CN113298081B - Image data processing method and system in Hunan embroidery plate making process - Google Patents
Image data processing method and system in Hunan embroidery plate making process Download PDFInfo
- Publication number
- CN113298081B CN113298081B CN202110841208.6A CN202110841208A CN113298081B CN 113298081 B CN113298081 B CN 113298081B CN 202110841208 A CN202110841208 A CN 202110841208A CN 113298081 B CN113298081 B CN 113298081B
- Authority
- CN
- China
- Prior art keywords
- image
- stitch
- stitches
- embroidery
- scaling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Sewing Machines And Sewing (AREA)
Abstract
The invention relates to the technical field of image processing, and discloses an image data processing method and system in a Hunan embroidery plate making process, which aim to improve plate making efficiency and resource reuse rate. The method comprises the following steps: scanning to obtain a hand embroidery pattern scanning image with a ratio of 1:1 corresponding to the manual hand embroidery pattern, dividing image areas with different colors into different layers according to RGB values of the hand embroidery pattern scanning image, and dividing the image areas with the same color into the same layer; carrying out image identification processing on each image layer, and identifying stitches, connecting lines and texture features of the image layer in the identification process of any image layer; mapping the zoomed corresponding number of stitch points to a border outline trend curve with the zoomed size, and reconstructing connecting lines among the zoomed stitch points; then, cutting the connecting line exceeding the maximum distance range of the embroidery machine; and finally, overlapping the zoomed layers to obtain a plate making data file for the embroidery machine.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an image data processing method and system in a Hunan embroidery plate making process.
Background
Hand embroidery refers to an art that is added to any existing fabric by hand, by needle and thread, and is one of the most precious species in the protection of Chinese non-material cultural heritage.
The types of manual embroidery which is left in China at present are not many, and the manual embroidery is mainly divided into Suzhou embroidery, Guangdong embroidery, Hunan embroidery, Shu embroidery and the like.
Hunan embroidery is one of the four famous embroideries in China, is a general name of Hunan embroidery products with bright Hunan cultural characteristics centering on Changsha in Hunan, originates from folk embroidery in Hunan, develops by absorbing the advantages of Su embroidery and Guangdong embroidery, and has a history of over 2000 years. The Hunan embroidery is mainly embroidered by silk, pure silk, hard satin, soft satin, transparent yarn, silk threads and woolen yarns with various colors, has strict composition and bright color, is compatible with expressive force by various stitches, enables the embroidered characters, animals, mountains and birds and the like to have special artistic effects by abundant color threads and changeable stitches, and fully exerts the expressive force of the stitches no matter plain embroidery, woven embroidery, net embroidery, knot embroidery, seed embroidery, velvet cutting embroidery, disordered stitch embroidery and the like.
The hand-imitating embroidery is machine embroidery with hand-made needle method and needs plate making during the preparation process. At present, most of plate making of imitation handmade embroidery based on Hunan embroidery is manually traced in a plate making software (such as Tajima DG/ML by plus, Wilckm software and the like) by adopting a manual stitch method one by referring to a handmade embroidery sample; the overall plate making efficiency is very slow and the plate making of a simple image also requires more than a worship for an operator skilled in operating the software. Moreover, the reuse rate of the finally formed plate making file is low, and one size needs to correspond to an exclusive plate making file; for example: the same image corresponding to a square work needs a special platemaking file, and the same image corresponding to four squares, nine squares … … and the like needs new platemaking files corresponding to one respectively; the main reasons for this result are: the thicknesses of the silk threads used in different sizes are fixed, and the sizes of the silk threads cannot be synchronously scaled along with the scaling of the presented sizes; and even if synchronous scaling is realized, the finished product of silk threads with different thicknesses selected based on the same platemaking file has far less effect than the expected scaling effect. Therefore, further improvements are awaited.
Disclosure of Invention
The invention aims to disclose an image data processing method and system in a Hunan embroidery plate making process, so as to improve the plate making efficiency and the resource reuse rate.
In order to achieve the purpose, the invention discloses an image data processing method in a Hunan embroidery plate making process, which comprises the following steps:
setting characteristic points in a non-pattern area of the artificial hand embroidery pattern;
scanning the pattern area of the artificial hand embroidery pattern and each characteristic point;
according to the mapping relation between each characteristic point in a scanned image and the characteristic points arranged in the non-pattern area of the substantial manual hand embroidery pattern, carrying out integral rotation and scaling processing on the scanned image to obtain a hand embroidery pattern scanning image with a ratio of 1:1 corresponding to the manual hand embroidery pattern, and carrying out whitening processing on the image area of each characteristic point by using the hand embroidery pattern scanning image;
dividing image areas of different colors into different layers according to the RGB values of the hand embroidery pattern scanning image, and dividing image areas of the same color into the same layer;
carrying out image recognition processing on each image layer, recognizing the connection lines between the stitches of the image layer and the stitches in the recognition process of any image layer, automatically sequencing the recognized stitches according to the progressive relation of the connection lines between the stitches, and then obtaining the texture features of the image layer after responding to the correction processing of a user on the recognition result, wherein the texture features comprise: the mapping relation between the continuous stitch sequencing range and the stitch method for representing the texture trend of each area;
storing the mapping relation between the hand embroidery pattern scanning images corresponding to different manual hand embroidery patterns in a ratio of 1:1 and the texture characteristics between the corresponding image layers in a database, and calibrating the size data of the actual line corresponding to each manual hand embroidery pattern;
acquiring a plate making request of a user, wherein the plate making request carries a hand embroidery pattern scanning image with a ratio of 1:1 and a designated scaling ratio, which are selected by the user from the database;
scaling the size of the corresponding boundary contour trend curve of each layer according to the scaling, wherein the grid size data of each layer is unchanged in the scaling process;
determining the number of the stitches of each subsection on the boundary contour of each area in each layer according to the scaling, the size of the scaled contour trend curve, the number of the stitches in each subsection corresponding to the corresponding 1:1 ratio manual embroidery pattern and the size data of the actual line type, mapping the corresponding number of the stitches after scaling to the boundary contour trend curve after scaling the size, and reconstructing connecting lines among the various stitches after scaling according to the mapping relation between the continuous stitch sequencing range and the stitches of the corresponding 1:1 ratio manual embroidery pattern;
traversing whether the distance between two stitch points of the zoomed connecting line exceeds the maximum distance range of the embroidery machine or not, if so, automatically cutting the connecting line exceeding the maximum distance range of the embroidery machine according to a preset cutting rule, or prompting a user to manually cut and then acquiring a manual cutting result of the user; the size data of the connecting line is consistent with the size data of the actual line type corresponding to each of the 1:1 ratio artificial hand embroidery patterns calibrated in the database;
and superposing the scaled layers, and generating a plate making data file which is scaled according to the proportion after the correction processing of a response user on the superposition result, so that the embroidery machine can carry out corresponding processing.
Preferably: in the process of carrying out image identification processing on each image layer, carrying out amplification processing on a local or whole area of the image layer; therefore, the wiring between the stitches can be clearly identified.
Preferably, the method of the present invention further comprises: respectively setting corresponding constraint conditions for at least two stitches of the artificial hand embroidery pattern; and in the process of carrying out image identification processing on each image layer, automatically extracting the mapping relation between the continuous stitch sequencing range representing the texture trend of each region and the stitch according to the constraint condition of each stitch.
Preferably, before extracting the mapping relationship between the continuous stitch sequencing range and the stitches for representing the texture trend of each region, the method further comprises: constructing a texture image dataset, wherein the same texture image only presents one stitch; training a classification model capable of identifying at least one method according to the image data set; judging whether any region of each image layer belongs to the same type of stitch or not through the classification model, if so, omitting the step of screening corresponding stitches according to stitch sequencing ranges, and directly marking the mapping relation between all stitch sequencing in the region and the corresponding stitches according to the judgment result of the classification model; thus, the plate making efficiency can be further improved.
Preferably, in the process of mapping the zoomed corresponding number of stitch points to the size-zoomed boundary contour trend curve, based on the stitch points in the same segment of the before-zoom and after-zoom analogy, the distance between the stitch point and the adjacent stitch point in the small-size image is consistent with the distance between the stitch point and the adjacent stitch point in the large-size image; and in the process of reconstructing the connecting lines between the zoomed stitch points according to the mapping relation between the continuous stitch sequencing range and the stitches, the ratio of the area ranges covered by the adjacent units of different stitches after reconstruction in the same analog region is consistent with that before reconstruction. Therefore, the rendering effect of the zoomed image size is closer to the traditional manual embroidery effect corresponding to the size, namely the imitation is more vivid.
Preferably, in the image recognition processing process of each image layer, the invention obtains the boundary contour trend curve of each region through fitting, and in the process of mapping the scaled corresponding number of stitches to the boundary contour trend curve with the scaled size, the stitches corresponding to the same boundary contour segment are mapped to the outside, the inside or the top of the scaled boundary contour trend curve corresponding to the boundary contour segment randomly based on the set scaled variance constraint condition and the random algorithm. Similarly, the rendering effect of the scaled image size can be closer to the embroidery effect corresponding to the size manually based on the operation, namely, the rendering effect can be simulated more vividly.
In order to achieve the above object, the present invention further discloses an image data processing system in a Hunan embroidery plate making process, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method when executing the computer program.
The invention has the following beneficial effects:
according to the invention, the image scanning, image recognition and image processing technology is combined with the traditional manual Hunan embroidery, so that the hand embroidery pattern scanning images corresponding to different manual embroidery patterns in the database in the ratio of 1:1 and the texture feature data between the corresponding image layers can be multiplexed to the image presenting sizes in other scales, and the resource multiplexing rate is greatly improved; and in the processing process, the number of the stitches of each section on the boundary contour of each area in each layer is determined according to the scaling, the size of the scaled contour trend curve, the number of the stitches in each section corresponding to the artificial hand embroidery pattern with the corresponding 1:1 ratio and the size data of the actual line type, so that the adverse effect on scaling caused by unchanged line type size is solved. Meanwhile, the system and the processing method fully consider the user interactivity, and reserve flexible space for the user to further modify the automatic image identification and data processing effect. And the whole process is simple, effective and convenient to realize, and the plate making efficiency is greatly improved.
The present invention will be described in further detail below with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flow chart of an image data processing method in a Hunan embroidery plate making process disclosed in the embodiment of the present invention.
Detailed Description
The embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways as defined and covered by the claims.
Example 1
The embodiment discloses an image data processing method in a Hunan embroidery plate making process, as shown in FIG. 1, comprising:
and step S1, scanning to obtain a hand embroidery pattern scanning image with a ratio of 1:1 corresponding to the manual hand embroidery pattern.
In this embodiment, the step may be further subdivided into:
and step S11, setting characteristic points in the non-pattern area of the artificial hand embroidery pattern. Wherein the feature points may be four vertices including the artificial hand embroidery pattern and forming a rectangle corresponding to the X-axis and the Y-axis. Each feature point can be marked by a convenient-to-position pasted marker or a regular pattern for hand-drawing auxiliary positioning.
And step S12, scanning the pattern area and each characteristic point of the manual hand embroidery pattern.
Step S13, according to the mapping relationship between each feature point in the scanned image and the feature point set in the non-pattern area of the actual manual embroidery pattern, performing overall rotation and scaling processing on the scanned image to obtain a manual embroidery pattern scanning image with a ratio of 1:1 corresponding to the manual embroidery pattern, and performing whitening processing on the image area of each feature point by using the manual embroidery pattern scanning image.
In the step, the scanning image is subjected to integral rotation and scaling processing according to the mapping relation between the angle and distance parameters between the characteristic points in the scanning image and the angle and distance parameters between the characteristic points set in the non-pattern area of the calibrated actual manual embroidery pattern, so as to obtain a 1:1 ratio manual embroidery pattern scanning image corresponding to the manual embroidery pattern. Preferably, the resulting scanned image is generally a normal rectangular box image.
In this step, the "white-out processing" is to set the RGB (i.e., colors representing three channels of red, green, and blue) values of the region corresponding to the "positioned affixed marker or the hand-drawn positioning-assisted regular pattern" in step S11 to 255, so that the entire effect is displayed in white, and the image region of the positioning-assisted non-hand embroidery is essentially removed from the scanned image in step S12.
And step S2, dividing the image areas with different colors into different layers according to the RGB values of the hand embroidery pattern scanning image, and dividing the image areas with the same color into the same layer.
Step S3 is to perform image recognition processing for each layer.
The method specifically comprises the following steps: in the process of identifying any layer, identifying the stitches (i.e. the needle hole marks punctured by the embroidery needles, which are not described later) of the layer and the connection lines between the stitches, automatically sequencing the identified stitches according to the progressive relation of the connection lines between the stitches, and then obtaining the texture features of the layer after the modification processing of the identification result by the response user, wherein the texture features comprise: the mapping relation between the continuous stitch sequencing range and the stitch method for representing the texture trend of each area.
In this step, the boundary contour trend curves of the regions are adjacent to each other according to the needle track points on the boundary of the regions, and the formed curves are calculated through fitting. Taking a leaf as an example, the leaf can be divided into an upper half part and a lower half part, or a left half part and a right half part, and the wiring of the needle method also corresponds to the contour curve, so that the contour curve of the leaf can be divided into an upper section and a lower section or a left section and a right section; for another example, a pentagon may segment, i.e., specifically divide, each side into 10 segments. In the invention, the segmentation of the contour curve can be automatically judged by calling a delineation plug-in of a third party or according to the distance between stitches in a certain range of the sequence, and when necessary, each segment of the contour curve can be traced in a manual mode. It is worth mentioning that: in the present embodiment, for a complex hand embroidery work, for example: a small pattern is formed inside the large pattern in a covering mode, and the large pattern and the small pattern can be split into two hand embroidery patterns respectively in a splitting mode to execute the series of steps of the embodiment; firstly, making a zoomed large pattern imitating hand embroidery by using a plate making file of the large pattern, and then carrying out scaling on the small pattern in the same proportion to obtain the plate making file; and finally, aligning the hand embroidery prevention big pattern and embroidering the machine embroidery part of the small pattern by an embroidery machine.
Generally, the same stitch has repeatability and continuity and covers continuous area of the region, so that the main texture feature of the region is extracted by extracting the mapping relation between the stitch sequencing range and the stitch. Preferably: in the process of carrying out image identification processing on each image layer, carrying out amplification processing on a local or whole area of the image layer; therefore, the wiring between the stitches can be clearly identified.
Preferably, the method of this embodiment further includes: setting corresponding constraint conditions (usually in the form of an equation set, generally including the length relation, the angle relation and the like between adjacent series stitch connecting lines) aiming at least two stitches of the artificial hand embroidery pattern; and in the process of carrying out image identification processing on each image layer, automatically extracting the mapping relation between the continuous stitch sequencing range representing the texture trend of each area and the stitch according to the constraint condition of each stitch.
Preferably, before extracting the mapping relationship between the continuous stitch sequencing range and the stitches for representing the texture trend of each region, the step further includes: constructing a texture image dataset, wherein the same texture image only presents one stitch; training according to the image data set to obtain a classification model capable of identifying at least one method; judging whether any region of each image layer belongs to the same type of stitch or not through a classification model, if so, omitting the step of screening corresponding stitches according to stitch sequencing ranges, and directly marking the mapping relation between all stitch sequencing in the region and the corresponding stitches according to the judgment result of the classification model; thus, the plate making efficiency can be further improved. Alternatively, the specific recognition mechanism of the classification model can be implemented by using a convolutional neural network algorithm.
Further, in this embodiment, the stitches available for the image recognition processing at least include at least one of the following three stitches with more xiang embroidery feature: blending needles, disorderly needles and hair needles. In contrast, in the existing partial plate making software, most of the preset fillable stitches are only used for machine embroidery general stitches, such as cross-stitch stitches and plain stitches, and other unpredicted stitches are uniformly presented and compensated by hand embroidery stitches. Therefore, the method of the embodiment is more convenient for customizing and developing the characteristic stitch of the Hunan embroidery, and improves the universality of the system.
And step S4, storing the mapping relation between the hand embroidery pattern scanning images corresponding to different manual hand embroidery patterns in the ratio of 1:1 and the texture features between the corresponding image layers in a database, and calibrating the size data of the actual line corresponding to each manual hand embroidery pattern.
And step S5, obtaining a plate making request of the user, wherein the plate making request carries the hand embroidery pattern scanning image with the ratio of 1:1 and the appointed scaling ratio selected by the user from the database.
And step S6, mapping the corresponding number of the stitch points after the zooming to the boundary contour trend curve after the size zooming, and reconstructing connecting lines among the stitch points after the zooming.
This step can be subdivided into:
and step S61, carrying out scaling processing on the size of the corresponding boundary trend curve of each layer according to the scaling, wherein the grid size data of each layer is unchanged in the scaling processing process. The implementation of the scaling in the step is similar to the storage of the 1 inch face certificate photo image as the 2 inch certificate photo image, and the description is omitted.
Step S62, determining the number of stitches of each segment on the boundary contour of each area in each layer according to the scaling, the size of the scaled contour trend curve, the number of stitches in each segment corresponding to the corresponding 1: 1-scaled artificial hand embroidery pattern and the size data of the actual line type, mapping the scaled corresponding number of stitches to the size-scaled boundary contour trend curve, and reconstructing the connecting lines between the scaled stitches according to the mapping relation between the continuous stitch sequencing range and the stitch method of the corresponding 1: 1-scaled artificial hand embroidery pattern.
Preferably, in this step, in the process of mapping a corresponding number of stitch points after scaling into the boundary contour trend curve after size scaling, based on the stitch points in the same segment of the analogy before and after scaling, the distance between the stitch point and the adjacent stitch point in the small-size image is consistent with the distance between the stitch point and the adjacent stitch point in the large-size image; in the process of reconstructing the connecting lines between the zoomed stitch points according to the mapping relation between the continuous stitch sequencing range and the stitches, the ratio of the area ranges covered by the adjacent units of different stitches after reconstruction in the same analog region is consistent with that before reconstruction. Therefore, the rendering effect of the zoomed image size is closer to the traditional manual embroidery effect corresponding to the size, namely the imitation is more vivid. In the step, the mapping of the corresponding number of the trace points and the reconstruction of the trace connecting lines after the scaling can form a corresponding equation set through the superposition of series constraint conditions for simultaneous solution, so that the texture of each reconstructed part is consistent with the stitch of the original stitch range mapped, the number of the corresponding traces of the same stitch is basically scaled according to the proportion, and the number of the trace points on each contour trend curve of the corresponding proportion is also scaled approximately according to the proportion; when the number of the zoomed stitches is not enough to complete a certain stitch in multiple, the stitches are added to fill up or the surplus part which is not enough to complete a stitch is brought into the coverage range of the next adjacent stitch; generally, as one of the constraints, the stitches rendered in the scanned image should be rendered at least once in the zoomed image; for another example, in the mapping process of the corresponding number of trace points after zooming, the contour trend curve after zooming is subjected to finer segmentation in a successive dichotomy mode, and the number of the trace points of each fine segmentation is determined; preferably, the aforementioned ways of filling up stitches and incorporating a small number of extra stitches into the next adjacent stitch are alternately spaced so that the final rendering effect is closer to the theoretically scaled effect. Thus, in a non-ideal situation, when the above-mentioned "stitch filling and a way of incorporating a small amount of extra stitches into the next adjacent stitch" exist, the above-mentioned "the ratio of the area ranges covered by the adjacent cells of different stitches after reconstruction in the same analog region is consistent with that before reconstruction" means "substantially consistent", and usually, the error between the two is considered as "substantially consistent" within 5%; such variations fall within the scope of the present invention and are not described in detail in the following; similarly, the aforementioned "the pitch between the stitch point and the adjacent stitch point in the small-size image coincides with the pitch between the stitch point and the adjacent stitch point in the large-size image" also includes "substantial coincidence" in which the error is within a certain range (e.g., 5%).
Preferably, the result after the solution based on the above constraint conditions and the equation system can be presented on the human-computer interaction interface for further revision by the user. Further, the magnification scale of the present embodiment is preferably increased in 50% steps, such as: 1.5, 2.0, 2.5 … …; while the reduction scale is preferably decreased in 25% steps, such as: 75%, 50% and 25%. The optimized scaling can reduce the complexity of data processing in the processes of stitch mapping and connecting line reconstruction to a certain extent and make the final effect presented after scaling more ideal.
Based on the above technical means, the average length of the connecting lines between stitches in any stitch region after reconstruction is also scaled by approximately the same proportion as the average length of the connecting lines between stitches in an analogous stitch region (only the connecting lines before segmentation in the subsequent step S7) in the hand-embroidery pattern scan of the 1:1 proportion.
Further, in the step, in the process of image recognition processing of each image layer, a boundary contour trend curve of each region is obtained through fitting, and in the process of mapping the zoomed corresponding number of stitches to the boundary contour trend curve with the zoomed size, the stitches corresponding to the same boundary contour segment are mapped to the outside, the inside or the top of the zoomed boundary contour trend curve corresponding to the boundary contour segment randomly based on the set zoomed variance constraint condition and the random algorithm. Similarly, the rendering effect of the scaled image size can be closer to the embroidery effect corresponding to the size manually based on the operation, namely, the rendering effect can be simulated more vividly.
And step S7, dividing the connecting line exceeding the maximum distance range of the embroidery machine.
The method specifically comprises the following steps: traversing whether the distance between two stitch points of the zoomed connecting line exceeds the maximum distance range of the embroidery machine or not, if so, automatically cutting the connecting line exceeding the maximum distance range of the embroidery machine according to a preset cutting rule, or prompting a user to manually cut and then acquiring a manual cutting result of the user; the dimension data of the connecting line is consistent with the dimension data of the actual line type corresponding to each 1:1 ratio manual embroidery pattern calibrated in the database. The automatic division function is consistent with the realization of the similar function of the Wilckm software, and is not described in detail.
In this embodiment, based on the fact that the maximum distance range of the embroidery machine is not continuously considered in the manual embroidery process, and in order to match with the implementation of this embodiment, in the actual conventional manual embroidery process, the size of the manual embroidery can be reduced so that the two stitches do not need to be divided, and therefore, it is not necessary to identify whether the divided stitches exist in the step S3; the design difficulty of the system is simplified, and the overall production efficiency of the hand-embroidered product from manual embroidery to batch multi-scaling-ratio large-scale production is ensured. Therefore, the present embodiment can preferably make the manual hand-embroidered pattern with the smallest size, and then magnify the manual hand-embroidered pattern with different corresponding magnification in the machine-embroidering process. In other words, the scaling of the present embodiment may be just scaling. Preferably, in order to fully utilize the labor achievement of the existing large-size manual embroidery, the reduced digital processing capacity is increased; in this case, if there are divided stitch points in the large-size manual embroidery, the stitch points may be manually removed by the user in step S3.
And step S8, overlapping the scaled layers, and generating a scaled plate-making data file for the embroidery machine to perform corresponding processing after responding to the correction processing of the user on the overlapping result.
In this step, optionally, during the process of overlaying the observed effect, the user may fine-tune the gap between the layers to ensure a smooth transition between the layers and an overall zooming effect.
Example 2
Corresponding to the above method embodiments, this embodiment further discloses an image data processing system in the xiangxiu plate-making process, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps corresponding to the above method are implemented.
In summary, the image data processing method and system in the xiangxiu plate-making process disclosed in the embodiments of the present invention have the following beneficial effects:
according to the invention, the image scanning, image recognition and image processing technology is combined with the traditional manual Hunan embroidery, so that the hand embroidery pattern scanning images corresponding to different manual embroidery patterns in the database in the ratio of 1:1 and the texture feature data between the corresponding image layers can be multiplexed to the image presenting sizes in other scales, and the resource multiplexing rate is greatly improved; and in the processing process, the number of the stitches of each section on the boundary contour of each area in each layer is determined according to the scaling, the size of the scaled contour trend curve, the number of the stitches in each section corresponding to the artificial hand embroidery pattern with the corresponding 1:1 ratio and the size data of the actual line type, so that the adverse effect on scaling caused by unchanged line type size is solved. Meanwhile, the system and the processing method fully consider the user interactivity, and reserve flexible space for the user to further modify the automatic image identification and data processing effect. And the whole process is simple, effective and convenient to realize, and the plate making efficiency is greatly improved.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. An image data processing method in a Hunan embroidery plate making process is characterized by comprising the following steps:
setting characteristic points in a non-pattern area of the artificial hand embroidery pattern;
scanning the pattern area of the artificial hand embroidery pattern and each characteristic point;
according to the mapping relation between each characteristic point in a scanned image and the characteristic points arranged in the non-pattern area of the substantial manual hand embroidery pattern, carrying out integral rotation and scaling processing on the scanned image to obtain a hand embroidery pattern scanning image with a ratio of 1:1 corresponding to the manual hand embroidery pattern, and carrying out whitening processing on the image area of each characteristic point by using the hand embroidery pattern scanning image;
dividing image areas of different colors into different layers according to the RGB values of the hand embroidery pattern scanning image, and dividing image areas of the same color into the same layer;
carrying out image recognition processing on each image layer, recognizing the connection lines between the stitches of the image layer and the stitches in the recognition process of any image layer, automatically sequencing the recognized stitches according to the progressive relation of the connection lines between the stitches, and then obtaining the texture features of the image layer after responding to the correction processing of a user on the recognition result, wherein the texture features comprise: the mapping relation between the continuous stitch sequencing range and the stitch method for representing the texture trend of each area;
storing the mapping relation between the hand embroidery pattern scanning images corresponding to different manual hand embroidery patterns in a ratio of 1:1 and the texture characteristics between the corresponding image layers in a database, and calibrating the size data of the actual line corresponding to each manual hand embroidery pattern;
acquiring a plate making request of a user, wherein the plate making request carries a hand embroidery pattern scanning image with a ratio of 1:1 and a designated scaling ratio, which are selected by the user from the database;
scaling the size of the corresponding boundary contour trend curve of each layer according to the scaling, wherein the grid size data of each layer is unchanged in the scaling process;
determining the number of the stitches of each subsection on the boundary contour of each area in each layer according to the scaling, the size of the scaled contour trend curve, the number of the stitches in each subsection corresponding to the corresponding 1:1 ratio manual embroidery pattern and the size data of the actual line type, mapping the corresponding number of the stitches after scaling to the boundary contour trend curve after scaling the size, and reconstructing connecting lines among the various stitches after scaling according to the mapping relation between the continuous stitch sequencing range and the stitches of the corresponding 1:1 ratio manual embroidery pattern;
traversing whether the distance between two stitch points of the zoomed connecting line exceeds the maximum distance range of the embroidery machine or not, if so, automatically cutting the connecting line exceeding the maximum distance range of the embroidery machine according to a preset cutting rule, or prompting a user to manually cut and then acquiring a manual cutting result of the user; the size data of the connecting line is consistent with the size data of the actual line type corresponding to each of the 1:1 ratio artificial hand embroidery patterns calibrated in the database;
and superposing the scaled layers, and generating a plate making data file which is scaled according to the proportion after the correction processing of a response user on the superposition result, so that the embroidery machine can carry out corresponding processing.
2. The method of claim 1, further comprising:
and in the process of carrying out image identification processing on each image layer, carrying out amplification processing on a local or whole area of the image layer.
3. The method of claim 1, further comprising:
respectively setting corresponding constraint conditions for at least two stitches of the artificial hand embroidery pattern; so that:
and in the process of carrying out image identification processing on each image layer, automatically extracting the mapping relation between the continuous stitch sequencing range representing the texture trend of each region and the stitch according to the constraint condition of each stitch.
4. The method of claim 3, before extracting the mapping relationship between the continuous stitch sequencing range and the stitch representing the texture trend of each region, further comprising:
constructing a texture image dataset, wherein the same texture image only presents one stitch;
training a classification model capable of identifying at least one method according to the image data set;
and judging whether any region of each image layer belongs to the same type of stitch or not through the classification model, if so, omitting the step of screening the corresponding stitch according to the stitch sequencing range, and directly marking the mapping relation between all stitch sequencing in the region and the corresponding stitch according to the judgment result of the classification model.
5. The method according to any one of claims 1 to 4, wherein in the process of mapping a corresponding number of stitch points after scaling into the boundary contour trend curve after size scaling, the distance between the stitch point and the adjacent stitch point in the small-size image is identical to the distance between the stitch point and the adjacent stitch point in the large-size image based on the stitch points in the same segment as compared before and after scaling; and in the process of reconstructing the connecting lines between the zoomed stitch points according to the mapping relation between the continuous stitch sequencing range and the stitches, the ratio of the area ranges covered by the adjacent units of different stitches after reconstruction in the same analog region is consistent with that before reconstruction.
6. The method according to claim 5, wherein during the image recognition processing of each image layer, a boundary contour trend curve of each region is obtained by fitting, and during the process of mapping a corresponding number of scaled stitch points into a boundary contour trend curve with scaled size, stitch points corresponding to the same boundary contour segment are randomly mapped outside, inside or above the scaled boundary contour trend curve corresponding to the boundary contour segment based on a set scaled variance constraint condition and a random algorithm.
7. An image data processing system in Hunan embroidery-making process, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any one of claims 1 to 6 when executing the computer program.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110841208.6A CN113298081B (en) | 2021-07-26 | 2021-07-26 | Image data processing method and system in Hunan embroidery plate making process |
| ZA2022/08197A ZA202208197B (en) | 2021-07-26 | 2022-07-22 | Image data processing method and system in process of xiang embroidery plate making |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110841208.6A CN113298081B (en) | 2021-07-26 | 2021-07-26 | Image data processing method and system in Hunan embroidery plate making process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113298081A CN113298081A (en) | 2021-08-24 |
| CN113298081B true CN113298081B (en) | 2021-11-09 |
Family
ID=77330982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110841208.6A Active CN113298081B (en) | 2021-07-26 | 2021-07-26 | Image data processing method and system in Hunan embroidery plate making process |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113298081B (en) |
| ZA (1) | ZA202208197B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116579934B (en) * | 2023-04-06 | 2024-04-16 | 湖南师范大学 | Embroidery plate making processing method and system based on edge detection |
| CN117909534B (en) * | 2024-03-15 | 2024-06-18 | 大连理工大学 | Embroidery pattern optimization processing method and system |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007020644A (en) * | 2005-07-12 | 2007-02-01 | Brother Ind Ltd | Embroidery data processor and program |
| JP2013192579A (en) * | 2012-03-16 | 2013-09-30 | Brother Ind Ltd | Embroidery data creating device, embroidery data creating program and computer-readable medium storing embroidery data creating program |
| CN103294864B (en) * | 2013-05-29 | 2016-01-20 | 南京大学 | A kind of computer-aided crewel embroidery production method adopting genetic algorithm |
| CN103793549B (en) * | 2013-11-04 | 2016-06-22 | 南京大学 | Adopt the computer-aided crewel embroidery production method of fuzzy clustering and random walk |
| CN108170907B (en) * | 2017-12-12 | 2020-06-23 | 南京大学 | Program-passing computer-aided disordered needlework manufacturing method |
| CN111950568A (en) * | 2020-08-20 | 2020-11-17 | 西南科技大学 | Embroidery stitch pattern feature point extraction and matching method |
-
2021
- 2021-07-26 CN CN202110841208.6A patent/CN113298081B/en active Active
-
2022
- 2022-07-22 ZA ZA2022/08197A patent/ZA202208197B/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| CN113298081A (en) | 2021-08-24 |
| ZA202208197B (en) | 2022-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113298081B (en) | Image data processing method and system in Hunan embroidery plate making process | |
| CN100350087C (en) | Making process of color jacquard fabric with different speciality | |
| CN103938371B (en) | The preparation method of clothes solely spent by a kind of imitative plain embroidery picture-weaving in silk | |
| CN1786308A (en) | Digitel code manufacturing method jacquard weave double surface beautiful and glorious | |
| CN103088515B (en) | Method suitable for weaving colored jacquard fabric with large texture thread number | |
| US8655474B2 (en) | Embroidery data generating apparatus, embroidery data generating method, and non-transitory computer-readable medium storing embroidery data generating program | |
| CN107012601B (en) | High-resolution machine embroidery method | |
| CN102181984B (en) | Method for manufacturing pattern card of jacquard fabric with straight stitching effect | |
| CN112131724A (en) | Three-dimensional design simulation system and method for knitted product | |
| CN108984857B (en) | Method and device for generating braiding pattern | |
| CN1217048C (en) | Method for producing embroidery and applique on netting cloth | |
| CN114622322B (en) | System and method for quickly generating brocade makeup satin fabric pattern | |
| US7272462B2 (en) | Process for producing a knitted fabric | |
| US8793009B2 (en) | Data generator, computer readable recording medium, and sewing machine | |
| CN109371555B (en) | Design method and device for manufacturing process of full-forming knitted garment program file | |
| CN106521838A (en) | Computerized embroidery plate-making method | |
| Sterman et al. | A computational design tool for gradual transition of knit structures in seamless circular knitting | |
| CN1480579A (en) | Computerized anglicanum | |
| CN109790669B (en) | Method for preparing a tufting process for tufting a fabric, in particular a carpet, and method for tufting a fabric | |
| CN114897673B (en) | Brocade style image generation method based on layering principle | |
| CN105956279A (en) | Pattern design system for Wangjiang cross stitch | |
| US20130054001A1 (en) | Embroidery method | |
| CN116579934B (en) | Embroidery plate making processing method and system based on edge detection | |
| CN115125740B (en) | Figured satin-like printed fabric and preparation method thereof | |
| JPH06195421A (en) | Textile designing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |