Production process optimization method and device for embroidery
Through digital pattern processing and process optimization algorithms, adaptively setting needle jump parameters is solved, which solves the problem of low automation in traditional embroidery production processes and achieves high-precision and high-efficiency embroidery production.
Patent Information
- Application Number
- CN202411722787.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-28
- Publication Date
- 2025-03-11
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In traditional embroidery production processes, the jumping needle parameters need to be set subjectively, resulting in a low degree of automation and cannot meet the needs of modern embroidery industry for high-precision and high-efficiency production.
By obtaining the digital patterns of the embroidery sample, color segmentation and needle jump analysis, pattern and layer needle jump point sequence are generated, and process evaluation functions are constructed in combination with the evaluation factor set, and optimization is carried out to generate the recommended needle jump point sequence and embroidery path to achieve adaptive optimization of the production process.
It improves the degree of automation of embroidery production, ensures the quality and stability of products, and meets the demands of modern embroidery industry for high-precision and high-efficiency production.
Smart Images
Figure CN119593152A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of data recognition, and in particular to a production process optimization method and device for embroidery. Background Art
[0002] With the rapid development of the global textile and apparel industry, embroidery, as an important decorative craft, has a growing market demand. However, the traditional embroidery production process faces many challenges. First, the setting of skip stitch parameters in the traditional process mainly depends on the operator's experience and subjective judgment, which not only leads to low production efficiency, but also makes it difficult to ensure the quality and stability of embroidery products. Secondly, due to the low degree of automation control, traditional embroidery machines are unable to cope with complex patterns and special materials, and cannot meet the modern embroidery industry's demand for high-precision and high-efficiency production. In order to meet these challenges, the embroidery industry urgently needs to introduce new technical solutions to improve the automation and intelligence level of the production process. Summary of the invention
[0003] The present application provides a method and device for optimizing the production process of embroidery, which solves the technical problem that the traditional embroidery production process requires subjective setting of skip stitch parameters, resulting in a poor degree of automation.
[0004] In view of the above problems, the embodiments of the present application provide a method and device for optimizing the production process of embroidery.
[0005] In a first aspect of an embodiment of the present application, a method for optimizing a production process for embroidery is provided, the method comprising: Obtain an embroidery sample digital pattern, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has several layers of digital patterns; traverse the first sub-pattern, the second sub-pattern to the M-th sub-pattern, perform color segmentation on the several layers of digital patterns, and generate M several layers of digital pattern color partitioning results; perform pattern skipping analysis based on the first sub-pattern, the second sub-pattern to the M-th sub-pattern, and generate a pattern skipping point sequence; traverse the M several layers of digital pattern color partitioning results to perform layer skipping analysis, and generate M layer skipping point sequences; based on the pattern skipping point sequence and the M layers A skipped stitch point sequence is used to analyze the color partition results of the M layers of digital patterns and generate a digital pattern embroidery path; a process evaluation function is constructed based on an evaluation factor set, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set includes at least the number of long skipped stitches, the number of skipped stitches, the total length of skipped stitches, and the total length of the embroidery path; the pattern skipped stitch point sequence, the M layer skipped stitch point sequences, and the digital pattern embroidery path are optimized according to the process evaluation function to generate a recommended pattern skipped stitch point sequence, M recommended layer skipped stitch point sequences, and a recommended digital pattern embroidery path; the production process is initialized according to the recommended pattern skipped stitch point sequence, the M recommended layer skipped stitch point sequences, and the recommended digital pattern embroidery path.
[0006] A second aspect of the embodiment of the present application provides a production process optimization device for embroidery, the device comprising: A pattern acquisition module, the pattern acquisition module is used to obtain an embroidery sample digital pattern, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has several layers of digital patterns; a color segmentation module, the color segmentation module is used to traverse the first sub-pattern, the second sub-pattern to the M-th sub-pattern, perform color segmentation on the several layers of digital patterns, and generate M several layers of digital pattern color partition results; a first skipped stitch analysis module, the first skipped stitch analysis module is used to perform pattern skipped stitch analysis based on the first sub-pattern, the second sub-pattern to the M-th sub-pattern, and generate a pattern skipped stitch point sequence; a second skipped stitch analysis module, the second skipped stitch analysis module is used to traverse the M several layers of digital pattern color partition results to perform layer skipped stitch analysis, and generate M layer skipped stitch point sequences; a path analysis module, the path analysis module The analysis module is used to perform embroidery path analysis on the color partition results of the M layers of digital patterns based on the pattern skip stitch point sequence and the M layer skip stitch point sequences, and generate a digital pattern embroidery path; the function construction module is used to construct a process evaluation function based on an evaluation factor set, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set at least includes the number of long skip stitches, the number of skip stitches, the total length of skip stitches, and the total length of the embroidery path; the optimization module is used to optimize the pattern skip stitch point sequence, the M layer skip stitch point sequences, and the digital pattern embroidery path according to the process evaluation function, and generate a recommended pattern skip stitch point sequence, M recommended layer skip stitch point sequences, and a recommended digital pattern embroidery path; the process control module is used to initialize the production process according to the recommended pattern skip stitch point sequence, the M recommended layer skip stitch point sequences, and the recommended digital pattern embroidery path.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, an embroidery sample digital pattern is obtained, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has several layers of digital patterns. Next, the first sub-pattern, the second sub-pattern to the M-th sub-pattern are traversed, and the several layers of digital patterns are color segmented to generate M several layers of digital pattern color partition results. Then, pattern skipping analysis is performed based on the first sub-pattern, the second sub-pattern to the M-th sub-pattern to generate a pattern skipping point sequence. Layer skipping analysis is performed by traversing M several layers of digital pattern color partition results to generate M layer skipping point sequences. Next, based on the pattern skipping point sequence and the M layer skipping point sequence, the M several layers of digital pattern color partition results are analyzed for embroidery paths to generate digital pattern embroidery paths. Based on the evaluation factor set, a process evaluation function is constructed, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set at least includes the number of long skipping stitches, the number of skipping stitches, the total length of skipping stitches, and the total length of the embroidery path. Furthermore, the pattern skipping stitch point sequence, M layer skipping stitch point sequences and digital pattern embroidery paths are optimized according to the process evaluation function to generate the recommended pattern skipping stitch point sequence, M recommended layer skipping stitch point sequences and recommended digital pattern embroidery paths. Finally, the production process is initialized according to the recommended pattern skipping stitch point sequence, M recommended layer skipping stitch point sequences and recommended digital pattern embroidery paths. The technical problem of the poor degree of automation caused by the subjective setting of skipping stitch parameters in the traditional embroidery production process is solved. By adaptively setting the skipping stitch parameters, the technical effect of improving the degree of automation is achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0009] Figure 1 A schematic diagram of a process flow of an embroidery production process optimization method provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a production process optimization device for embroidery provided in an embodiment of the present application.
[0010] Explanation of the reference numerals: pattern acquisition module 11 , color segmentation module 12 , first skip stitch analysis module 13 , second skip stitch analysis module 14 , path analysis module 15 , function construction module 16 , optimization module 17 , process control module 18 . DETAILED DESCRIPTION
[0011] The embodiments of the present application provide a method and device for optimizing the production process of embroidery, thereby solving the technical problem that in the traditional embroidery production process, the skipping stitch parameters need to be set subjectively, resulting in a low degree of automation.
[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0013] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices. Embodiment 1
[0014] like Figure 1 As shown, the embodiment of the present application provides a production process optimization method for embroidery, wherein the method includes: An embroidery sample digital pattern is obtained, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has several layers of digital patterns; the first sub-pattern, the second sub-pattern to the M-th sub-pattern are traversed, the several layers of digital patterns are color segmented, and M several layers of digital pattern color partition results are generated.
[0015] The embroidery sample digital pattern contains the detailed design of the entire embroidery, usually including multiple sub-patterns, from the first sub-pattern to the Mth sub-pattern, and each sub-pattern has its own unique design and details. Each of these sub-patterns is composed of several layers of digital patterns. These layers can represent different colors, textures or other design elements, which together constitute the complex structure of the entire embroidery. By obtaining the embroidery sample digital pattern and performing color segmentation on the several layers of digital patterns of each sub-pattern, different color areas in the same layer are distinguished, and M several layers of digital pattern color partitioning results are generated, so that these colors can be accurately identified and processed in the subsequent embroidery process.
[0016] Specifically, traversing the first sub-pattern, the second sub-pattern, and finally the Mth sub-pattern, performing color segmentation on the plurality of layers of digital patterns, and generating M color partitioning results of the plurality of layers of digital patterns, includes: Obtain the j-th layer digital pattern of the i-th sub-pattern, wherein i∈[1,M], j∈[1,N], i is an integer, j is an integer, and N is the total number of embroidery layers of the i-th sub-pattern; perform neighborhood color aggregation on the j-th layer digital pattern to generate a j-th layer color partition result; add the j-th layer color partition result to the i-th several layers digital pattern color partition result; add the i-th several layers digital pattern color partition result to the M several layers digital pattern color partition results.
[0017] Preferably, the j-th layer digital pattern of the i-th sub-pattern is obtained from the embroidery sample digital pattern, wherein i∈[1,M], M is the total number of sub-patterns in the embroidery sample digital pattern, j∈[1,N], N is the total number of embroidery layers of the i-th sub-pattern, i is an integer, j is an integer, and N is the total number of embroidery layers of the i-th sub-pattern; neighborhood color aggregation is performed on the j-th layer digital pattern. Neighborhood color aggregation is an image processing technology used to combine similar color pixels into larger color areas, thereby reducing color diversity and simplifying subsequent processing. The result of neighborhood color aggregation is one or more color partitions, each Each partition contains a group of pixels with similar colors and adjacent in space. Specifically, through the conversion of color space (such as from RGB to HSV), the color similarity is measured in the color space, and similar color pixels are merged into the same area; after completing the neighborhood color aggregation of the j-th layer of digital pattern, the j-th layer color partition result is generated, and the j-th layer color partition result is added to the i-th several layers of digital pattern color partition result; after completing the i-th several layers of digital pattern color segmentation, the i-th several layers of digital pattern color partition result is added to the M several layers of digital pattern color partition results.
[0018] Furthermore, performing neighborhood color aggregation on the j-th layer of digital patterns to generate a j-th layer of color partitioning results includes: Randomly extract the color feature value of the first coordinate from the j-th layer of digital patterns; randomly extract the color feature value of the first adjacent coordinate of the first coordinate from the j-th layer of digital patterns; when the color deviation distance between the first adjacent coordinate color feature value and the first coordinate color feature value is less than or equal to the color deviation distance threshold, aggregate the first adjacent coordinate and the first coordinate to generate a first updated coordinate, wherein the first updated coordinate has the number of first updated coordinate pixels and the first updated coordinate color feature value; based on the first updated coordinate pixel number and the first updated coordinate color feature value, cyclic analysis is performed, and when the color deviation distance between any two neighborhoods is greater than the color deviation distance threshold, output the j-th layer color partition result.
[0019] A coordinate point is randomly selected from the j-th layer of digital pattern as the first coordinate, and the color feature value of the first coordinate is extracted. The color feature value is usually a multidimensional vector used to represent various attributes of color, such as RGB value, HSV value or LAB value, etc. For the first coordinate, its adjacent coordinate points are found (usually in four directions of up, down, left and right, or eight directions), and the first adjacent coordinate color feature value of the first coordinate is extracted; the color deviation distance between the first adjacent coordinate color feature value and the first coordinate color feature value is calculated, and the color deviation distance is obtained by calculating the Euclidean distance and the Manhattan distance between the two color feature values; if the calculated color deviation distance is less than or equal to the preset color deviation distance threshold, it is considered that the colors of the two coordinate points are similar, and they can be aggregated together. The result of the aggregation is a new coordinate as the first updated coordinate, and the first updated coordinate contains all pixels after aggregation. The number of points is the number of pixels in the first updated coordinate and a new color feature value is the color feature value of the first updated coordinate (usually the average of the color feature values of these pixels); based on the first updated coordinate, continue to perform the same color deviation distance calculation and aggregation process on its adjacent coordinate points; continuously iterate this process until the stop condition is met, wherein the stop condition may be that no new adjacent coordinate points meet the color deviation distance threshold, or the preset maximum number of iterations is reached; when the color deviation distance between any two neighborhoods is greater than the color deviation distance threshold, it means that the colors of these neighborhoods are no longer similar and can be regarded as different color areas; at this time, the j-th layer digital pattern color partitioning result will be output, and the j-th layer digital pattern color partitioning result is usually a group of color areas, each area contains a group of pixels with similar colors, as well as the number of these pixels and a color feature value representing the color of the area.
[0020] Further, when the color deviation distance between the first adjacent coordinate color feature value and the first coordinate color feature value is less than or equal to a color deviation distance threshold, aggregating the first adjacent coordinate and the first coordinate to generate a first updated coordinate includes: The number of the first updated coordinate pixel points is the sum of the number of the first adjacent coordinate pixel points and the number of the first coordinate pixel points; The first updating coordinate color feature value calculation step comprises: A first fitting weight is obtained by comparing the number of pixels in the first adjacent coordinates with the number of pixels in the first updated coordinates; a second fitting weight is obtained by comparing the number of pixels in the first coordinates with the number of pixels in the first updated coordinates; and a weighted mean of the first adjacent coordinate color feature value and the first coordinate color feature value is calculated based on the first fitting weight and the second fitting weight to generate the first updated coordinate color feature value.
[0021] The number of pixels in the first update coordinate is the sum of the number of pixels in the first adjacent coordinate and the number of pixels in the first coordinate. That is, when the first adjacent coordinate is aggregated with the first coordinate, the new coordinate (i.e., the first update coordinate) will contain all the pixels of the two coordinates. The ratio of the number of pixels in the first adjacent coordinate to the number of pixels in the first update coordinate is used as the first fitting weight to indicate the relative importance of the pixels in the first adjacent coordinate in the first update coordinate; the ratio of the number of pixels in the first coordinate to the number of pixels in the first update coordinate is used as the second fitting weight to indicate the relative importance of the pixels in the first coordinate in the first update coordinate; since the number of pixels in the first update coordinate is the sum of the number of pixels in the first adjacent coordinate and the number of pixels in the first coordinate, the sum of the first fitting weight and the second fitting weight should be 1. According to the calculated first fitting weight and second fitting weight, the weighted mean of the color feature value of the first adjacent coordinate and the color feature value of the first coordinate is calculated to generate the color feature value of the first update coordinate.
[0022] A pattern skipping stitch analysis is performed based on the first sub-pattern, the second sub-pattern, and up to the Mth sub-pattern to generate a pattern skipping stitch point sequence.
[0023] In embroidery production, pattern skipping analysis is used to determine the positions that need to be skipped or specially processed during the embroidery process to avoid possible thread breakage, color mixing or other quality problems. Pattern skipping analysis is performed based on the first sub-pattern, the second sub-pattern, and the Mth sub-pattern to generate a pattern skipping point sequence, which reflects the positional relationship of each skipping point in the overall pattern.
[0024] Further, performing pattern skipping analysis based on the first sub-pattern, the second sub-pattern, and up to the Mth sub-pattern to generate a pattern skipping point sequence includes: Configuring a pattern skip stitch constraint distance, wherein the pattern skip stitch constraint distance represents a maximum distance between skip stitches of patterns; The first sub-pattern has a first skipped stitch constraint area, the first sub-pattern has a second skipped stitch constraint area, until the M-th sub-pattern has an M-th skipped stitch constraint area, and the skipped stitch constraint area is a preset start embroidery area for each pattern; According to the skip stitch constraint distance, in combination with the first skip stitch constraint area, the second skip stitch constraint area up to the Mth skip stitch constraint area, the first sub-pattern, the second sub-pattern up to the Mth sub-pattern are randomly allocated between patterns to generate the pattern skip stitch point sequence.
[0025] Preferably, based on factors such as the performance of the embroidery machine, the toughness of the wire, the material of the embroidery, and the aesthetics of the design, a pattern skipping stitch constraint distance is set to limit the maximum distance when jumping from one sub-pattern to another sub-pattern during the embroidery process. Each sub-pattern has its preset starting embroidery area, which is set as a skipping stitch constraint area. This means that during the embroidery process, when the embroidery machine jumps from the current sub-pattern to another sub-pattern, it must first reach the skipping stitch constraint area of the sub-pattern, and then start embroidering from this area, thereby ensuring the accuracy and continuity of the embroidery. For the first sub-pattern, there is a first skipping stitch constraint area; for the second sub-pattern, there is a second skipping stitch constraint area; and so on, until the Mth skipping stitch constraint area of the Mth sub-pattern. After setting the skipping stitch constraint distance and the skipping stitch constraint areas of each sub-pattern, the random allocation of the skipping stitch positions between patterns is performed. By randomly allocating the skipping stitch positions between patterns, a pattern skipping stitch point sequence can be finally generated. The pattern skipping stitch point sequence records the coordinates of each skipping stitch point, the sub-pattern to which it belongs, and the specific position in the sub-pattern.
[0026] The M layers of digital pattern color partition results are traversed to perform layer skipping stitch analysis to generate M layer skipping stitch point sequences.
[0027] The color partitioning results of several layers of digital patterns of each sub-pattern (a total of M) are traversed, and the skipping needle position is determined for each layer to generate M layers of skipping needle point sequences.
[0028] Furthermore, the M layers of digital pattern color partition results are traversed to perform layer skipping analysis to generate M layer skipping point sequences, including: A layer skip stitch constraint distance is configured, wherein the layer skip stitch constraint distance represents the maximum distance of skip stitches between layers; a first plurality of layers of digital pattern color partition results are extracted according to the M plurality of layers of digital pattern color partition results; the first plurality of layers of digital pattern color partition results have layer embroidery sequence labels and plurality of layers of skip stitch constraint areas; based on the layer skip stitch constraint distance, the first plurality of layers of digital pattern color partition results are randomly assigned skip stitch positions in combination with the layer embroidery sequence labels and the plurality of layers of skip stitch constraint areas, to generate a first layer of skip stitch point sequence, which is added to the M layer of skip stitch point sequences.
[0029] Preferably, based on the performance of the embroidery machine, the toughness of the wire, the material and design requirements of the embroidery, a layer skipping stitch constraint distance is set to limit the maximum distance when jumping from one color partition layer to another color partition layer during the embroidery process; the first several layers of digital pattern color partition results are extracted from the M several layers of digital pattern color partition results, and the layer embroidery sequence label of each layer is obtained for the extracted first several layers of digital pattern color partition results. The layer embroidery sequence label indicates the order of embroidery, such as the bottom layer needs to be embroidered first, and the second layer needs to be embroidered after the bottom layer is embroidered; at the same time, the skipping stitch constraint area of each layer is identified and defined. These areas are the preset positions that must be reached first when skipping stitches between layers to ensure the accuracy and continuity of the embroidery. Based on the layer skipping stitch constraint distance, combined with the layer embroidery sequence label and the layer skipping stitch constraint area, the first several layers of digital pattern color partition results are randomly assigned the skipping stitch positions. The random assignment of the skipping stitch positions should be performed on the premise of satisfying the layer skipping stitch constraint distance to ensure the rationality of the skipping stitch path. At the same time, the allocation of skip stitch positions also needs to consider the layer embroidery order label to ensure the correct order of embroidery. When allocating skip stitch positions, the layer skip stitch constraint area also needs to be considered to ensure that each skip stitch starts from the skip stitch constraint area of the current layer and ends after reaching the skip stitch constraint area of the next layer. According to the randomly assigned skip stitch positions, the first layer of skip stitch point sequence is generated. The first layer of skip stitch point sequence records the coordinates of each skip stitch point, the number of layers it belongs to, the layer embroidery order label, and other related information. The generated first layer of skip stitch point sequence is added to the M layer of skip stitch point sequences.
[0030] Based on the pattern skipped stitch point sequence and the M layer skipped stitch point sequences, an embroidery path analysis is performed on the color partitioning results of the M layers of digital patterns to generate a digital pattern embroidery path.
[0031] According to the pattern skipped stitch point sequence and the M layer skipped stitch point sequence, the embroidery path analysis is performed on the M layers of digital pattern color partition results to generate a digital pattern embroidery path in order to improve embroidery efficiency and quality, reduce scrap rate and reduce costs.
[0032] Furthermore, based on the pattern skipped stitch point sequence and the M layer skipped stitch point sequences, an embroidery path analysis is performed on the color partitioning results of the M layers of digital patterns to generate a digital pattern embroidery path, including: Extracting the first several layers of digital pattern color partition results according to the M several layers of digital pattern color partition results; extracting the first layer of digital pattern color partition results of the first several layers of digital pattern color partition results, wherein the first layer of digital pattern color partition results has a plurality of partitioned skipped stitch constraint areas; randomly distributing the first layer of digital pattern embroidery paths according to the plurality of partitioned skipped stitch constraint areas, the pattern skipped stitch point sequence and the M layer skipped stitch point sequences; The first layer of digital pattern embroidery path is added into the first several layers of digital pattern embroidery paths; and the first several layers of digital pattern embroidery paths are added into the digital pattern embroidery paths.
[0033] From the M digital pattern color partitioning results of several layers, select and extract the first digital pattern color partitioning results of several layers; select the first digital pattern color partitioning results of several layers in the first digital pattern color partitioning results; identify multiple partition skipped stitch constraint areas in the first digital pattern color partitioning results; combine the pattern skipped stitch point sequence and the M layer skipped stitch point sequence to randomly distribute the embroidery path of the first digital pattern color partitioning results to generate the first digital pattern embroidery path, which contains the specific embroidery sequence, path and skipped stitch point information of the embroidery machine in each color partition. If the first digital pattern of several layers contains multiple layers, the above process needs to be repeated to generate a corresponding embroidery path for each layer. Then, the embroidery paths of these layers are combined into a complete first digital pattern embroidery path of several layers, and the first digital pattern embroidery path of several layers is added as a whole to the entire digital pattern embroidery path.
[0034] Based on the evaluation factor set, a process evaluation function is constructed, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set at least includes the number of long skipped stitches, the number of skipped stitches, the total length of the skipped stitches, and the total length of the embroidery path.
[0035] The evaluation factor set includes at least the number of long skipped stitches, the number of skipped stitches, the total length of skipped stitches, and the total length of the embroidery path. The number of long skipped stitches refers to the number of long-distance skipped stitches during the embroidery process, which usually increases the running time and possible errors of the machine; the number of skipped stitches refers to the total number of skipped stitches during the embroidery process, including long skipped stitches and short skipped stitches; the total length of skipped stitches refers to the total length of all skipped stitches during the embroidery process. The longer this length is, the farther the machine needs to move; the total length of the embroidery path refers to the total path length required to complete the entire embroidery task. According to the evaluation factor set, a process evaluation function is constructed to evaluate different embroidery process schemes, and the scheme with the smallest evaluation function value is selected as the optimal scheme.
[0036] Furthermore, based on the evaluation factor set, a process evaluation function is constructed, including: The process evaluation function includes: ; when When it is a negative evaluation factor, ; when When is a positive evaluation factor, ; in, The process quality evaluation value representing the h-th group of pattern skipped stitch point sequences, M-layer skipped stitch point sequences and digital pattern embroidery path is referred to as the h-th solution. The negative eigenvalue representing the eigenvalue of the z-th dimension evaluation factor of the h-th solution, Characterizes the eigenvalue of the z-th evaluation factor of the h-th solution, The minimum eigenvalue of the z-th dimension evaluation factor representing all solutions, The maximum eigenvalue of the z-th dimension evaluation factor of all solutions, represents the weight of the evaluation factor of the zth dimension, and Y represents the total dimension of the evaluation factor.
[0037] is the process quality evaluation value of the hth group of pattern skipped stitch point sequence, M layers of skipped stitch point sequence and digital pattern embroidery path. When it is a negative evaluation factor, based on To standardize, The negative eigenvalue representing the eigenvalue of the z-th dimension evaluation factor of the h-th solution. The larger the negative eigenvalue, the smaller the quality evaluation value. Characterizes the eigenvalue of the z-th evaluation factor of the h-th solution, The minimum eigenvalue of the z-th dimension evaluation factor representing all solutions, The maximum eigenvalue of the z-th dimension evaluation factor of all solutions, represents the weight of the evaluation factor of the zth dimension, and Y represents the total dimension of the evaluation factor. When it is a positive evaluation factor, based on The standardized evaluation factor Multiply by the corresponding weight , and then add all the weighted evaluation factors to get the final evaluation function value .
[0038] The pattern skipped stitch point sequence, the M layer skipped stitch point sequences and the digital pattern embroidery path are optimized according to the process evaluation function to generate a recommended pattern skipped stitch point sequence, M recommended layer skipped stitch point sequences and a recommended digital pattern embroidery path.
[0039] According to the process evaluation function, the pattern skipping stitch point sequence, M layer skipping stitch point sequences and digital pattern embroidery path are optimized, that is, the process evaluation function is used to evaluate the current pattern skipping stitch point sequence, layer skipping stitch point sequence and embroidery path, and the process quality evaluation value is calculated. A suitable optimization algorithm, such as genetic algorithm, particle swarm optimization algorithm, simulated annealing algorithm, etc., is selected to search for the optimal skipping stitch point sequence and embroidery path. According to the selected optimization algorithm, the current solution (i.e., skipping stitch point sequence and embroidery path) is iteratively optimized. In each iteration, the process evaluation function is used to evaluate the new solution, and the solution with a better evaluation value is retained as a candidate solution. Several iterations are performed. When the termination condition is met, the optimization process is stopped, such as reaching the maximum number of iterations, the evaluation value is no longer improved for multiple consecutive iterations, etc. After the optimization process is completed, the solution with the smallest evaluation value is selected as the recommended solution, which includes the recommended pattern skipping stitch point sequence, M recommended layer skipping stitch point sequences and the recommended digital pattern embroidery path.
[0040] The production process is initialized according to the recommended pattern skip stitch point sequence, the M recommended layer skip stitch point sequences and the recommended digital pattern embroidery path.
[0041] According to the recommended pattern skip stitch point sequence, M recommended layer skip stitch point sequences and recommended digital pattern embroidery path, the parameters of the embroidery machine are set in the initialization stage of the production process to ensure that the production process can perform efficient and accurate embroidery production according to the recommended pattern skip stitch point sequence, M recommended layer skip stitch point sequences and recommended digital pattern embroidery path.
[0042] In summary, the embodiments of the present application have at least the following technical effects: First, an embroidery sample digital pattern is obtained, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has several layers of digital patterns. Next, the first sub-pattern, the second sub-pattern to the M-th sub-pattern are traversed, and the several layers of digital patterns are color segmented to generate M several layers of digital pattern color partition results. Then, pattern skipping analysis is performed based on the first sub-pattern, the second sub-pattern to the M-th sub-pattern to generate a pattern skipping point sequence. Layer skipping analysis is performed by traversing M several layers of digital pattern color partition results to generate M layer skipping point sequences. Next, based on the pattern skipping point sequence and the M layer skipping point sequence, the M several layers of digital pattern color partition results are analyzed for embroidery paths to generate digital pattern embroidery paths. Based on the evaluation factor set, a process evaluation function is constructed, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set at least includes the number of long skipping stitches, the number of skipping stitches, the total length of skipping stitches, and the total length of the embroidery path. Furthermore, the pattern skipping stitch point sequence, M layer skipping stitch point sequences and digital pattern embroidery paths are optimized according to the process evaluation function to generate the recommended pattern skipping stitch point sequence, M recommended layer skipping stitch point sequences and recommended digital pattern embroidery paths. Finally, the production process is initialized according to the recommended pattern skipping stitch point sequence, M recommended layer skipping stitch point sequences and recommended digital pattern embroidery paths. The technical problem of the poor degree of automation caused by the subjective setting of skipping stitch parameters in the traditional embroidery production process is solved. By adaptively setting the skipping stitch parameters, the technical effect of improving the degree of automation is achieved. Embodiment 2
[0043] Based on the same inventive concept as the production process optimization method for embroidery in the aforementioned embodiment, Figure 2 As shown, the present application provides a production process optimization device for embroidery, and the device and method embodiments in the present application are based on the same inventive concept. The device includes: A pattern acquisition module 11, the pattern acquisition module 11 is used to obtain an embroidery sample digital pattern, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has a plurality of layers of digital patterns; a color segmentation module 12, the color segmentation module 12 is used to traverse the first sub-pattern, the second sub-pattern to the M-th sub-pattern, perform color segmentation on the plurality of layers of digital patterns, and generate M plurality of layers of digital pattern color partition results; a first skipped stitch analysis module 13, the first skipped stitch analysis module 13 is used to perform pattern skipped stitch analysis based on the first sub-pattern, the second sub-pattern to the M-th sub-pattern, and generate a pattern skipped stitch point sequence; a second skipped stitch analysis module 14, the second skipped stitch analysis module 14 is used to traverse the M plurality of layers of digital pattern color partition results to perform layer skipped stitch analysis, and generate M layer skipped stitch point sequences; a path analysis module 15, the path analysis module The path analysis module 15 is used to perform embroidery path analysis on the color partitioning results of the M layers of digital patterns based on the pattern skip stitch point sequence and the M layer skip stitch point sequences, and generate a digital pattern embroidery path; the function construction module 16 is used to construct a process evaluation function based on an evaluation factor set, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set at least includes the number of long skip stitches, the number of skip stitches, the total length of skip stitches, and the total length of the embroidery path; the optimization module 17 is used to optimize the pattern skip stitch point sequence, the M layer skip stitch point sequences, and the digital pattern embroidery path according to the process evaluation function, and generate a recommended pattern skip stitch point sequence, M recommended layer skip stitch point sequences, and a recommended digital pattern embroidery path; the process control module 18 is used to initialize the production process according to the recommended pattern skip stitch point sequence, the M recommended layer skip stitch point sequences, and the recommended digital pattern embroidery path.
[0044] Furthermore, the color segmentation module 12 is used to perform the following method: Obtain the j-th layer digital pattern of the i-th sub-pattern, wherein i∈[1,M], j∈[1,N], i is an integer, j is an integer, and N is the total number of embroidery layers of the i-th sub-pattern; perform neighborhood color aggregation on the j-th layer digital pattern to generate a j-th layer color partition result; add the j-th layer color partition result to the i-th several layers digital pattern color partition result; add the i-th several layers digital pattern color partition result to the M several layers digital pattern color partition results.
[0045] Furthermore, the color segmentation module 12 is used to perform the following method: Randomly extract the color feature value of the first coordinate from the j-th layer of digital patterns; randomly extract the color feature value of the first adjacent coordinate of the first coordinate from the j-th layer of digital patterns; when the color deviation distance between the first adjacent coordinate color feature value and the first coordinate color feature value is less than or equal to the color deviation distance threshold, aggregate the first adjacent coordinate and the first coordinate to generate a first updated coordinate, wherein the first updated coordinate has the number of first updated coordinate pixels and the first updated coordinate color feature value; based on the first updated coordinate pixel number and the first updated coordinate color feature value, cyclic analysis is performed, and when the color deviation distance between any two neighborhoods is greater than the color deviation distance threshold, output the j-th layer color partition result.
[0046] Furthermore, the color segmentation module 12 is used to perform the following method: The number of pixels in the first updated coordinate is the sum of the number of pixels in the first adjacent coordinate and the number of pixels in the first coordinate; the step of calculating the color feature value of the first updated coordinate includes: obtaining a first fitting weight by comparing the number of pixels in the first adjacent coordinate with the number of pixels in the first updated coordinate; obtaining a second fitting weight by comparing the number of pixels in the first coordinate with the number of pixels in the first updated coordinate; and performing a weighted mean calculation on the color feature value of the first adjacent coordinate and the color feature value of the first coordinate according to the first fitting weight and the second fitting weight to generate the color feature value of the first updated coordinate.
[0047] Furthermore, the first needle skipping analysis module 13 is used to execute the following method: A pattern skipping stitch constraint distance is configured, wherein the pattern skipping stitch constraint distance represents the maximum distance of skipping stitches between patterns; wherein the first sub-pattern has a first skipping stitch constraint area, the first sub-pattern has a second skipping stitch constraint area, until the M-th sub-pattern has an M-th skipping stitch constraint area, and the skipping stitch constraint area is a preset start embroidery area for each pattern; according to the skipping stitch constraint distance, in combination with the first skipping stitch constraint area, the second skipping stitch constraint area until the M-th skipping stitch constraint area, the first sub-pattern, the second sub-pattern until the M-th sub-pattern are randomly assigned skipping stitch positions between patterns to generate the pattern skipping stitch point sequence.
[0048] Furthermore, the second needle skipping analysis module 14 is used to execute the following method: A layer skip stitch constraint distance is configured, wherein the layer skip stitch constraint distance represents the maximum distance of skip stitches between layers; a first plurality of layers of digital pattern color partition results are extracted according to the M plurality of layers of digital pattern color partition results; the first plurality of layers of digital pattern color partition results have layer embroidery sequence labels and plurality of layers of skip stitch constraint areas; based on the layer skip stitch constraint distance, the first plurality of layers of digital pattern color partition results are randomly assigned skip stitch positions in combination with the layer embroidery sequence labels and the plurality of layers of skip stitch constraint areas, to generate a first layer of skip stitch point sequence, which is added to the M layer of skip stitch point sequences.
[0049] Furthermore, the path analysis module 15 is used to perform the following method: According to the M layers of digital pattern color partitioning results, extract the first several layers of digital pattern color partitioning results; for the first layer of digital pattern color partitioning results of the first several layers of digital pattern color partitioning results, wherein the first layer of digital pattern color partitioning results has multiple partitioned skipping stitch constraint areas; according to the multiple partitioned skipping stitch constraint areas, the pattern skipping stitch point sequence and the M layer skipping stitch point sequences, randomly distribute the first layer of digital pattern embroidery path; add the first layer of digital pattern embroidery path into the first several layers of digital pattern embroidery path; add the first several layers of digital pattern embroidery path into the digital pattern embroidery path.
[0050] Furthermore, the function construction module 16 is used to execute the following method: The process evaluation function includes: ; when When it is a negative evaluation factor, ; when When is a positive evaluation factor, ; in, The process quality evaluation value representing the h-th group of pattern skipped stitch point sequences, M-layer skipped stitch point sequences and digital pattern embroidery path is referred to as the h-th solution. The negative eigenvalue representing the eigenvalue of the z-th dimension evaluation factor of the h-th solution, Characterizes the eigenvalue of the z-th evaluation factor of the h-th solution, The minimum eigenvalue of the z-th dimension evaluation factor representing all solutions, The maximum eigenvalue of the z-th dimension evaluation factor of all solutions, represents the weight of the evaluation factor of the zth dimension, and Y represents the total dimension of the evaluation factor.
[0051] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0052] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0053] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A method for optimizing the production process of embroidery, characterized in that: The method comprises: Obtaining an embroidery sample digital pattern, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has a plurality of layers of digital patterns; Traversing the first sub-pattern, the second sub-pattern until the Mth sub-pattern, performing color segmentation on the multiple layers of digital patterns, and generating M multiple layers of digital pattern color partitioning results; Performing pattern skipping analysis based on the first sub-pattern, the second sub-pattern, and up to the Mth sub-pattern to generate a pattern skipping point sequence; Traversing the color partitioning results of the M layers of digital patterns to perform layer skipping analysis, and generating M layer skipping point sequences; Based on the pattern skipped stitch point sequence and the M layer skipped stitch point sequences, an embroidery path analysis is performed on the color partitioning results of the M layers of digital patterns to generate a digital pattern embroidery path; Based on the evaluation factor set, a process evaluation function is constructed, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set at least includes the number of long skipped stitches, the number of skipped stitches, the total length of the skipped stitches, and the total length of the embroidery path; Optimizing the pattern skipped stitch point sequence, the M layer skipped stitch point sequences and the digital pattern embroidery path according to the process evaluation function to generate a recommended pattern skipped stitch point sequence, M recommended layer skipped stitch point sequences and a recommended digital pattern embroidery path; The production process is initialized according to the recommended pattern skip stitch point sequence, the M recommended layer skip stitch point sequences and the recommended digital pattern embroidery path.
2. The method according to claim 1, characterized in that Traversing the first sub-pattern, the second sub-pattern, and finally the Mth sub-pattern, performing color segmentation on the multiple layers of digital patterns, and generating M multiple layers of digital pattern color partitioning results, including: Obtain the j-th layer digital pattern of the i-th sub-pattern, where i∈[1,M], j∈[1,N], i is an integer, j is an integer, and N is the total number of embroidery layers of the i-th sub-pattern; Performing neighborhood color aggregation on the j-th layer of digital patterns to generate a j-th layer of color partitioning results; Adding the j-th layer color partitioning result to the i-th layer digital pattern color partitioning result; The i-th digital pattern color partition result of several layers is added to the M digital pattern color partition results of several layers.
3. The method according to claim 2, characterized in that Performing neighborhood color aggregation on the j-th layer of digital patterns to generate a j-th layer of color partitioning results includes: Randomly extracting a first coordinate color feature value from the j-th layer of digital patterns; Randomly extracting a first adjacent coordinate color feature value of a first coordinate from the j-th layer of digital patterns; When the color deviation distance between the first adjacent coordinate color feature value and the first coordinate color feature value is less than or equal to a color deviation distance threshold, aggregating the first adjacent coordinate and the first coordinate to generate a first updated coordinate, wherein the first updated coordinate has a first updated coordinate pixel number and a first updated coordinate color feature value; Based on the cyclic analysis of the number of pixels of the first update coordinates and the color feature values of the first update coordinates, when the color deviation distance between any two neighborhoods is greater than the color deviation distance threshold, the j-th layer color partition result is output.
4. The method according to claim 3, characterized in that When the color deviation distance between the first adjacent coordinate color feature value and the first coordinate color feature value is less than or equal to a color deviation distance threshold, aggregating the first adjacent coordinate and the first coordinate to generate a first updated coordinate, including: The number of the first updated coordinate pixel points is the sum of the number of the first adjacent coordinate pixel points and the number of the first coordinate pixel points; The first updating coordinate color feature value calculation step comprises: The first fitting weight is obtained by comparing the number of the first adjacent coordinate pixel points with the number of the first updated coordinate pixel points; Comparing the number of pixel points at the first coordinate with the number of pixel points at the first update coordinate to obtain a second fitting weight; A weighted mean calculation is performed on the first adjacent coordinate color feature value and the first coordinate color feature value according to the first fitting weight and the second fitting weight to generate the first updated coordinate color feature value.
5. The method according to claim 1, characterized in that Performing pattern skipping analysis based on the first sub-pattern, the second sub-pattern, and up to the Mth sub-pattern to generate a pattern skipping point sequence, including: Configuring a pattern skip stitch constraint distance, wherein the pattern skip stitch constraint distance represents a maximum distance between skip stitches of patterns; The first sub-pattern has a first skipped stitch constraint area, the first sub-pattern has a second skipped stitch constraint area, until the Mth sub-pattern has an Mth skipped stitch constraint area, and the skipped stitch constraint area is a preset start embroidery area for each pattern; According to the skip stitch constraint distance, in combination with the first skip stitch constraint area, the second skip stitch constraint area up to the Mth skip stitch constraint area, the first sub-pattern, the second sub-pattern up to the Mth sub-pattern are randomly allocated between the patterns to generate the pattern skip stitch point sequence.
6. The method according to claim 1, characterized in that Traversing the M layers of digital pattern color partition results to perform layer skipping analysis, generating M layer skipping point sequences, including: Configure a layer jump stitch constraint distance, wherein the layer jump stitch constraint distance represents the maximum distance of jump stitches between layers; Extracting the first plurality of layers of digital pattern color partition results according to the M plurality of layers of digital pattern color partition results; The first several layers of digital pattern color partition results have layer embroidery order labels and several layers of skipped stitch constraint areas; Based on the layer skip stitch constraint distance, in combination with the layer embroidery order label and the several layers of skip stitch constraint areas, the skip stitch positions of the first several layers of digital pattern color partition results are randomly allocated to generate a first layer of skip stitch point sequence, which is added to the M layer skip stitch point sequences.
7. The method according to claim 1, characterized in that Based on the pattern skipped stitch point sequence and the M layer skipped stitch point sequences, an embroidery path analysis is performed on the color partitioning results of the M layers of digital patterns to generate a digital pattern embroidery path, including: Extracting the first plurality of digital pattern color partition results according to the M plurality of digital pattern color partition results; A first layer of digital pattern color partition results for the first plurality of layers of digital pattern color partition results, wherein the first layer of digital pattern color partition results has a plurality of partition skip stitch constraint areas; Randomly distribute the first layer of digital pattern embroidery paths according to the multiple partitioned skip stitch constraint areas, the pattern skip stitch point sequence and the M layer skip stitch point sequences; Adding the first layer of digital pattern embroidery paths into the first plurality of layers of digital pattern embroidery paths; The first several layers of digital pattern embroidery paths are added into the digital pattern embroidery path.
8. The method according to claim 1, characterized in that Based on the evaluation factor set, a process evaluation function is constructed, including: The process evaluation function includes: ； when When it is a negative evaluation factor, ; when When is a positive evaluation factor, ; in, The process quality evaluation value representing the h-th group of pattern skipped stitch point sequences, M-layer skipped stitch point sequences and digital pattern embroidery path is referred to as the h-th solution. The negative eigenvalue representing the eigenvalue of the z-th dimension evaluation factor of the h-th solution, Characterizes the eigenvalue of the z-th evaluation factor of the h-th solution, The minimum eigenvalue of the z-th dimension evaluation factor representing all solutions, The maximum eigenvalue of the z-th dimension evaluation factor of all solutions, represents the weight of the evaluation factor of the zth dimension, and Y represents the total dimension of the evaluation factor.
9. A production process optimization device for embroidery, characterized in that: Used to implement the production process optimization method for embroidery according to any one of claims 1 to 8, the device comprises: A pattern acquisition module, wherein the pattern acquisition module is used to obtain an embroidery sample digital pattern, wherein the embroidery sample digital pattern has a first sub-pattern, a second sub-pattern to an M-th sub-pattern, and any one of the first sub-pattern, the second sub-pattern to the M-th sub-pattern has a plurality of layers of digital patterns; A color segmentation module, the color segmentation module is used to traverse the first sub-pattern, the second sub-pattern to the Mth sub-pattern, perform color segmentation on the multiple layers of digital patterns, and generate M multiple layers of digital pattern color partition results; A first skip stitch analysis module, the first skip stitch analysis module is used to perform pattern skip stitch analysis based on the first sub-pattern, the second sub-pattern to the Mth sub-pattern to generate a pattern skip stitch point sequence; A second skipped stitch analysis module, the second skipped stitch analysis module is used to traverse the color partition results of the M layers of digital patterns to perform layer skipped stitch analysis and generate M layer skipped stitch point sequences; A path analysis module, the path analysis module is used to perform embroidery path analysis on the color partition results of the M layers of digital patterns based on the pattern skipped stitch point sequence and the M layer skipped stitch point sequences, and generate a digital pattern embroidery path; A function construction module, wherein the function construction module is used to construct a process evaluation function based on an evaluation factor set, wherein the process evaluation function is a minimization optimization function, and the evaluation factor set at least includes the number of long skipped stitches, the number of skipped stitches, the total length of the skipped stitches, and the total length of the embroidery path; An optimization module, the optimization module is used to optimize the pattern skipped stitch point sequence, the M layer skipped stitch point sequences and the digital pattern embroidery path according to the process evaluation function, and generate a recommended pattern skipped stitch point sequence, M recommended layer skipped stitch point sequences and a recommended digital pattern embroidery path; A process control module is used to initialize the production process according to the recommended pattern skipping stitch point sequence, the M recommended layer skipping stitch point sequences and the recommended digital pattern embroidery path.