Intelligent discharging method and device based on array combinatorial optimization and storage medium

Abstract

The invention discloses an intelligent discharging method and device based on array combinatorial optimization and a storage medium, and relates to the technical field of image processing, and the method comprises the steps: analyzing a target graphic file to extract cutting piece information in the target graphic file; classifying the cutting piece information so as to classify the outline of the flexible material and sub cutting pieces in the flexible material; constructing an initial typesetting scheme according to the flexible material contour and the sub-cutting pieces; selecting a reference typesetting scheme from the initial typesetting schemes according to a preset fitness function; performing cross processing on the reference typesetting scheme; performing variation processing on the part of the reference typesetting schemes after the cross processing; carrying out correction processing on the reference typesetting scheme which is subjected to cross processing and is not subjected to variation processing and the reference typesetting scheme which is subjected to variation processing; and taking the reference typesetting scheme with the highest fitness as a target typesetting scheme. Typesetting of the sub-cutting pieces is regarded as a combinatorial optimization problem, and the material utilization rate is improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an intelligent nesting method, device and storage medium based on array combination optimization. Background Art

[0002] Smart nesting technology is an advanced method that combines computer-aided design and artificial intelligence technology, mainly used to optimize material cutting and arrangement during the manufacturing process. This technology has a wide range of applications in many fields, including but not limited to cutting, welding, processing of flexible materials (such as cloth, leather) and CNC equipment such as woodworking engraving.

[0003] Although intelligent nesting technology has made some important breakthroughs in the fields of artificial intelligence, big data, etc., in some cases, the existing intelligent nesting system may not achieve the desired effect.

[0004] For example, for the layout of special-shaped plates, although intelligent nesting technology can provide a preliminary solution, it often requires further optimization by humans; this is because the automatically generated layout solution may not always be optimal, especially when dealing with complex shapes.

[0005] Therefore, with regard to the cutting problem in the manufacturing industry, how to maximize material utilization through optimization algorithms is an issue that needs to be addressed urgently. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an intelligent nesting method, device and storage medium based on array combination optimization, which can improve the material utilization rate.

[0007] In order to solve the above technical problems, the present invention provides an intelligent nesting method based on array combination optimization, including: S1, parsing a target graphic file to extract piece information in the target graphic file; S2, classifying the piece information to classify the flexible material contour and the sub-pieces in the flexible material; S3, constructing an initial layout scheme according to the flexible material contour and the sub-pieces; S4, selecting a benchmark layout scheme from the initial layout scheme according to a preset fitness function; S5, cross-processing the benchmark layout scheme; S6, mutating part of the benchmark layout scheme after cross-processing; S7, correcting the benchmark layout scheme that has been cross-processed and not mutated and the benchmark layout scheme after mutation processing; S8, repeating iterative steps S4 to S7 until the fitness of the benchmark layout scheme meets the convergence condition; S9, taking the benchmark layout scheme with the highest fitness as the target layout scheme.

[0008] As an improvement of the above solution, the fitness function is: f (x)=U-aS, wherein U is the utilization rate of the flexible material, S is the overlapping area between the sub-pieces, and a is a weight factor for controlling the penalty of the overlapping area.

[0009] As an improvement of the above solution, the goal of the fitness function is to maximize the utilization rate of the flexible material and minimize the overlapping area between the sub-pieces.

[0010] As an improvement of the above-mentioned scheme, the step of constructing an initial layout scheme based on the flexible material contour and the sub-pieces includes: randomly generating a gene sequence for representing the initial layout scheme based on preset constraints, the flexible material contour and the sub-pieces; the genes in the gene sequence are used to represent the position information and angle information of the sub-pieces in the flexible material.

[0011] As an improvement of the above scheme, the step of cross-processing the benchmark typesetting scheme includes: grouping the benchmark typesetting scheme, each group includes two benchmark typesetting schemes; exchanging some genes of the two benchmark typesetting schemes in the group to generate two new benchmark typesetting schemes.

[0012] As an improvement of the above-mentioned scheme, the step of mutating part of the baseline typesetting scheme after cross processing includes: extracting a preset proportion of the baseline typesetting scheme from the baseline typesetting scheme after cross processing as the typesetting scheme to be mutated; randomly adjusting the position information of the sub-piece in the typesetting scheme to be mutated; randomly adjusting the angle information of the sub-piece in the typesetting scheme to be mutated.

[0013] As an improvement of the above-mentioned scheme, the step of correcting the baseline typesetting scheme that has been cross-processed and not mutated and the baseline typesetting scheme after mutating includes: performing a boundary check on the baseline typesetting scheme that has been cross-processed and not mutated and the baseline typesetting scheme after mutating to adjust the position information of the sub-pieces that exceed the boundary of the flexible material; performing an overlap check on the baseline typesetting scheme that has been cross-processed and not mutated and the baseline typesetting scheme after mutating to adjust the position information and / or angle information of the sub-pieces whose overlapping area exceeds a preset allowable value.

[0014] As an improvement of the above-mentioned solution, the intelligent nesting method based on array combination optimization also includes: converting the target layout scheme into line information and displaying the line information.

[0015] Correspondingly, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein the processor implements the steps of the above-mentioned intelligent nesting method based on array combination optimization when executing the computer program.

[0016] Correspondingly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-mentioned intelligent nesting method based on array combination optimization.

[0017] The implementation of the present invention has the following beneficial effects: The intelligent nesting method based on array combination optimization of the present invention regards the layout of sub-pieces as a combination optimization problem, and explores the optimal layout scheme by optimizing the spacing, angle and overlapping area between sub-pieces, thereby improving the material utilization rate; Furthermore, the present invention introduces the overlapping area into the fitness function, that is, the sub-pieces are allowed to overlap slightly, thereby realizing material-saving operation; More preferably, the present invention helps to generate new typesetting schemes through crossover operations and mutation operations; wherein, the crossover operation generates new individuals by exchanging some genes in the gene sequences of two parent individuals (i.e., the baseline typesetting scheme), and the mutation operation randomly adjusts some positions or angles of genes in the gene sequence to avoid falling into a local optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of the first embodiment of the intelligent nesting method based on array combination optimization of the present invention; Figure 2 It is a schematic diagram of the neutron cut piece of the present invention; Figure 3 is a schematic diagram of the target typesetting scheme in the present invention; Figure 4 It is a flow chart of the second embodiment of the intelligent nesting method based on array combination optimization of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. It is hereby stated that the directional terms such as up, down, left, right, front, back, inside, outside, etc. that appear or will appear in the text of the present invention are only based on the accompanying drawings of the present invention, and are not specific limitations of the present invention.

[0020] See also Figure 1 , Figure 1 The flowchart of the first embodiment of the intelligent nesting method based on array combination optimization of the present invention is shown, which includes: S101, parsing a target graphic file to extract piece information in the target graphic file; The software can be used to parse target graphic files of various formats to extract the cutting piece information in the target graphic files.

[0021] S102, classifying the cut piece information to classify the flexible material contour and the sub-cut pieces in the flexible material; The piece information in the target graphic file is classified to identify the flexible material contours and the sub-pieces in the flexible material, so as to facilitate comparison and sorting operations.

[0022] like Figure 2 As shown, the sub-pieces in the present invention include a target shape area M1 and a sewing area M2 surrounding the target shape area M1; wherein the sewing area M2 is used for sewing, and the sewing area M2 can be blocked / hidden by the target shape area M1 after sewing; for example, a bag sub-piece, a shoe upper sub-piece, etc. Therefore, the sewing areas M2 between different sub-pieces can overlap slightly, and in the case of slight overlap, there is still enough space for sewing and ensuring the integrity of the target shape area M1.

[0023] S103, constructing an initial layout plan according to the flexible material contour and the sub-pieces; Specifically, a gene sequence for representing an initial layout scheme can be randomly generated according to preset constraints, flexible material contours and sub-pieces; wherein the genes in the gene sequence are used to represent position information and angle information of the sub-pieces in the flexible material.

[0024] For example, a number (e.g., 50-100) of initial layout plans are randomly generated. Each initial layout plan can be represented as a gene sequence. Each gene in the gene sequence represents the position information and angle information of a sub-piece. The specific gene can be represented by a triple as follows: [x,y,θ] in, (x, y) represents the position information of the sub-piece; θ represents the angle information of the sub-piece (i.e. the rotation angle).

[0025] The corresponding constraint conditions are: the randomly generated initial layout scheme should meet the basic flexible material contour restrictions, that is, the sub-pieces cannot exceed the flexible material boundaries.

[0026] It should be noted that the initial population of existing genetic algorithms is usually generated completely randomly and lacks domain knowledge guidance. Unlike the prior art, the present invention generates the initial gene sequence based on the flexible material contour and the sub-piece shape, combined with preset constraints (such as boundary restrictions), to ensure that the sub-piece does not exceed the material range; at the same time, the present invention uses the triple (x, y, θ) to represent the position and rotation angle of the sub-piece, directly related to the physical characteristics of the typesetting problem, and improves the feasibility of the initial solution.

[0027] S104, selecting a reference layout scheme from the initial layout schemes according to a preset fitness function; It should be noted that most of the existing fitness functions have a single objective (such as maximizing utilization only) and do not explicitly weigh the overlap penalty. Unlike the prior art, the fitness function of the present invention adopts a multi-objective balance strategy, dynamically combines material utilization with overlap area, controls overlap tolerance through weight factors, and allows slight overlap to maximize material savings.

[0028] Specifically, the present invention introduces overlapping area into the fitness function, that is, slight overlap is allowed between sub-pieces, thereby achieving material-saving operation; the goal of the fitness function is to maximize the utilization rate of flexible materials and minimize the overlapping area between sub-pieces under the premise that the allowed overlap is as close to the set tolerance as possible.

[0029] Accordingly, the fitness function of the present invention is: f (x) = U - aS in, U is the utilization rate of flexible materials; S is the overlapping area between the sub-pieces; a is the weight factor that controls the overlap area penalty.

[0030] It should be noted that the flexible material utilization rate is the ratio of the coverage area of ​​all sub-cut pieces in the current initial layout plan to the total area of ​​the flexible material, and this ratio needs to be maximized; At the same time, the overlapping area can be penalized according to the allowed overlapping area of ​​each sub-piece. The more overlap, the lower the fitness; In addition, in general, the value of the weight factor can be flexibly adjusted according to the characteristics of materials and processes to enhance the versatility of the algorithm; for example, when applied to leather, the value of the weight factor needs to be less than 0.03.

[0031] After calculating the fitness of the initial layout schemes according to the fitness function, select N initial layout schemes with higher fitness as the benchmark layout schemes, where N is preferably 10, but is not limited to this and can be set according to actual conditions. When selecting, you can use strategies such as Roulette Wheel Selection or Tournament Selection.

[0032] S105, cross-processing the benchmark typesetting scheme; Specifically, the steps of cross-processing the benchmark typesetting scheme include: (1) The benchmark typesetting schemes are grouped, with each group including two benchmark typesetting schemes; (2) Exchange some genes of the two benchmark typesetting schemes within the group to generate two new benchmark typesetting schemes.

[0033] Among them, the commonly used crossover methods include: (1) Single-point crossover: Cut at a random position in the gene sequence and exchange the front and back parts to generate two new individuals; (2) Two-point crossover: Select two positions in the gene sequence for cutting and exchange.

[0034] It should be noted that the existing crossover and mutation lack targeted optimization of layout geometric characteristics; unlike the existing technology, the present invention adopts a layout-guided crossover strategy, using single-point / double-point crossover to exchange gene sequences, retaining effective local structures in the parent layout (such as densely arranged areas) and improving the quality of offspring.

[0035] S106, performing mutation processing on some of the reference typesetting schemes after the cross processing; Specifically, the step of performing mutation processing on the part of the reference typesetting schemes after the cross processing includes: (1) Extracting a preset proportion of the benchmark typesetting scheme from the cross-processed benchmark typesetting scheme as the typesetting scheme to be mutated; The preset ratio generally adopts a smaller probability, preferably 5%, but it is not limited to this and can be adjusted according to actual conditions as long as excessive randomization is avoided.

[0036] (2) Randomly adjust the position information (x, y) of the sub-piece in the layout scheme to be mutated; (3) Randomly adjust the angle information θ of the sub-piece in the layout scheme to be mutated.

[0037] It should be noted that by adjusting the position (x, y) and angle (θ), the diversity of solutions can be increased to meet the rotation requirements of complex shapes.

[0038] Combining the crossover operation of step S105 and the mutation operation of step S106 helps to generate a new layout scheme; wherein the crossover operation generates a new individual by exchanging some genes in the gene sequences of two parent individuals (i.e., the benchmark layout scheme), while the mutation operation randomly adjusts some positions or angles of genes in the gene sequence to avoid falling into a local optimal solution.

[0039] S107, performing correction processing on the cross-processed and unmutated benchmark typesetting scheme and the mutated benchmark typesetting scheme; It should be noted that, in actual situations, the baseline typesetting scheme after processing through steps S101 to S106 is rarely densely packed (i.e., the utilization rate is 100%); in general, its utilization rate is about 70% to 80%, i.e., the remaining spaces can be fine-tuned.

[0040] Therefore, in order to ensure that the generated layout scheme is valid, it is necessary to check the boundary conditions and overlap conditions. Specifically, the steps of correcting the cross-processed and unmutated baseline layout scheme and the mutated baseline layout scheme include: (1) Performing boundary checks on the cross-processed and unmutated baseline layout scheme and the mutated baseline layout scheme to adjust the position information of the sub-pieces that exceed the boundary of the flexible material; That is, if the child piece exceeds the boundary of the flexible material, reposition it.

[0041] Therefore, boundary checking can be used to force the position of sub-pieces that exceed the material boundary to ensure the feasibility of the solution.

[0042] (2) performing an overlap check on the cross-processed and unmutated baseline layout scheme and the mutated baseline layout scheme to adjust the position information and / or angle information of the sub-pieces whose overlapping area exceeds a preset allowable value.

[0043] That is, if the overlap between sub-pieces exceeds the allowed value, try to reduce the overlap by fine-tuning the position or angle.

[0044] Therefore, by fine-tuning the position and angle of the sub-pieces, unnecessary overlaps can be reduced and invalid solutions can be reduced.

[0045] S108, repeating the iteration steps S104 to S107 until the fitness of the benchmark typesetting scheme meets the convergence condition; Generally, step S104 to step S107 may be repeated and iterated a certain number of times (eg, 500 times), and the best layout scheme (ie, the individual with the highest fitness) is retained each time.

[0046] In addition, you can also set convergence conditions, such as stopping iteration when the fitness does not improve significantly between several generations.

[0047] Existing iterative algorithms usually only rely on fitness or number of iterations as termination conditions; unlike the prior art, the iterative algorithm adopted by the present invention adopts dynamic convergence judgment, that is, combining fitness stability and number of iterations (such as 500 times) to avoid premature convergence, and also adopts an elite retention mechanism, that is, each iteration retains the individual with the highest fitness to ensure that the optimization direction does not deviate.

[0048] S109, taking the benchmark typesetting scheme with the highest fitness as the target typesetting scheme.

[0049] like Figure 3 As shown in FIG. 1 , after all iterations are completed, the layout scheme with the highest fitness is output as the final target layout scheme.

[0050] See also Figure 4 , Figure 4 The flowchart of the second embodiment of the intelligent nesting method based on array combination optimization of the present invention is shown, which includes: S201, parsing the target graphic file to extract the piece information in the target graphic file; S202, classifying the cut piece information to classify the flexible material contour and the sub-cut pieces in the flexible material; S203, constructing an initial layout plan according to the flexible material contour and the sub-pieces; S204, selecting a reference layout scheme from the initial layout schemes according to a preset fitness function; S205, cross-processing the benchmark typesetting scheme; S206, performing mutation processing on some of the reference typesetting schemes after the cross processing; S207, performing correction processing on the cross-processed and unmutated benchmark typesetting scheme and the mutated benchmark typesetting scheme; S208, repeating the iteration steps S204 to S207 until the fitness of the benchmark typesetting scheme meets the convergence condition; S209, taking the benchmark typesetting scheme with the highest fitness as the target typesetting scheme.

[0051] S210, converting the target typesetting scheme into line information, and displaying the line information.

[0052] and Figure 1 Different from the first embodiment shown, this embodiment also outputs all sub-piece lines in the target layout scheme after processing in steps S201 to S209 as visual line information, and displays it on the software interface, allowing the operator to check the final layout effect and start cutting after confirming that it is correct.

[0053] In summary, the intelligent layout method based on array combination optimization of the present invention regards the layout of sub-pieces as a combinatorial optimization problem, and explores the optimal layout scheme by optimizing the spacing, angles and overlapping areas between sub-pieces, thereby achieving improved material utilization.

[0054] Correspondingly, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein the processor implements the steps of the above-mentioned intelligent nesting method based on array combination optimization when executing the computer program.

[0055] At the same time, the present invention also discloses a computer-readable storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the above-mentioned intelligent nesting method based on array combination optimization are implemented.

[0056] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. An intelligent nesting method based on array combination optimization, characterized in that: include: S1, parsing a target graphic file to extract piece information in the target graphic file; S2, classifying the piece information to classify the flexible material contour and the sub-pieces in the flexible material; S3, constructing an initial layout plan according to the flexible material contour and the sub-pieces; S4, selecting a reference layout scheme from the initial layout schemes according to a preset fitness function; S5, performing cross processing on the reference typesetting scheme; S6, performing mutation processing on some of the benchmark typesetting schemes after the cross processing; S7, modifying the cross-processed and unmutated benchmark typesetting scheme and the mutated benchmark typesetting scheme; S8, repeating the iteration steps S4 to S7 until the fitness of the benchmark typesetting scheme meets the convergence condition; S9, taking the benchmark layout scheme with the highest fitness as the target layout scheme.

2. The intelligent nesting method based on array combination optimization according to claim 1, characterized in that: The fitness function is: f (x) = U - aS Among them, U is the utilization rate of flexible materials, S is the overlapping area between sub-pieces, and a is the weight factor that controls the penalty of overlapping area.

3. The intelligent nesting method based on array combination optimization as claimed in claim 2, characterized in that: The goal of the fitness function is to maximize the utilization of the flexible material and minimize the overlapping area between the sub-panels.

4. The intelligent nesting method based on array combination optimization according to claim 1, characterized in that: The step of constructing an initial layout scheme according to the flexible material contour and the sub-pieces comprises: Randomly generate a gene sequence for representing an initial layout plan according to preset constraints, the flexible material contour and the sub-pieces; The genes in the gene sequence are used to represent the position information and angle information of the sub-piece in the flexible material.

5. The intelligent nesting method based on array combination optimization according to claim 1 or 4, characterized in that: The step of cross-processing the benchmark typesetting scheme comprises: The benchmark typesetting schemes are grouped, each group including two benchmark typesetting schemes; Part of the genes of two benchmark typesetting schemes in the group are exchanged to generate two new benchmark typesetting schemes.

6. The intelligent nesting method based on array combination optimization according to claim 1, characterized in that: The step of performing variation processing on the part of the reference typesetting schemes after the cross processing comprises: Extracting a preset proportion of the benchmark typesetting scheme from the cross-processed benchmark typesetting scheme as the typesetting scheme to be mutated; Randomly adjust the position information of the sub-pieces in the layout scheme to be mutated; Randomly adjust the angle information of the sub-pieces in the layout scheme to be mutated.

7. The intelligent nesting method based on array combination optimization according to claim 1, characterized in that: The step of correcting the cross-processed and unmutated benchmark typesetting scheme and the mutated benchmark typesetting scheme comprises: Perform boundary check on the cross-processed and unmutated baseline layout scheme and the mutated baseline layout scheme to adjust the position information of the sub-pieces beyond the boundary of the flexible material; An overlap check is performed on the cross-processed and unmutated benchmark layout scheme and the mutated benchmark layout scheme to adjust the position information and / or angle information of the sub-pieces whose overlapping area exceeds a preset allowable value.

8. The intelligent nesting method based on array combination optimization according to claim 1, characterized in that: Also includes: The target typesetting scheme is converted into line information, and the line information is displayed.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent nesting method based on array combination optimization described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent nesting method based on array combination optimization described in any one of claims 1 to 8 are implemented.

Images