JPH0274284A - Embroidery data forming device for embroidery machine - Google Patents

Abstract

PURPOSE:To easily form embroidery data in a short time by forming the embroidery data from compression data by a means for extracting the feature points of respective graphics and design data for designating the stitch method of a design by a design designating means. CONSTITUTION:At the time of forming the embroidery data of an automatic embroidery machine, the original picture of an embroidery pattern is inputted by a CCD camera 1. A mechanical post-processing (extraction of an outline, compression of data and the like) is completely automatically processed by a computer. The correction of the graphic and the setting of a design which requires a sense of beauty are associated with the designer of the data. The correction and the like of the graphic and the setting of the design are executed by the use of a display device 8. While the graphic is displayed, the correction and the like of the graphics to be required is carried out. The design is set by displaying the design selected from the basic patterns of the design which are previously formed in the outline. In such a way, the embroidery data for stitching an embroidery pattern high in the quality of beauty can be easily formed in a short time.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Field of Application) The present invention relates to an embroidery data creation device using an image input device for an automatic embroidery machine. (compression, etc.) are all processed by computers, while tasks that require aesthetic sensibility can be performed by humans.

(Prior art and problems to be solved by the invention) Conventionally,
There were two methods for creating embroidery data:

The first method is to place the original drawing of the embroidery pattern on a digitizer, plot all seam positions with a stylus pen, taking into account the placement of the seams so that they will look beautiful when sewn, and use this coordinate data. Let me remember it. Then, in synchronization with the vertical movement of the needle, the coordinate data is read out and a stitch is formed. This method requires a lot of time to input data, has poor work efficiency, and requires a certain level of skill on the part of the operator. Furthermore, since the final stitch coordinates themselves are input, data processing is complicated when enlarging, reducing, rotating, etc. the embroidery pattern.

The second method is to place the original drawing of the embroidery pattern on a digitizer, consider the outline of the embroidery pattern as a collection of many straight lines and circular arcs, and visually determine whether it is a straight line or circular arc. Points are input with a stylus pen, and compressed contour line data and a code (hereinafter referred to as a function code) indicating whether linear approximation or circular arc approximation is to be applied are stored. To find the actual stitch coordinates, find the outline of the embroidery pattern from the compressed outline data and the function code,
The internal seams surrounded by the contour line were determined in relation to the contour line. According to this method, the work efficiency of data input has been improved, and data processing such as enlarging, reducing, and rotating embroidery patterns has become relatively easy. However, the inside surrounded by the contour line can only be filled with uniform stitching, for example, a tight line using straight zigzag stitching.For example, when configuring the inside of the contour line included in the contour line with another pattern, etc. I still had to rely on the lt method.

(Means for Solving the Problems and Effects of the Invention) In order to solve the above-mentioned problems, the present invention, when creating embroidery data for an automatic embroidery machine, inputs an original image of an embroidery pattern to an image input device such as a camera or an image scanner. and subsequent processing that can be done mechanically (contour extraction, data compression, etc.)
The data creator is automatically processed by the computer, and the data creator is involved in the modification of shapes and pattern settings that require an aesthetic sense. Using a display device, necessary modifications to the figure are made while displaying the figure, and the pattern is set while displaying a pattern selected from among the basic patterns created in advance within the outline. By adopting this method, you can instantly check the appearance of finished stitching on the display screen before actually embroidering, so you can easily create embroidery data for sewing high-quality embroidery patterns in a short time. It has the effect of being able to.

(Example) The present invention will be explained below with reference to Examples.

- Outline of the embroidery data creation device First, the outline of the embroidery data creation device will be explained with reference to the block diagram shown in FIG. In the figure, in this embodiment, a COD (Charge Coup) is used as an image input device for an original image.
(LED Device) Camera l is used, and linear approximation is used for data compression. The image input by the COD camera 1 is converted into a video signal and amplified by the amplifier 2 to an appropriate signal level.

The amplifier 2 is connected to an A/D converter 3, and the amplified video signal is converted into a digital signal in synchronization with a signal from a timing circuit 4 connected to the central processing unit cPU via a pass line BL and an interface 101. The A/D converter 3, which serves as a converting means, converts the data into digital data of 256 density levels. The digitized video signal is stored in the video memory 5. On the other hand, central processing unit C
The PU executes the following processing according to the processing procedure (program) stored in the control memory ROM. video memory 5
256!1 stored in! The image data of Q+ is sequentially taken into the central processing unit CPU via the pass line BL, and the data string is subjected to differential processing in order to detect a place where the density changes rapidly, that is, an edge portion of the original image. Based on the result, Ill is given as the value of the point to the coordinate that is an inflection point of the edge strength and the edge strength is above a certain value, and is converted into a binary value. Store in rear D. Feature points, which will be described later, are extracted from the contour line formed by this set of point coordinates, and the contour line is approximated by a straight line. The coordinates of this feature point are registered in the compressed data area C of the working memory 20 as compressed contour data. Further, according to instructions from the central processing unit CPU, the linearly approximated contour line data is transferred to the VRAM 7 connected via the CRT controller 6, and displayed on the display device 8 under the control of the CRT controller 6. A tablet 9 is connected via an interface 102 in order to correct or add defective parts of the displayed outline and to specify a pattern to be embroidered within the outline.

Note that the position specified by operating the tablet 9 can be recognized by a cursor on the display device 8. Also, interface ■1
Keyboard 0 is connected via 03. These devices allow workers to visually check the condition of the shape,
Original image data can be transformed into a design suitable for embroidery. The completed design is written into the working memory 20 in accordance with the format as embroidery data based on compressed outline data, pattern data, etc. The embroidery data generated in this way is transferred to the floppy disk controller FDC.
Floppy disk drive F connected via
The data is transferred to the DD and recorded on the floppy disk FD. Furthermore, if it is desired to immediately embroider the created embroidery data, the embroidery machine I2 can be connected to the network via the LAN controller 11 and the embroidery data can be directly transferred to the embroidery machine I2.

The processing procedure for creating embroidery data according to the structure of 1 μm construction work (411 stitches) is recorded as a program in the control memory ROM of the system.The processing procedure will be explained below mainly with reference to the flowchart of FIG. Processing a is a program module (hereinafter referred to as a module) that performs the processing of importing the original image into the video memory 5, starts the timing circuit 4 via the interface r101, and converts the analog video signal to digital. Move the screen to the bridge direction 256
All data is input by converting it with a resolution of 256 dots in the vertical direction and 256 gradations. The processing module is a module that extracts the contour of the 256-gradation data by differential processing and converts it into binary data. Details will be explained with reference to FIG.

Processing C is a module that performs labeling processing.

The binarized data (pixels) are continuous and form one outline or linear figure component. Generally, a plurality of such graphical components are scattered in an image, so to simplify image processing, each component is separated and given a different name (label) to distinguish it. Details will be explained with reference to FIG. Process d corrects when the graphic components cannot be completely separated in process C, and also searches for small graphics and removes them as noise by measuring the length of the contour of the graphic component. Details will be explained with reference to FIG.

Process e extracts feature points in order to approximate each figure component with a straight line, determines the coordinates of the feature points, and compresses the contour data.

Then, processing is performed to reproduce the contour on the display screen based on the compressed data. Details will be explained with reference to FIG.

Through processes a to e, the outline of the original image could be automatically converted into compressed data faithfully to the original image. However, to express this through embroidery using only thread, unnecessary line segments and outlines may be included or may be missing.

In this case, by executing process r and process g, figure components are deleted, line segments are added, and contour lines are divided. These details will be explained with reference to FIGS. 7 and 8.

Processing is a module that specifies the pattern for filling the outline with thread. Create a needle movement method to form the most suitable pattern for the shape, or select from the basic needle movement patterns prepared in advance, and select the sewing pitch and sewing direction (
tilt) using a tablet to improve aesthetic quality. Details will be explained in FIG. 9.

In process I, the above process results are displayed on the display device 8, and the operator repeats the pattern selection operation so as to make it look the most beautiful according to the operator's judgment.

Labels and pattern information of the embroidery data created by process j,
Each piece of contour information data is recorded on a floppy disk according to the format shown in FIG.

If embroidery is to be performed immediately using already created embroidery data, the embroidery machine 12 can be connected to the network via the LAN controller 11 and the embroidery data can be directly transferred.

-Binarization Process Next, the process of extracting the outline of the original image and binarizing it will be explained with reference to FIG. In the same figure, in process b1, a rask scan is performed from the upper left of the original image, and a portion where the density changes rapidly, that is, an edge portion, is detected by performing edge intensity calculation by differential processing. The calculation formula is shown below.

In process b2 and process b3, the inflection point of the edge strength is found and the inflection point is where the strength at that time becomes greater than or equal to the specified value M^(x
, y) and assigns the value Ill to that point by processing b4.
give. These processes are performed for the entire screen according to the judgment in process b5, the results are written into the binary data area D of the working memory 20, and the binary screen is displayed on the display device 8.

-Labeling Process Next, the labeling process will be explained with reference to FIG. In the figure, the binarized screen is scanned from the upper left by processing cl to find the point D(x, y)=1. Through process c3, four points CD(x-1, y), D(x, y-1), D shown in FIG. 12 are created around the current point D(x, y).
Check the label numbers of (x+ 1° y-1) and D(x(, y-1)). If no label number has been attached yet as determined in process C4, a new label number is attached to point D (x, y) in process C6, and the label number and coordinate data of (x, y) are stored in the working memory 20. Register in area L for coordinate data.

Furthermore, when there is already a label number, a determination is made in process C5, and if there are two or more types of label numbers, in process c7, the fact that they are the same is recorded in the combined information area ■ of the working memory 20. put.

Then, in step c9, the oldest label number is assigned to D(XJ). If there is only one type of label number, process C
8, the label number is attached to D(x, y).

The process is continued until the screen ends as determined by the process CLO.

・Connection processing of graphic components and noise removal processing Next, referring mainly to FIG. 5, the connection processing and noise removal processing of graphic components will be explained. When a figure like the one shown in Fig. 13 is labeled by rask scanning from the upper left, two labels appear even though it is originally one figure.
It is divided into types. Therefore, a need arises for modification.

This modification is performed using the combination information recorded in the working memory, and the label numbers are rewritten to unify them to the oldest one.

In FIG. 5, the next coordinate data L is determined by the process di.
If (n, i+1) exists, L(n) is obtained by processing d2.

Find the distance between i) and L(n, i+1). [n indicates the label number, and i indicates the first coordinate data. ] In other words, the first
As shown in Figure 1(a), when the two points are in a vertical or horizontal positional relationship, the distance is Ill, and as shown in Figure 14(b), when the two points are in a diagonal positional relationship, the distance is 'Ill'. /
Find it as "T". The obtained distances are summed in process d3, and the circumference of the contour line or the length of the line segment is calculated. Also,
When the next coordinate L(n, i+1) is no longer available, it is checked in step d4 whether there is a label number for the outline to be connected, and if it exists, the coordinate data of that label number is transferred to the coordinate data L( of the same label number). Add after n, i) and connect the contour lines.

If the length of the outline or line segment of the graphic component is smaller than a certain value as determined in step d6, the graphic component is regarded as noise and the label number is deleted. In other words, it removes noise. These processes (from di to d7) are performed until all labels are completed as determined in process d8.

・Data compression processing method using linear approximation The process of extracting feature points of each figure component and compressing data using linear approximation will be explained mainly with reference to FIG.

In the same figure, the direction (VoVυ (Fig. 15)) at the start point of the contour line is determined by process e1 and used as the reference direction.The process e2 determines the direction (V, V, ) is calculated.The vector value is determined by the rotation angle with respect to the reference direction as shown in Table 1.In other words, the vector value in the same direction as the reference direction is 0, and the vector value when the direction is +45° different from the reference direction. As with +11, the value is +2 at +90° and + at +135°.
3. At -45°! , -2 at -90°, -135
It is set as −3 when the temperature is °. In process e3, each vector value is added, and the number of appearances of the same vector as the reference direction is counted. The feature points of the vector are detected by the determinations in processes e4 and e5, and the starting point V is detected. and the current point V1 are approximated by a straight line connecting the two points. Process e4 has a total vector value of 3
The point before the above is defined as a feature point, and in process e5, the number of occurrences of the same vector as the reference direction is equal to or greater than a predetermined number N, and the period until the next direction change is approximated by a straight line. . An example is shown in FIG. ■ By processing e6.

The coordinate values of and the coordinate values of 1 are registered as compressed data in the compressed data area C of the working memory 20 together with the label number to create a compressed data table. Also, the coordinate value of V is misplaced in Vo and preparations are made for the next reference direction. In process e7, a straight line pattern is drawn on the display device 8 using the compressed data obtained by the method described above.

Based on the determination in process e8, the process (from el to e7) is repeated over the entire circumference or entire length of this figure component. Furthermore, by the judgment of process e9, the process (el
Execute e8) from By performing the above processing,
The original picture in FIG. 10(a) is drawn as a contour line similar to the same outline.

- Deletion Process of Graphic Components Next, with reference to FIG. 7, a process of deleting a portion of the graphic components converted into compressed data through image processing as instructed by the operator's tablet operation will be described.

In process r1, a contour line or line segment to be deleted is designated by a tablet operation. Keyboard 10 in process r2
With the key input from , it is possible to select whether to delete all or part of the specified outline or line segment.

In the case of partial deletion, the range is specified by operating the tablet in process r3. In process f4, the drawing color of the portion designated to be deleted is changed and confirmation of the deletion portion is prompted. In process r5, deletion is executed to delete a part of the compressed data table, or the graphic component is deleted by deleting the label number attached to the graphic component from the registration. An example of executing the deletion process is shown in FIG. 10(e).

- Addition of line segments and division of contour lines Next, with reference to FIG. 8, processing for adding lines that could not be extracted from the original image, dividing contour lines, and processing figures will be explained. In the figure, processing g! Input two points to draw a straight line using the tablet 9, and process g2.
The following three processes (g3, g4, g5
). Process g3 is a process for adding line segments, and new coordinate data is added to the already existing compressed data table of graphical components. Process g4 is a process for dividing the contour line, and a new label number is attached to the new contour line created by the division and registered in the table. Process g5 is the generation of a new figure component, which creates a new label number and displays it on the tablet 9.
Register the data entered. An example of execution of the additional processing is shown in FIG. 10(d).

- Pattern Designation Process Next, with reference to FIG. 9, a needle movement method for filling the outline with embroidery thread, that is, a pattern designation process will be explained. In the figure, when an outline is specified by the tablet 9 in process h1 and a pattern designation function is executed in process h2, the type of pattern (straight line, triple stitch, zigzag, sand stitch, cross stitch, etc.) is displayed in the right corner of the display device 8. The basic patterns (such as stitches, etc.) are displayed as patterns, and the operator selects the most suitable pattern and color from among them using the tablet 9. If an optimal pattern is not available, create and register a basic pattern of stitches that will form the pattern using a tablet. The basic pattern consists of about a few dozen stitches, and the needles for each stitch and their needle drop points are entered on a tablet while looking at the display screen.By repeating this basic pattern, the outline is filled with embroidery. be able to. By processing h3, select the sewing pitch of the pattern on the keyboard I.
0, and input the sewing direction in angle from the keyboard in process h4.

Note that the order in which the patterns are specified is the order in which each graphic component is embroidered, and the compressed data are arranged in this order. However, when each figure is embroidered in this order, each figure is connected by a thread, and other figures often embroider the thread. There is no problem if the embroidery is done with straight, close stitches, but since the embroidery is done by filling in the pattern, the sewn-in crossover threads degrade the aesthetic quality. Therefore, it is necessary to provide a crossing thread to escape such graphic components. This designation is carried out in process h5, and a crossing thread between the previously designated pattern and the current graphic component is determined, and its needle drop point is designated by the tablet and added to the compressed data. When these specifications are completed, the pattern will be drawn within the outline on the display screen, allowing you to check how it looks when it is sewn.

By repeating the above operations, the operator can easily create embroidery data while improving the aesthetic quality of the embroidery pattern.

In the example shown in Fig. 10(e), the instructions are to sew the leaf part with two types of straight lines in different directions, embroidery the leaf vein part with triple stitches, and embroidery the base of the leaf with a zigzag stitch. The branches are also filled with cross-stitch embroidery.

The pattern information (pitch, direction, hand movement data) specified by the operator in this way is stored in the working memory 20 in the embroidery data format shown in FIG. It is registered in the storage area C for compressed data and recorded on the floppy disk FD.

Further, as described above, when embroidering is to be performed immediately using embroidery data that has already been created, the embroidery machine 12 is connected to the network via the LAN controller 11 and the embroidery data is directly transferred.

(Effects) As described above, according to the present invention, when creating embroidery data for an automatic embroidery machine, an original image of an embroidery pattern is captured using a camera or an image.
An aesthetic sense is required when creating data that is input using an image input device such as a scanner, and all subsequent processing that can be performed mechanically (extraction of contour lines, compression of data, etc.) is automatically processed by a computer. The data creator is involved in modifying figures and setting patterns, and furthermore, a display device is used for modifying figures and setting patterns, and necessary corrections to figures are made while displaying the figures. In addition, by adopting a method of setting the pattern while displaying the pattern selected from among the basic patterns created in advance within the outline, you can see how it will look when it is sewn on the screen before actually embroidering. Since the embroidery data can be checked immediately, it is possible to easily create embroidery data for sewing embroidery patterns of high aesthetic quality in a short time.

[Brief explanation of the drawing]

The drawings relate to embodiments of the present invention; FIG. 1 is a block diagram showing the configuration of an embroidery data creation device, FIG. 2 is a flowchart corresponding to a program of processing steps for creating embroidery data, and FIGS. 3 to 9 are 3 is a subroutine flowchart corresponding to each program module in FIG. 2, FIG. 3 corresponds to module b in FIG. 2, FIG. 4 corresponds to module C in FIG. 6 corresponds to module e in FIG. 2, FIG. 7 corresponds to module r in FIG. 2, FIG. 8 corresponds to module g in FIG. FIG. 9 corresponds to module h in FIG. Figure 1θ is a diagram showing the process of image processing, where (a) shows the original image of the embroidery pattern, (b) shows the outline of the original image, and (c)
, (d) shows an example of the modification of the figure by the operator, and (e) shows the created embroidery data displayed on the display device. Figure 11 is a diagram showing the format of embroidery data for one figure component, Figures 12 and 13 are diagrams for explaining labeling processing, and Figures 14 (a) and (b) are directions and distances of unit vectors. Figure 15 shows the relationship between
FIG. 6 is a diagram showing an example of linear approximation for explaining feature point extraction processing, and Table 1 is a table showing the relationship between vector deflection angle and vector value. 1 is an image input device, 3 is a digitizing means, b is a contour extraction means, bl is a differential processing means, e is a means for extracting feature points of each figure, f is a figure component deletion means, g is a contour line , and h is a pattern designation means.

Claims (1)

[Claims] I. An embroidery data creation device for an embroidery machine, comprising: an image input device for converting an original image of an embroidery pattern into a video signal; and converting the video signal into digital values of a plurality of gradations according to density values. and a differential processing means to detect an edge portion where the digital value changes rapidly from the digital value, convert each coordinate into a binary value based on the edge strength of the edge portion, and generate a contour line as a figure component. a contour extraction means for extracting data; and a means for extracting feature points of each figure to approximate the contour line of the figure component by a set of a plurality of straight lines and compress the contour data to create a new figure component. , a figure part deletion means for removing unnecessary parts of the new figure part, a means for adding necessary line segments to the new figure part and dividing the contour line, and an inside of each divided figure part. A pattern specifying means for specifying a needle movement method, that is, a pattern, for filling in the stitches, and specifying the compressed data obtained by the means for extracting feature points of each figure and the needle movement method of the pattern obtained by the pattern specification means. An embroidery data generation device for an embroidery machine that generates embroidery data consisting of pattern data. II. The embroidery according to claim I, wherein the pattern specifying means creates data on a needle movement method in stitches for forming the pattern, and specifies the pitch of the stitch row and the inclination of the stitch row. Machine embroidery data creation device. III. The embroidery data creation device for an embroidery machine according to claim I or claim II, wherein the pattern designation means designates addition of color data for embroidering with threads of multiple colors.