to a preset neighborhood scanning rule, and obtaining the next valid boundary image data, and repeating path planning on the next valid boundary image data until the planning of the original binarized image data is completed, to obtain a complete coverage path plan, including:
Step S21, path planning is carried out on the effective boundary image data according to a preset neighborhood scanning rule, and boundary coverage path planning data are obtained;
And S22, deleting the boundary coverage path planning in the target image data, carrying out boundary extraction on the processed target image number to obtain next effective boundary image data, carrying out path planning on the next effective boundary image data based on the preset neighborhood scanning rule until the path planning of the target area is completed, and obtaining the complete coverage path planning according to all the boundary coverage path planning data.
In some embodiments, in step S21, the performing path planning on the effective boundary image data according to a preset neighborhood scanning rule to obtain boundary coverage path planning data includes:
Step S211, determining a corresponding initial path point based on the effective boundary image data, starting from the initial path point, sequentially determining the path points in the effective boundary image data according to the preset neighborhood scanning rule until reaching a target path point, and obtaining the boundary coverage path planning data.
In some embodiments, in step S211, starting from the initial path point, sequentially determining path points in the valid boundary image data according to the preset neighbor scanning rule until reaching a target path point, to obtain the boundary coverage path planning data, including:
step T1, determining a first path point in the effective boundary image data based on the initial path point and the preset neighborhood scanning rule;
And step T2, determining a second path point in the rest of the effective boundary image data according to the first path point and a preset neighborhood scanning rule until the target path point is reached, and obtaining the boundary coverage path planning data.
In some embodiments, the preset neighbor scanning rule includes an initial neighbor point selection condition and a preset scanning direction, and step T1, the determining a first path point in the effective boundary image data based on the initial path point and the preset neighbor scanning rule includes:
Step T11, judging whether a neighborhood point exists in the effective boundary image data according to the preset scanning direction according to the current initial path point and the initial neighborhood point selection condition;
Step T12a, if yes, the neighborhood point is used as the first path point;
And step T12b, if not, re-acquiring the next valid boundary image data, determining the initial path point of the next valid boundary image data according to the next valid boundary image data and the current initial path point, and taking the initial path point of the next valid boundary image data as the first path point.
Specifically, in the path planning process, an initial path point, namely a current path point, is set in initial (first) effective boundary image data, and a next path point is acquired according to a preset neighborhood scanning rule, wherein the preset neighborhood scanning rule comprises an initial neighborhood point selection condition and a preset scanning direction, for example, as shown in fig. 4, a current search boundary is a current effective boundary, a closed target area is a target area (an initial target area or a planned residual target area), an 8 neighborhood searching criterion is shown as an initial neighborhood point which is set at the right upper corner of the current path point in the 8 neighborhood directions (namely, the right upper corner of the current path point in the 8 neighborhood directions of the current path point is used as an initial neighborhood point selection condition), and a clockwise direction (the direction of a thick arrow) is used as a preset scanning direction (a defined searching method), namely, 8 neighborhood scannable points (neighborhood points to be searched) around the current path point are set in the clockwise direction (a small arrow), the initial scanning point and the continuous scanning sequence of the current path point is set for ensuring the continuous neighborhood scanning process, the continuous scanning sequence of the current neighborhood point is required, the initial scanning point and the continuous neighborhood point is set in the continuous sequence, the embodiment is set as an initial neighborhood point, the initial neighborhood point which is set at the right upper corner, the first neighborhood point is set as an initial neighborhood point, the clockwise direction is set in the first neighborhood point, and the first neighborhood point is set in the first neighborhood point, and the clockwise direction is used as a first neighborhood point in the initial neighborhood point, the method is characterized in that a preset neighborhood scanning rule is facilitated to search a neighborhood point of a corresponding current path point in initial (first) effective boundary image data, the neighborhood point is used as a first path point (next path point), the searching (planning) process is repeated until a target path point is reached, namely the initial effective boundary image data is planned, and the target path point is the last path point in boundary coverage path planning data. In the planning process, the tabu list is updated after each search is completed, namely, the path points of each path are added into the tabu list in real time, so that the planned path points are prevented from being repeatedly covered.
After the planning of the initial effective boundary image data is completed, each path in the tabu list is removed from the initial target image data, the effective boundary image data is re-acquired from the rest target image data, and the initial path point of the current effective boundary image data is selected by calculating the distance between each pixel point (path point) in the current effective boundary image data and the initial effective boundary image data, namely the last path point in the last boundary coverage path planning data, taking the point closest to the last path point in the last boundary coverage path planning data as the initial point (initial path point) of the current effective boundary image data, and repeating the planning process, namely carrying out path planning on the initial path point in the current effective boundary image data according to a preset neighborhood scanning rule based on the current initial point, so as to obtain the current effective boundary image data, and finally obtaining the complete coverage path planning until the path planning of the corresponding target area is completed.
In some embodiments, the valid boundary image data includes a plurality of sub-path points, step T12b, the determining the initial path point of the next valid boundary image data from the next valid boundary image data and the initial path point includes:
When the neighborhood point of the current initial path point does not exist in the effective boundary image data, generating the distance between the current initial path point and each sub path point in the next effective boundary image data to obtain a plurality of distance data;
And selecting the sub-path point corresponding to the minimum value from all the distance data as the initial path point of the next effective boundary image data.
Specifically, when planning the effective boundary image data, when the current path point does not exist in the effective boundary image data, namely, when no path point exists in the 8 fields, the effective boundary image data is directly obtained again from the rest of the target image data, the distance between each path point in the effective boundary image data and the path point where no neighbor point exists currently is calculated, the point with the smallest distance is taken as the initial path point of the current effective boundary image data, namely, the next path point of the path point where no neighbor point exists currently, the path planning is repeatedly performed on the current effective boundary image data based on the preset neighbor scanning rule, and the process is iterated until the complete coverage path planning of the target area is completed. Where a greedy algorithm may be used to traverse the nearest waypoint from the previous beat of the overlapping path end point (last waypoint in the upper boundary overlapping path planning data) at the new target edge (current valid boundary image data).
For the case that the current path point does not have a neighborhood point, it can be considered that the current effective boundary image data has completed full coverage or that the target area may have a dead zone, that is, if the path is at a certain edge point, if the neighborhood range of the current path point does not have a searchable target point (neighborhood point) in the searching process, the neighborhood searching judges (presets a neighborhood scanning rule) whether the neighborhood of the current path point has an effective part to be cleaned, that is, if the scanning coordinate point value is 1, there is a searchable target point, and if all 8 neighborhood coordinate points are 0, there is no searchable target point. If the current binary image (current effective boundary) is judged to be 0, the complete coverage is considered to be finished, if the current binary image is judged to still have a coordinate point with a value of 1, the complete coverage is considered to be not finished, a new round of effective boundary needs to be extracted and corresponding path planning is carried out until the whole area is covered, and the completion of the path planning is marked.
The method has the advantages of ensuring the generation of the convex angle of the overall planning path as much as possible, ensuring the dynamic tracking precision requirement in the track tracking link based on the beam deflection control, and simultaneously inhibiting the boundary ablation effect caused by the traditional bow-shaped reciprocating scanning motion. In the improved grating heuristic method, initial effective boundary data can be accurately obtained through expanding a background area, and a preset neighborhood scanning rule is combined, so that a laser cleaning path can be accurately planned by continuously planning the effective boundary image data and obtaining the effective boundary image data of the next step, the cleaning efficiency and the coverage area are improved, the randomly generated target area to be cleaned can be effectively identified and covered, the operation is simple, and the labor and the time cost are saved.
It should be noted that, for the complete path-showing planning process, in case scan unit size is equivalent to pixel unit size, i.e., the discrete coordinate system of the path point is equivalent to the pixel coordinate system of the image.
In some embodiments, the laser cleaning intelligent coverage path planning method further comprises:
Step S3, obtaining the total distance of the complete coverage path according to the complete coverage path planning;
the total distance is:
Where L is the total distance, b tz is an information matrix for determining whether to go to the path point t and the path point z, dtz is a distance matrix for the path point t and the path point z, and D is the total number of the path points.
Specifically, referring to a partial description of the TSP (TSP, acronym for english total suspended particulate, i.e., total suspended particulates, also known as total suspended particulates) problem, the expression of the total distance that completely covers the target area is first determined as:
Where L is the total distance, b tz is an information matrix for determining whether to go to the path point t and the path point z, dtz is a distance matrix for the path point t and the path point z, and D is the total number of the path points.
The single access limit for the seek is:
s is a set containing all path points t and path points z, and V is a non-negative set.
The binary constraint of the waypoint selection is:
btz∈{0,1}，t,z＝1,2,...,D，t≠z； (3)
Based on the description of the TSP problem, all the path points to be planned in the purge target area are regarded as D path points to be accessed, and the distance cost matrix D tz is related only to linear distance factors in calculation units of the size of the discrete units.
The path points t and d are valid position point sets in the target area to be cleaned, which have not been updated to the path tabu list, and d tz may be considered as a probability or weight information matrix in the path searching process from the current path point to the next path point, and in the raster scanning in this embodiment, the information matrix is determined by a neighborhood criterion (preset neighborhood scanning rule) and a greedy criterion (greedy algorithm), for example: an information matrix, assuming 3 position points, from the first position point to the second position point is [01 0; 00 0; 00 < 0 > d is a row, t is a column, and a semicolon is a line feed; the information matrix from the first position point to the third position point is 0 < 1 >; 00 0; 00 ], the information matrix from the second position point to the third position point is [ 00 0; 0.1; 000 ], and so on. The information matrix iterates according to the pheromones. The external scan may determine that the current point is available with the next point.
In one embodiment, the method for planning the intelligent coverage path of the laser cleaning comprises the following specific steps:
Step A1, collecting a target image of a workpiece to be cleaned, and generating digital matrix input information (original binarized image data) for path planning.
Firstly, fixing a workpiece to be cleaned, performing graph sampling on an action section of the workpiece to be cleaned, calibrating a workpiece reference coordinate standard, and adjusting and matching a coordinate set of a target area. The workpiece to be cleaned illustrated in the embodiment is a natural rust floating area on the surface of a carbon steel workpiece in engineering.
Setting a color gamut offset blue range 133-192 by utilizing color space distribution corresponding to rust, obtaining a recognition result of a cleaning target by combining closed operation, and obtaining corresponding original binarized image data (comprising a binarized characteristic image and a binarized coordinate matrix), wherein the binarized characteristic image is shown in a figure 2, a white area is a rust area which is randomly generated on the surface of a recognized sample, namely a target area, and a black area is a non-rust part area of the surface of the sample, namely a background area of the sample to be cleaned;
And step A2, path planning initialization definition. In order to completely extract original binary image data, cleaning an initial boundary of a target area part, expanding an original binary coordinate matrix generated in the step A1 by utilizing padarray functions based on an improved raster heuristic algorithm to obtain target image data (comprising a binary matrix and a corresponding binary pattern) after expanding a background boundary, performing bias processing on the expanded matrix coordinate, and performing coordinate matching alignment with the original target area. And performing first boundary extraction by utilizing bwperim functions, wherein a schematic diagram of the first boundary extraction is shown in fig. 3. The scan initial coordinate value is determined or defined from the boundary in a non-fixed manner, and in this embodiment, the scan initial value is defined as the intersection point of the smallest row and the smallest column existing in the cleaning target area, that is, the lower right corner end point of the cleaning target area, and the original path position is recorded and the path processing of the next step is performed.
And step A3, setting an initial search point (initial neighborhood point) and a search direction (preset scanning direction) of an 8 neighborhood based on an improved raster heuristic algorithm, taking the initial search point (initial neighborhood point) and the search direction (preset scanning direction) as criteria for judging a current path searching state (current path point), and if the current search is effective, namely when the current path point exists in the field point, indicating that the current effective boundary is not covered, performing sequential coverage of the current boundary by using neighborhood scanning (preset neighborhood scanning rule) in a beat-by-beat mode, and simultaneously recording coordinates and sequence numbers of covered path points one by one to form continuous path information. Marking the covered path points as planned paths, wherein the planned path points are not identified by repeated searching in subsequent neighborhood searching; if the current neighborhood search has no effective target, namely the front path point does not have a domain point, the search state is switched to: the path planning is in a stage of cutting into inwards stepping one layer (namely, the current effective boundary is completely covered and the next effective boundary is acquired), or in a dead zone condition processing stage, namely, if the current boundary is completely covered or a dead zone is entered, extracting a new boundary (extraction of the new effective boundary), if the set unit distance is 1, obtaining the nearest boundary position (initial point of the new effective boundary) from the current path point by combining a greedy algorithm with traversal distance sequencing, continuously moving the planned path under different search states, simultaneously carrying out the path planning of the next round based on a preset neighborhood scanning rule until the area is not provided with an identifiable target, recording the whole path information, and obtaining the complete coverage path for the laser cleaning target. As shown in fig. 5, the final complete coverage path planning is schematically illustrated.
According to the laser cleaning intelligent coverage path planning method, original binarization image data corresponding to a workpiece (target area) to be cleaned are obtained, and an improved grating heuristic algorithm is utilized to conduct boundary extraction on the image data, so that effective boundary image data are obtained; the method comprises the steps of identifying boundary conditions (such as shape, size and the like) of a target area more accurately, planning a path of effective boundary image data based on a preset neighborhood scanning rule, acquiring new effective boundary image data again, namely effective boundary image data of the next step, planning the path of the effective boundary image data, repeating the operation until coverage planning of the target area is completed, namely effectively planning the path of the target area by utilizing an improved grating heuristic algorithm and combining the preset neighborhood scanning rule, and thus obtaining complete coverage path planning. Therefore, the invention can more accurately plan the laser cleaning path by continuously planning the path of the effective boundary image data and acquiring the effective boundary image data of the next step, thereby improving the cleaning efficiency and coverage area. The boundary extraction and path planning are carried out by utilizing the improved grating heuristic algorithm, so that the waste of energy and time can be reduced, the cleaning quality is improved, meanwhile, the method can effectively identify and cover the path of the randomly generated target area to be cleaned, is simple to operate, saves labor and time cost, can greatly reduce the maintenance cost and resource consumption of a laser cleaning system, and improves the cleaning efficiency and benefit.
Corresponding to the laser cleaning intelligent coverage path planning method, the embodiment of the invention also provides a laser cleaning intelligent coverage path planning system. As shown in fig. 6, the laser cleaning intelligent coverage path planning system includes:
The acquisition unit is used for acquiring original binarized image data corresponding to the target area;
And the processing unit is used for extracting the boundary of the original binarized image data based on an improved grating heuristic algorithm to obtain effective boundary image data, planning a path of the current effective boundary image data according to a preset neighborhood scanning rule, obtaining the next effective boundary image data adjacent to the current effective boundary image data, and planning the path of the next effective boundary image data until the planning of the original binarized image data is completed, so as to obtain a complete coverage path plan.
It should be noted that in this document, terms such as "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Although the invention is disclosed above, the scope of the invention is not limited thereto. Various changes and modifications may be made by one skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications will fall within the scope of the invention.
Claims (9)
1. The intelligent coverage path planning method for laser cleaning is characterized by comprising the following steps of:
acquiring original binarized image data corresponding to a target area;
And carrying out boundary extraction on the original binary image data based on an improved raster heuristic algorithm to obtain effective boundary image data, carrying out path planning on the current effective boundary image data according to a preset neighborhood scanning rule, obtaining next effective boundary image data adjacent to the current effective boundary image data, and carrying out path planning on the next effective boundary image data until the planning of the original binary image data is completed, so as to obtain a complete coverage path planning.
2. The method for planning a laser cleaning intelligent coverage path according to claim 1, wherein the performing boundary extraction on the original binarized image data to obtain effective boundary image data comprises:
expanding the original binarized image data to obtain target image data;
And extracting the boundary of the target image data to obtain the effective boundary image data.
3. The method for planning a laser cleaning intelligent coverage path according to claim 1, wherein the steps of planning a path of the current effective boundary image data according to a preset neighborhood scanning rule, obtaining the next effective boundary image data adjacent to the current effective boundary image data, planning a path of the next effective boundary image data until the planning of the original binary image data is completed, and obtaining a complete coverage path plan include:
Carrying out path planning on the effective boundary image data according to the preset neighborhood scanning rule to obtain boundary coverage path planning data;
deleting the boundary coverage path planning in the target image data, extracting the boundary of the processed target image number to obtain the next effective boundary image data, planning the path of the next effective boundary image data based on the preset neighborhood scanning rule until the path planning of the target area is completed, and obtaining the complete coverage path planning according to all the boundary coverage path planning data.
4. The method for planning a laser cleaning intelligent coverage path according to claim 3, wherein the performing path planning on the effective boundary image data according to a preset neighborhood scanning rule to obtain boundary coverage path planning data comprises:
And determining corresponding initial path points based on the effective boundary image data, starting from the initial path points, sequentially determining the path points in the effective boundary image data according to the preset neighborhood scanning rule until reaching a target path point, and obtaining the boundary coverage path planning data.
5. The method for planning a laser cleaning intelligent coverage path according to claim 4, wherein starting from the initial path point, sequentially determining path points in the effective boundary image data according to the preset neighborhood scanning rule until reaching a target path point, and obtaining the boundary coverage path planning data, includes:
Determining a first path point in the effective boundary image data based on the initial path point and the preset neighborhood scanning rule;
And determining a second path point in the rest effective boundary image data according to the first path point and a preset neighborhood scanning rule until reaching the target path point, and obtaining the boundary coverage path planning data.
6. The method for planning a laser cleaning intelligent coverage path according to claim 5, wherein the preset neighborhood scanning rule comprises an initial neighborhood point selection condition and a preset scanning direction; the determining a first path point in the effective boundary image data based on the initial path point and the preset neighborhood scanning rule includes:
Judging whether a neighborhood point exists in the effective boundary image data according to the preset scanning direction according to the current initial path point and the initial neighborhood point selection condition;
If yes, the neighborhood point is used as the first path point;
And if the effective boundary image data does not exist, acquiring the next effective boundary image data again, determining the initial path point of the next effective boundary image data according to the next effective boundary image data and the current initial path point, and taking the initial path point of the next effective boundary image data as the first path point.
7. The laser cleaning intelligent coverage path planning method of claim 6, wherein the effective boundary image data comprises a plurality of sub-path points; said determining said initial waypoint for the next said valid boundary image data from the next said valid boundary image data and said initial waypoint comprising:
When the neighborhood point of the current initial path point does not exist in the effective boundary image data, generating the distance between the current initial path point and each sub path point in the next effective boundary image data to obtain a plurality of distance data;
And selecting the sub-path point corresponding to the minimum value from all the distance data as the initial path point of the next effective boundary image data.
8. The laser cleaning intelligent coverage path planning method of claim 1, further comprising:
obtaining the total distance of the complete coverage path according to the complete coverage path planning;
the total distance is:
Where L is the total distance, b tz is an information matrix for determining whether to go to the path point t and the path point z, dtz is a distance matrix for the path point t and the path point z, and D is the total number of the path points.
9. A laser cleaning intelligent coverage path planning system, comprising:
The acquisition unit is used for acquiring original binarized image data corresponding to the target area;
And the processing unit is used for extracting the boundary of the original binarized image data based on an improved grating heuristic algorithm to obtain effective boundary image data, planning a path of the current effective boundary image data according to a preset neighborhood scanning rule, obtaining the next effective boundary image data adjacent to the current effective boundary image data, and planning the path of the next effective boundary image data until the planning of the original binarized image data is completed, so as to obtain a complete coverage path plan.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410155735.5A CN118015015A (en) | 2024-02-04 | 2024-02-04 | Laser cleaning intelligent coverage path planning method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410155735.5A CN118015015A (en) | 2024-02-04 | 2024-02-04 | Laser cleaning intelligent coverage path planning method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118015015A true CN118015015A (en) | 2024-05-10 |
Family
ID=90957909
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410155735.5A Pending CN118015015A (en) | 2024-02-04 | 2024-02-04 | Laser cleaning intelligent coverage path planning method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118015015A (en) |
-
2024
- 2024-02-04 CN CN202410155735.5A patent/CN118015015A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110378879B (en) | Bridge crack detection method | |
| JP4806230B2 (en) | Deterioration dictionary generation program, method and apparatus | |
| Lee et al. | Fusion of lidar and imagery for reliable building extraction | |
| US7245766B2 (en) | Method and apparatus for determining a region in an image based on a user input | |
| KR101199195B1 (en) | Method for generating label image and image processing system | |
| JP2951814B2 (en) | Image extraction method | |
| JP6872670B2 (en) | Dimension measuring device, dimensional measuring program and semiconductor manufacturing system | |
| JP2002328015A (en) | Semiconductor test system | |
| JP2000137804A (en) | Method and system for abnormality detection of digital image and storage medium for same | |
| Rakhmadi et al. | Connected component labeling using components neighbors-scan labeling approach | |
| KR102502486B1 (en) | Image generating system | |
| CN112508952B (en) | Pathological section double-objective lens self-adaptive scanning control method and system | |
| CN106203496B (en) | Hydrographic curve extracting method based on machine learning | |
| US5761385A (en) | Product and method for extracting image data | |
| CN118015015A (en) | Laser cleaning intelligent coverage path planning method and system | |
| JP3718967B2 (en) | Image feature amount extraction apparatus and method, and recording medium on which image feature amount extraction program is recorded | |
| JPH10269347A (en) | Method for eliminating shade shadow element on geographic image, geographic image processor and recording medium | |
| JPH0981740A (en) | Diagram input device | |
| JPH08194716A (en) | Picture processing method and its device | |
| Busch | Revision of built-up areas in a GIS using satellite imagery and GIS data | |
| CN114202665A (en) | Image similarity determining method and device, equipment and storage medium | |
| Sugahara et al. | An automatic image analysis system for RLGS films | |
| van den Broek et al. | Weighted distance mapping (WDM) | |
| JPH09245166A (en) | Pattern matching device | |
| JP2010074416A (en) | Image data processing method, image data processing apparatus, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |