JP2678228B2 - Knit design system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a knit design system, and in particular, conversion of a knit design into knitting data of a knitting machine, simulation by a pseudo image representing a loop of a knit, and fitting of a pseudo image of a knit on a mannequin Regarding the design system that was designed to do. In this specification, a knitted fabric means not only a knitted fabric having a form such as a sweater but also a knitted fabric having no specific form.

[0002]

2. Description of the Related Art The applicant has proposed a system for designing a knitted fabric on a monitor (Japanese Patent Publication No. 3-21661). In this system, data of a knitted fabric in a design process is stored in a frame memory, for example, one pixel is assigned to one loop, and tissue data such as a pattern is stored in a color of a pixel. For example, assuming that the degree of freedom of color is 256, 256 types of data can be expressed for a loop knitting method. The type of the color thread to be used is optionally specified, and specified for each type of organization in the organization data. In this system, the data in the frame memory after the design is completed is converted into data to be knitted by a knitting machine using a reference table or the like.

The inventor plans to improve this system and develop it into a system that satisfies the following problems. 1) Generate a pseudo knitting image from the design data and artificially try on the mannequin. As a result, it is possible to generate a pseudo sample without actually knitting a knitted fabric, which increases flexibility in design and can be applied to high-mix low-volume production. 2) Provide a realistic design image with the loop aspect ratio corrected while using a data format that does not impair the strictness of correspondence with knitting data at the design stage. That is, the aspect ratio of the loop is generally not 1, and if the aspect ratio is not corrected, the design image does not correspond to the actual product. Therefore, the data is corrected so that the data can be accurately converted into knitting data. 3) Generate a pseudo image of the knitted fabric without actually knitting the knitted fabric. This makes it easier to evaluate the design, and if the design is bad, you can immediately go back to design modification. 4) The generation of pseudo images eliminates the need for knitting loops, and enables the generation of pseudo images expressing all loops, even when using various types of loops. 5) Make a pseudo image try on the mannequin so that the design can be evaluated three-dimensionally. As a result, a sample having a pseudo three-dimensional effect can be obtained without actually knitting the knitted fabric. 6) Allow knitting to be designed independently according to the type of knitting method, and increase design flexibility. For example, it is possible to correct only the organization data and not the design data such as intarsia, and to copy, move, and delete only the organization data, or vice versa, copy only the data such as intarsia, Allows you to move, save, and delete, and makes the design data related to the organization independent from other design data. 7) Correct the aspect ratio of the pseudo image of the knitted product to make it an image close to the real thing. 8) Connect the knitting machine to the design system and realize CIM (Computer Integrated Manufacturing). For example, the design system and the knitting machine are integrated, or they are connected by a local area LAN or data of a floppy disk. The design is evaluated with pseudo images on the frame memory, only the main sizes of promising designs are actually knitted on the knitting machine, and other sizes are sampled by loop simulation and pseudo fitting to the mannequin. In this way, the design efficiency and the flexibility thereof are increased, and the production can be immediately started.

[0004]

An object of the present invention is to satisfy the above points. The invention of claim 1 is the above 1) to 5), and the invention of claim 2 is the above 1) to 6). The invention of 3 has the above 1) to 5) and 7),
The invention of claim 4 is to provide a knit design system which satisfies the above 1) to 5) and 8).

[0005]

According to the knit design system of the present invention, a frame memory for storing design data of a knitted fabric in a predetermined number of pixels per loop, and a means for inputting pattern data of the knitted fabric into the frame memory. When,
A means for correcting the design data of the frame memory, a means for correcting the design data of the frame memory according to the aspect ratio of the loop, and a monitor for displaying the design data after the aspect ratio correction,
Automatic control means for converting the design data of the frame memory into knitting data for a knitting machine, and based on the knitting data of the automatic control means, the shape and brightness of each loop from the yarn sample stored in the memory. A loop simulation means for generating a loop image expressing the above, storing each loop image in a frame memory for loop simulation to generate a pseudo image of a knit, and displaying the generated pseudo image on a monitor, and a loop simulation. Mesh mapping means for mesh-mapping the pseudo image generated by the means onto an image representing the mannequin to artificially try on the pseudo image of the knitted fabric on the mannequin.

Preferably, at least two frame memories for knitting design data are provided, design data relating to the structure of the knitting is stored in the first frame memory, and design data for other knitting methods is stored in the second frame memory. It is stored so that the design data of each frame memory can be independently corrected. Further preferably, the loop simulation means is provided with means for correcting the pseudo image stored in the frame memory for loop simulation according to the aspect ratio of the loop. More preferably, a knitting machine is connected to the automatic control means.

Correcting the aspect ratio means correcting the aspect ratio for each loop or correcting the aspect ratio of the entire knitted fabric. The number of pixels per loop for storing the design data may be different for each type of loop. Further, connecting a knitting machine means connecting the knitting machine to the design system, connecting with a LAN or the like, or connecting with data such as a floppy disk. Preferably, the knitting machine is literally integrated with or connected to the LAN. , Etc. through the line,
Allow the knitting machine to be driven by the design system command. As the frame memory in this specification, the same memory may be used both at the stage of design data creation and at the stage of loop simulation. If there are two frame memories, the area of the frame memory allocated to each image is two. It means that there are seeds, not physically two.

[0008]

According to the present invention, the design data is stored in a predetermined number of pixels per loop, and is converted into the knitting data of the knitting machine after the design is completed. Each loop corresponds to a predetermined pixel, and there is no ambiguity in the conversion from design data to knitting machine data. The data in the design process is displayed on the monitor after the loop aspect ratio is corrected, so you can design with an image similar to the actual knitting. The data after the design is converted into knitting data for the knitting machine, so you can move to trial production or production at any time. A loop image is generated based on the created knitting data, and the loop image is created from the thread sample stored in the memory. The loop image is generated without actually knitting the loop. this is,
This is because there are many types of loops that occur in one knitted fabric, considering the color of the yarn, the relationship with the loops of the base, and the type of structure, and it is difficult to simulate these based on the actual loops. Therefore, a loop image is generated from the thread sample in the memory, and the loop image is realistic in which the shape and the brightness of the loop are reflected. The loop image is preferably one in which the overlap with the background loop is taken into consideration. When a pseudo image of a knit is generated from the loop image and displayed on the monitor,
The design can be evaluated without knitting the knit, and the design can be modified if it is not preferable. Moreover, the pseudo image after the loop simulation is mesh-mapped to the mannequin image, and the mannequin is artificially tried on the knitted fabric. For this reason,
The design of the knitted fabric to the three-dimensional evaluation on the mannequin can be easily performed, the efficiency of the design is increased, and the design can be flexibly changed (claim 1).

Preferably, at least two frame memories for knitting design data are provided, design data relating to the knitting structure is stored in the first frame memory, and design data of other knitting methods is stored in the second frame memory. It is stored so that the design data of each frame memory can be independently corrected. As a result, organizational data and other data can be created, copied, moved, deleted, saved, and modified independently of each other, thereby increasing design flexibility and flexibility. For example, in designing an organization, it is not necessary to consider other designs, and the organization such as a pocket can be freely moved, or can be reduced / enlarged, copied, and modified. In addition, the design of the intarsia or the like can be freely changed without considering the organization such as pockets and braids (Claim 2).

Preferably, the pseudo image obtained by the loop simulation is displayed with the loop aspect ratio corrected. As a result, the realism of the pseudo image is increased, and in many cases, the aspect ratio correcting means can also be used as the aspect ratio correcting means in the design data (claim 3).

Preferably, a knitting machine is connected to the automatic control means, a design is roughly evaluated, for example, by fitting a pseudo image on a mannequin, and a real size knitting machine is knitted by the knitting machine, and the others are mannequins. A sample of the try-on image is generated by changing the size and changing the pattern and color. Then, if the knitting machine is driven by the knitting data of the knitting machine, it is possible to flexibly process from design to sample preparation and production in a short time. This is extremely efficient in promoting small-volume production of other varieties.

[0012]

1 to 21 show an embodiment, and FIG. 1 shows a hardware configuration of a knit design system. Figure 2
FIG. 10 shows the knit paint processing at the preceding stage of the loop simulation, FIGS. 11 to 17 show the loop simulation processing of the embodiment, and FIGS. 18 to 20 show the subsequent mesh mapping processing. FIG. 21 shows an overview of the knit design system. Note that the display in FIG. 2 and subsequent figures is obtained by activating the hardware configuration in FIG. 1 by software.

In FIG. 1, 2 is a main bus, 4 is a host CPU, 6 is a main memory, and 8 is a graphic CPU. The input / output includes, for example, a scanner 10, an external storage 12 such as a floppy disk, hard disk, or magneto-optical disk, a digitizer 14, and a stylus 1
6. Use the keyboard 18 or the like. Graphic CPU 8
Is connected to a graphic bus 20, and a group of frame memory 22, work memory 23, affine transformation processing unit 24, and mesh mapping processing unit 26 are connected to the bus 20.
The data in the frame memory 22 is synthesized by the synthesis processing unit 28 and displayed on the graphic monitor 30.

The knitting data in the design process is stored in the frame memory 22 as internal data in which one loop has one pixel, for example. Similarly, polka dots, straight lines, repeat patterns, and the like in the creation process are stored in the work memory 23 in an internal data format. The affine transformation processing unit 24 corrects the internal data according to the loop aspect ratio to obtain monitor data, and stores the correction result in another area of the frame memory 22. The affine transformation is performed using the address (x, y)
At the address (ax + by + c, dx + ey +
f). Here, c and f represent offset terms, and are used for offset correction between the frame memory 22 and the display address of the graphic monitor 30. In the affine transformation, coordinate x is set to ax + by and coordinate y is set to d.
When converted to x + ey, not only expansion and contraction in the horizontal and vertical directions, but also deformation and rotation in an oblique direction can be performed. This means that the aspect ratio can be corrected even when the direction of the loop is oblique to the scanning direction of the frame memory 22 or the graphic monitor 30. However, the minimum necessary conversion is the address (x, y) of the frame memory 22.
Can be converted to an address (ax, by), so that b / a
Is the aspect ratio.

The mesh mapping processing unit 26 is for virtually trying on the data after the loop simulation on the mannequin.

FIG. 2 shows the knit paint processing section. In the figure, 32 is a storage area for intarsia data, 34
Is a storage area for organization data, and 36 is a storage area for jacquard data. These areas are designated and assigned to the frame memory 22, and data is stored as internal data. Here, it is divided into three types: intarsia, organization, and jacquard. However, at least two types may be used. For example, two types of intarsia or jacquard pattern and organization are used. 33
Is an organization information adding unit that generates organization data such as a change in the number of stitches from the pattern data, and stores the intarsia data storage area 3
2 is stored in the organization data storage area 34 in addition to the pattern data stored in the pattern data stored in Step 2.

Reference numerals 38, 40 and 42 denote affine transformation means 2
In the storage area of the monitor data after the loop size is corrected in step 4, the area is allocated to the frame memory 22 and stored. Numeral 38 is a storage area for the corrected intarsia data, 40 is a storage area for the tissue data after the correction, and 42 is a storage area for the jacquard data after the correction. These data are synthesized by the synthesis processing unit 28 or the synthesis processing unit 28
Is bypassed and displayed on the graphic monitor 30 as independent data. The synthesizing processing unit 28 does not need to be able to simultaneously synthesize three images, but only needs to be able to synthesize two images that are highly necessary. The display mode on the monitor 30 is determined by the display mode selection means 44.

Reference numeral 46 is a drawing processor, and 48 is a monitor 30 for input coordinates of external input means such as the stylus 16.
This is a coordinate affine transformation unit for performing coordinate affine transformation to the above coordinates. In this embodiment, the position of the external input is input as internal coordinates corresponding to the internal data, and is converted by the coordinate affine conversion means 48 into monitor coordinates (coordinates for the monitor 30).

Reference numeral 50 is a knitting method input, which uses the keyboard 18 and the external memory 12 to specify, for example, a knitting method (intarsia, jacquard, organization), the number of colored threads, the gauge, the number of stitches, and the like. 52 is a loop size determining means for determining the size of the loop according to the knitting method, the gauge, the number of stitches, and the like. Reference numeral 54 denotes a size input for measuring and numerically inputting data from the keyboard 18 or inputting data from the stylus 16, and inputting paper pattern data stored in the external storage 12 or paper pattern data read from the scanner 10. Reference numeral 56 denotes a pattern data creating means for creating pattern data in accordance with the data from the size input 54, and a pattern contour determining means 58 for determining the pattern contour. Reference numeral 60 denotes an internal data creation unit which inputs the size of the loop from the loop size determination unit 52 and the outline of the pattern from the pattern outline determination unit 58. As a result, one loop generates one pixel of internal data, and stores the internal data in the data storage areas 32, 34, 36 and the like.

To input a pattern or pattern to a knitted fabric, the stylus 16 or pattern drawing data input 62 (existing pattern or pattern stored in the external memory 12, existing pattern or pattern read from the scanner 10), etc. The input position is specified by the digitizer 14 using the stylus 16, is stored as an internal address, and is displayed as a monitor address. The drawing processor 46 processes the input data and writes it to the pixel specified by the internal address. On the other hand, in the display on the graphic monitor 30, the internal address specified by the digitizer 14 is converted into a monitor address by the coordinate affine conversion means 48 and displayed.

The monitor data after the loop size correction is not inversely affine-transformed into internal data in order to prevent the accuracy of the internal data from being deteriorated due to the inverse affine transformation. Took priority.

Reference numeral 64 denotes an automatic control processing means for converting internal data into knitting data of a knitting machine. It is known from Japanese Patent Publication No. 3-21661 that the internal data can be converted into knitting machine data (knitting data) when the tissue data is stored as color information in one pixel and the other data is stored as option information. It is. Therefore, if the three types of data of intarsia, organization, and jacquard can be converted into the data format of Japanese Patent Publication No. Hei 3-21661, the internal data can be converted into knitting machine data. Therefore, the data of the organization is used as it is, and the data of the option line is generated from the data of the intarsia and the jacquard.
Then, this data is processed by the automatic control processing means of Japanese Patent Publication No. 3-21661 and converted into knitting machine data. 6
Numeral 6 is a memory for knitting machine data, and 68 is a floppy disk for storing knitting machine data. Of course, the method of converting to knitting machine data is arbitrary.

FIG. 3 shows the relationship among the pattern data, the internal data, and the monitor data. When the pattern data on the left side of the figure is created by the internal data creating means 60, the data storage areas 32, 34, and 36 store the modified internal data as shown in the center of the figure. This is because one loop corresponds to one pixel in the internal data, and the aspect ratio of the loop is not considered. Next, the internal data is affine-transformed and the aspect ratio of the loop is corrected to obtain monitor data, which is displayed on the monitor 30 as shown on the right side of FIG.
When the cursor position of the stylus 16 is indicated by a cross mark, the cursor position is input by an internal address, converted by the coordinate affine conversion means 48, and displayed on the graphic monitor 30 by the monitor address. As a result, designers can take advantage of
Can design.

FIG. 4 shows an algorithm of loop size correction and coordinate affine transformation of the cursor position designated by the stylus 16. First, the display mode selection means 44 selects which data of intarsia, tissue, or jacquard is to be corrected, the coordinate position of the cursor is designated by the stylus 16 by an internal address, coordinate affine transformation is performed, and the monitor address is used for graphic display It is displayed on the monitor 30. Next, the drawing area is determined by the stylus 16 or the like, and the data in the drawing process is stored as internal data in, for example, the work memory 23, affine-transformed and displayed on the monitor 30. When the designer confirms the drawing, the work memory 2 is stored in the data storage areas 32, 34 and 36 using the drawing processor 46.
Input the internal data of No. 3. Work memory 2 for drawing process
The internal data of No. 3 is converted into monitor data by the affine conversion means 24, transferred to the data storage areas 38, 40 and 42, and displayed on the graphic monitor 30.

FIG. 5 shows the composition of three data, intarsia, organization and jacquard. With the algorithm of FIG.
Since the internal data and the internal address are virtualized and only the monitor data and the monitor address are visible, the monitor data after the loop size correction is shown in FIG. In these figures, there are two types of processing: design processing using internal data and internal addresses, and display processing using monitor data and monitor addresses. Each time the designer checks the processing on the monitor, the internal data is processed. And

The original image of the internal data creating means 60 is, for example, three types of intarsia, organization and jacquard,
Alternatively, two types of data, such as intarsia and organization, are written in the storage areas 32, 34, and 36, and are independently corrected by the drawing processor 46. Therefore, in designing the intarsia pattern, it is not necessary to pay attention to the organization or the jacquard pattern, and only the intarsia pattern needs to be designed. In organizational design, there is no need to pay attention to intarsia or jacquard patterns. As a result, the pattern and the tissue can be designed separately, and the design load is reduced. Intarsia, Jacquard, and organizational data can be copied, moved, deleted, and modified independently.
For example, it is possible to copy only the organization and organizational pattern of the pocket from existing data, and to copy, move, modify, and scale only the pattern of intarsia and jacquard from existing data.

As shown in FIG. 5, the position of the jacquard pattern with respect to the tissue pattern can be confirmed by synthesizing the images of the tissue data storage area 34 and the jacquard data storage area 36. Similarly, the relationship between the intarsia and the organizational pattern can be confirmed by combining the data storage areas 32 and 34. Further, since each image is displayed as monitor data in which the aspect ratio of the loop has been corrected, it is easy to move, copy, reduce and enlarge the pattern or pattern. For example, when moving the jacquard one-point mark on the composite screen of FIG. 5, it is difficult to imagine the actual position of the mark in the knitted fabric with the internal data in which the loop aspect ratio is not corrected. Therefore, it is difficult to move or copy the mark, and it is also difficult to understand the image given by the mark after copying or after moving. On the other hand, in the monitor data in which the aspect ratio has been corrected, the display shape of the monitor 30 is similar to the shape of the actual knitted fabric, and the position of the mark on the actual knitted fabric is displayed.
You can immediately understand the image of the mark. When the correction is completed, the three types of data are converted into one type of data using a look-up table or the like, and the automatic control processing means 64 converts the data into knitting machine knitting data.

FIG. 6 shows the support of polka dots, repeat patterns, and straight lines by the drawing processor 46. In the figure, 70 is a processor body, 72 is a zoom processing unit, 74 is a movement processing unit, 76 is a copy processing unit, 78 is a polka dot processing unit,
80 is a repeat processing unit, and 82 is a linear processing unit. Each of these processing units 72 to 82 performs each job with the help of the processor main body 70, and writes in the designated data storage areas 32, 34, and 36. The job of the zoom processing unit 72 is to reduce or enlarge the input pattern or pattern, and the job of the movement processing unit 74 is to move the input pattern or pattern. The job of the copy processing unit 76 stores the input pattern or pattern in the data storage area 32,
34, 36 to be copied to other parts. These processing units 72, 74, and 76 are well known in an ordinary paint system and can be easily realized. FIG. 10 shows algorithms of polka dot processing, repeat processing, and linear processing.

FIG. 7 shows the processing in the polka dot processing unit 78.
When designing a polka dot pattern, it is troublesome to draw polka dots with the stylus 16 and copy them one by one with the copy processing unit 76. Therefore, as shown on the right side of FIG. 7, circles A, B, C, and D serving as polka dot elements are input, and these polka dots A, B,
A reference point E for moving or copying C and D collectively is input. The straight line between the reference point E and each of the polka dots A to D is a line for indicating that the polka dots A to D are related to the reference point E, and is called a control rod. Polka dots A to D and reference point E
And the like are stored in the work memory 23 as internal data, affine-transformed and displayed, the cursor position is also designated by an internal address, coordinate affine-transformed and displayed. When the position where the polka dot pattern is copied on the monitor 30 is designated as the position of the reference point E, the polka dot pattern is copied based on the internal address of the reference point E. The actual copy is polka dot A,
The reference point E and the control line are not copied only by B, C, and D.

FIG. 8 shows the processing in the repeat processing section 80. C, D, E, and F in the figure are points for designating a repeat area, and circles A and B are intersections between the repeat area and the outline of the knitted fabric. Points A, A2, A3, and A4 are points for designating an area occupied by one unit of the pattern. In this case, the width between points A and A2 is the width of the unit of the pattern. And point A,
In the area designated by A2, A3, and A4, a pattern of one unit is input with the stylus 16 or the like, and the repeat area (C, D,
E, F) are picked up, and the repeat processing unit 80 repeatedly copies the unit pattern into the repeat area. As a result, it is possible to copy repeatedly exactly by inputting one pattern. G is a movement point, and one point in the repeat area is designated with the stylus 16 or the like and clicked. Next, when the stylus 16 is moved to an appropriate position, the pattern is moved to that position, or the pattern is copied to that position. In the repeat processing unit 80,
The data and the cursor position in the creation process are processed by the internal data and the internal address, and the writing to the data storage areas 32, 34, 36 is also performed in the internal data format. On the other hand, the display on the monitor 30 is subjected to the affine transformation, and is processed by the monitor data and the monitor address.

FIG. 9 shows the operation of the straight line processing section 82. Drawing straight lines with the stylus 16 is difficult per se, and it is even more difficult to specify a regular line with the least aliasing. Therefore, two angle adjustment points are specified for the work memory 23 and the like, and a straight line is interpolated between the two points. The data between the angle adjustment points is vector data, and a straight line closest to the vector data, that is, a straight line with the smallest aliasing is generated. The movement of the straight line is the same as the processing of the movement point in the repeat processing unit 80. When one point on the straight line is clicked and moved to a designated point, a new straight line is generated or the straight line is moved to that point. Can be translated. In the linear processing section 82 as well, the processing in the work memory 23 and the data storage areas 32, 34 and 36 is performed by internal data, and the graphic monitor 30 displays affine-transformed monitor data.

11 to 17 show a loop simulation processing section of the embodiment. FIG. 11 shows the architecture, which is realized by adding software to the hardware configuration of FIG. In FIG. 11, reference numeral 90 denotes a loop shape determining means, which analyzes the knitting machine data of the automatic control processing means 64 for each loop, and determines the type and color shape of the color yarn used for each loop, the lightness and darkness of each loop, and the loop of the background. Analyze the degree of overlap with. The analysis of the loop shape is performed one loop at a time along the course, and when the analysis of one course is completed, the next course is analyzed.

Reference numeral 92 denotes a thread sample storage means, which stores about 10 thread samples for one type of color thread when using, for example, 16 types of color threads. The difference between each thread sample is light and dark. The light and dark of the thread sample itself and the surrounding light and dark are changed, and the light and dark of the thread itself (light and dark depending on the difference in the front and back stitches and the position in the loop) is expressed. Enhance or blur edges. I prepared multiple thread samples by changing the lightness and darkness so that the lightness and darkness of each part of the loop could be processed only by cutting out which thread sample.
If the processing time may be slow, use one thread sample for each thread.
You may memorize only one and change the light and dark after cutting out.
As shown in FIG. 14, a line representing the yarn is written on the yarn sample by giving unevenness to the periphery, and a line expressing the twist of the yarn is written. Reference numeral 94 denotes a thread sample cutout mask creating unit which creates a circular mask as shown in FIG. 14, for example, and gradually changes the value of the mask in a transition region around the mask to make a soft mask. The radius of the mask is changed according to the length of the segment cut out from the thread sample. When the segment length is constant, the thread sample can be used as it is, and no mask is required. 96
Is a spline converter, which bends the segment at the bent portion between the base end and the tip of the loop. Although spline transformation may be applied to all the segments, the space between the base and the end of the loop is almost a straight line, and the spline transformation is omitted at this portion. A synthesis processing unit 98 synthesizes each segment of the loop to form one loop sample (an image showing one loop).

Reference numeral 100 denotes a mask forming means for the base loop, which forms a mask in order to express the degree of overlap of a new loop with the loop of one course before. Reference numeral 102 denotes a loop sample memory, which is realized by designating an area in, for example, the work memory 23, temporarily stores a created loop sample, and terminates processing of a sample of a basic loop having a high frequency of appearance. Save later. Reference numeral 46 denotes the drawing processor, and reference numeral 104 denotes a frame memory for a loop simulation image. The frame memory 104 is realized by allocating an area to the frame memory 22 in FIG.
The drawing processor 46 calls the data of the loop one course below from the frame memory 104, and protects the exposed part of the underlying loop (the loop one course before) by masking the underlying loop with a newly written loop sample. , A loop sample is called from the memory 102 and written into the frame memory 104.

The shape of the loop generally does not fit a square grid and the aspect ratio is not unity. Therefore, in order to simulate an actual knit, affine transformation means 2
4 to correct the aspect ratio. The minimum necessary conversion is to convert the image at the address (x, y) in the frame memory 104 into the image at the address (ax, by), and the conversion is not necessarily affine. Reference numeral 106 denotes a frame memory of an image whose loop size has been corrected by the affine transformation unit 24, which is realized by designating an area in a part of the frame memory 22. Then, the data of the frame memory 106 subjected to the affine transformation after the loop simulation is displayed on the graphic monitor 30.

12 to 17 show the processing in the loop simulation. FIG. 12 shows the main algorithm of the processing. When the knitting machine data from the automatic control processing means 64 is analyzed, the center position of the base end of each loop and the center position of the tip of the loop, the type of yarn used, the front or back stitch and other knitting data, and the knitting data of each part of the loop are obtained. Light and darkness etc. are found. Therefore, a loop sample is created accordingly.

13 to 16 show a loop sample preparation process. The loop shape determining means 90 checks whether or not a loop sample has been created.
Call a loop sample from 2 and use it. If the loop sample is not created yet and is a basic loop (a loop without swing), the created loop sample is stored in the memory 102.
When a new loop sample is created, the loop is divided into a plurality of segments in accordance with the curvature and the brightness, and the segments are cut out from the thread sample stored in the thread sample storage means 92. Since the basic loop sample is created only once, most of the created loop samples are loops with swings. Such a loop may have all segments newly created, but for each segment of the tip and the proximal end that is common to the basic loop, copy the loop sample of the basic loop and The processing time can be shortened by creating only the new segment.

As shown in FIG. 14, the yarn sample storage means 92
To change the contrast between the light and dark of the thread itself and the surroundings,
Approximately 10 types of yarn samples are stored for one yarn. Further, in each of the yarn samples, the thickness, color, and the like of the yarn are stored in the vicinity of the yarn. When the length of the segment is determined by the loop shape determining means 90, the diameter of the mask is determined, and a mask for extracting the segment is extracted from the mask creating unit 94. The mask has, for example, a circular shape as shown on the right side of FIG. 14, and has a transition area around the mask, which makes the periphery soft. Assuming that the value of the mask is 1 and the thread sample is cut out as it is, and the value of the mask is 0 and the thread sample is not cut out, the center of the mask is transparent and Z is 1; outside the mask, Z is 0; Z gradually decreases from 1 to 0. When the thread sample is cut out using such a mask, as shown in the lower part of FIG. 14, the segment cuts out the thread sample data as it is at the center, and the thread sample data is cut in the peripheral transition region (broken line in the figure). Will be cut out softly.

There is a curvature at the base and the tip of the loop,
Since the interval between the two is almost a straight line, the proximal end and the distal end are spline approximated and smoothly bent. Then, the respective segments are synthesized by the synthesis processing unit 98 to form one loop sample. FIG. 15 shows this process. Each part of the loop has lightness and darkness depending on the type of the front / back eyes, etc., the front end has a semicircular shape around the center of curvature C1, and the base end has a hyperbolic focal point C2. It is bent to the center. In the example, all the segments were spline-converted, but between the proximal end and the distal end is close to a straight line,
You may perform linear interpolation.

Simply joining each segment results in an edge at the segment-to-segment connection. Therefore, both ends of the segments are made soft by the mask shown in FIG. This process is shown in FIG. At the boundary between the segments S2 and S3, the brightness of each segment decreases smoothly, and by superimposing them, the segments can be connected smoothly.

When the loop sample is generated, the process returns to the main program shown in FIG.
Create a mask for the underlying loop. This mask is shown in FIG.
Shown in The mask of the base loop is obtained by masking and protecting a portion where the base loop is not covered by the new loop and is exposed. As shown on the right side of FIG. 17, the mask value Z is 1, The value is 0 at the portion not to be masked, and a smooth connection is made between them. When the mask is formed, the base loop image data is taken out from the frame memory 104, the loop sample is taken out from the memory 102, and the overlapping is processed by the drawing processor 46 using the mask.
The data is written back to the frame memory 104.

The image data in the frame memory 104 is a detailed simulation of a knitted fabric including its loop, but the aspect ratio is not corrected. Therefore, the aspect ratio of the loop is corrected by the affine transformation means 24, stored in the frame memory 106, and displayed on the monitor 30. In this way, without actually knitting the knitted fabric, and without taking in a sample of the loop with a scanner or the like, various types of loops, including light and dark, contrast with the surroundings, loop shape, overlap condition, etc. And can simulate accurately. Further, in the case where the knitting machine data is data that cannot be knitted, for example, data in which eyes are lost, a loop is simulated based on the knitting machine data. In the loop simulation, since the affine transformation is performed and displayed every time one loop analysis is completed, the number of loops displayed on the monitor 30 increases by one each time the analysis is completed. Therefore, a loop that cannot be easily knitted can be detected from the image on the monitor 30.

18 to 20 show trial fitting of a loop simulation image on a mannequin by mesh mapping. In FIG. 18, reference numeral 110 denotes a processor for mesh mapping, 106 denotes a frame memory of a loop size corrected image, 112 denotes a frame memory of a mannequin shape, 114 denotes a frame memory of a light and dark image of a mannequin, 1
A frame memory 16 represents a mask for the mannequin. Reference numeral 46 denotes the drawing processor, and reference numeral 118 denotes a frame memory for storing a result of the mesh mapping process.

In the mesh mapping, an image (image of the frame memory 106) whose aspect ratio is corrected after the loop simulation is used, and the shape of the mannequin, its brightness and darkness, a mask, etc. are used, and the frame memories 112 and 11 are used.
4, 116. Next, using the stylus 16,
The main points of the mannequin image and the loop simulation image are mapped (0 mark in FIG. 18). When the mapped points are connected by a framework (skeleton), for example, as shown in FIG. 19, square regions S1, S2, S3, S4 in the loop simulation image
Is deformed on the mannequin image as shown on the right side of the figure. Therefore, for each part surrounded by the framework, the loop simulation image and the mannequin image are called and interpolated so as to compensate for the increase or decrease in the number of pixels caused by fitting the knitted fabric to the mannequin. Scanning is performed in order, and an image deformed by the drawing processor 46 is generated. This image represents the value of the loop simulation image as P
Assuming that the value of the mannequin image is P2, the value of the light / dark image of the mannequin is P3, and the value of the mask image of the mannequin is Z, P1 · P3 · Z + (1-Z) · P2, and this value is stored in the frame memory 118. Write and display on monitor 30. And, for example, the frame memory 1
If 18 images are output by a color printer, a sample for presentation can be obtained.

In this way, the designed knitted fabric can be tried on the mannequin without actually knitting the knitted fabric. Also, by reflecting the shape and brightness of the mannequin, a three-dimensional effect can be expressed by fitting a flat knitted fabric to the mannequin, and the knitted fabric can be expanded or contracted accordingly.

FIG. 21 shows an overall image of a knit design system in which the techniques of FIGS. 1 to 20 are integrated. In the figure,
Reference numeral 120 denotes a knit paint processing unit shown in FIG.
Is a loop simulation processing unit shown in FIG. The data storage areas 32, 34, 36 are extracted from the knit paint processing unit 120 and displayed, and the monitor 30
Was omitted.

A new feature in FIG. 21 is that the dyeing machine 124 and the knitting machine 128 are incorporated in the design system. For example, these are integrated with the design system,
Alternatively, they are connected by a local area LAN or the like and driven by a design system command and organization data. Then, a plurality of design systems are provided, a larger number of knitting machines 128 are provided, and each knitting machine is connected to a plurality of design systems and used as a server of the design system. In this way, the knitting machine 128
Are allocated to the mass-production type knitted fabric, the other-kind small-quantity production-type knitted fabric, and the trial knitting, respectively, so that mass production, small-quantity production, and trial knitting can be efficiently performed. Further, the dyeing machine 124 is assigned to the out-of-stock item or the out-of-stock item of the types of yarns determined by the automatic control means 64, and automatically supplies the necessary colored yarns.

In the system shown in FIG. 21, for example, the body shape of a model is measured, pattern data is generated in the pattern outer shape generating means 58, and the knit paint processing section 120 designs a knit. The design result is loop simulation processing unit 12
The evaluation is made in 2, and the mesh mapping processing unit 26 artificially tries on the mannequin to perform a three-dimensional evaluation. If there is a design failure in these processes, the knit paint processing unit 12
It is easy to return to zero. The design result is divided into intarsia, jacquard, and organization and saved. Therefore, it is possible to modify only the necessary parts such as modifying only the design of the organization or only the design of the intarsia, and it is possible to make the library of existing designs easier to use. For example, the design of the organization can be referred to the existing knit A, and the design of the intarsia can be referred to another knit B.

When the design evaluation is completed, the knitting machine 128
Then, only the main size is trial-edited to make a real sample sweater, and other sizes are used in the loop simulation processing unit 122.
The result of the mesh mapping processing unit 26 is used as a sample. In addition, the version of the pattern change, the color change, or the like uses the results of the loop simulation processing unit 122 and the mesh mapping processing unit 26 as samples. As a result, at the same time as the end of the design, the real sample and the pseudo sample are prepared, and when the transition to the production is decided, the necessary number of knitting machines 128 are allocated and produced. Therefore, the knitting inventory and the leading time until the start of production are almost unnecessary.

The system of FIG. 21 makes it possible to produce easy-order sweaters and the like in boutiques and the like. First, the customer's body shape is measured, the pattern shape determining unit 58 selects the existing pattern paper that best fits the body shape, and the knit paint processing unit 120 designs it according to the customer's preference. The loop simulation processing unit 122 checks the design result, checks the image of the knitted product, and confirms that the design is actually knitting.
Next, using the mannequin image according to the body type of the customer, a three-dimensional pseudo sample is created by the mesh mapping processing unit 26, and if good, knitting is performed by the knitting machine 128. As a result, easy-order knitting can be knitted on the spot.

[0051]

According to the present invention, the following effects can be obtained. 1) It is possible to artificially generate an image of a knitted fabric from the design data and artificially try on the mannequin. As a result, a sample can be generated in a pseudo manner without actually knitting the knitted fabric, the flexibility in design is increased, and it is possible to cope with small-lot production of various kinds (claims 1 to 4). 2) It is possible to provide a realistic image with the loop aspect ratio corrected while using a data format that does not impair the strictness of correspondence with the organization data at the design stage.
That is, the aspect ratio of the loop is not generally 1, and an image that does not correspond to the actual object is obtained unless the aspect ratio is corrected.
Therefore, this is corrected and the data is converted into the knitting data accurately (claims 1 to 4). 3) It is possible to generate a pseudo image of a knitted fabric without actually knitting the knitted fabric. This facilitates the evaluation of the design, and if the design is defective, the design can be immediately corrected (claims 1 to 4). 4) When generating a pseudo image, it is not necessary to knit an actual loop, and even when various types of loops are used, it is possible to generate a pseudo image expressing all the loops (claims 1 to 4). 5) A pseudo image is try-on on the mannequin so that the design can be evaluated three-dimensionally (claims 1 to 4). 6) Allow knitting to be designed independently according to the type of knitting method, and increase design flexibility. For example, it is possible to modify only the organizational data and not the design data such as intarsia, and to copy, move, delete and save only the organizational data, or vice versa. Can be copied, moved, saved, and deleted, and the design data relating to the organization and other design data can be made independent (Claim 2). 7) Correct the aspect ratio of the pseudo image of the knitted product to obtain an image close to the actual product (claim 3). 8) Connect the knitting machine to the design system and realize CIM (Computer Integrated Manufacturing). For example, the design system and the knitting machine are integrated, or they are connected by a local area LAN or data of a floppy disk. The design is evaluated with pseudo images on the frame memory, only the main sizes of promising designs are actually knitted on the knitting machine, and other sizes are sampled by loop simulation and pseudo fitting to the mannequin. In this way, the design efficiency and the flexibility thereof are increased, and the production can be immediately started (claim 4).

[Brief description of the drawings]

FIG. 1 is a block diagram of a knit design system according to an embodiment.

FIG. 2 is a detailed block diagram of a main part of the knit paint system of the embodiment.

FIG. 3 is a diagram showing the relationship among pattern data, internal data, and monitor data in knit paint.

FIG. 4 is a flowchart showing coordinate conversion in knit paint.

FIG. 5 is a diagram showing a relationship between three data of intarsia, jacquard, and organizational pattern in knit paint.

FIG. 6 is a diagram showing an internal block of the drawing processor of FIG. 2;

FIG. 7 is a diagram showing a polka dot process in the embodiment.

FIG. 8 is a diagram showing a repeat process in the embodiment.

FIG. 9 is a diagram showing a straight line process in the embodiment.

FIG. 10 is a control flowchart of three processes of a polka dot, a repeat, and a straight line.

FIG. 11 is a block diagram of a loop simulation processing unit.

FIG. 12 is a flowchart showing an algorithm of a loop simulation.

FIG. 13 is a flowchart showing a work generation algorithm.

FIG. 14 is a diagram showing cutout of a yarn sample in a loop simulation.

FIG. 15 is a diagram showing synthesis from a segment to a work.

FIG. 16 is a diagram showing the superposition of segments in the synthesis of a workpiece.

FIG. 17 is a diagram showing a mask for a base loop in a loop simulation;

FIG. 18 is a block diagram of a mesh mapping processing unit.

FIG. 19 is a diagram showing deformation of an image in mesh mapping.

FIG. 20 is a flowchart showing an algorithm of mannequin processing;

FIG. 21 is a block diagram showing the overall configuration of a knit design system according to an embodiment.

[Explanation of symbols]

 2 Main bus 4 Host CPU 6 Main memory 8 Graphic CPU 10 Scanner 12 External storage 14 Digitizer 16 Stylus 18 Keyboard 20 Graphic bus 22 Frame memory 23 Work memory 24 Affine conversion processing unit 26 Mesh mapping processing unit 28 Synthesis processing unit 30 Graphic monitor 32 Intarsia data storage area 33 Organization information adding unit 34 Organization data storage area 36 Jacquard data storage area 38 Intarsia data storage area after correction 40 Tissue data storage area after correction 42 Jacquard data storage area after correction 44 Display mode selection means 46 drawing processor 48 coordinate affine transformation means 50 knitting method input 52 loop size determination means 54 size input 56 paper pattern data creation means 58 paper pattern outer shape determination Determination means 60 internal data creation means 62 pattern drawing data input 64 automatic control processing means 66 memory of knitting machine data 68 floppy disk 70 processor body 72 zoom processing part 74 movement processing part 76 copy processing part 78 polka dot processing part 80 repeat processing part 82 Straight line processing unit 90 Loop shape determination means 92 Thread sample storage means 94 Thread sample cutout mask creation unit 96 Spline conversion unit 98 Synthesis processing unit 100 Base loop mask creation means 102 Loop sample memory 104 Frame memory of loop simulation image 106 Loop size correction image Frame memory 110 mannequin-shaped frame memory 114 mannequin light and dark image frame memory 116 mannequin mask frame memory 118 mapping result file Lame memory 120 Knit paint processing unit 122 Loop simulation processing unit 124 Dyeing machine 128 Knitting machine

Claims (4)

(57) [Claims] 1. A frame memory for storing knitting design data in a predetermined number of pixels per loop, means for inputting knitting pattern data to the frame memory, and design data for the frame memory. A means for correcting, a means for correcting the design data of the frame memory according to the aspect ratio of the loop, a monitor for displaying the design data after the aspect ratio correction, and a design data for the frame memory. , development and automatic control means for converting the knitting data for a knitting machine, based on knitting data of the automatic control means, a loop image representing the shape and brightness of each loop from the yarn sample having stored in the memory Then, each loop image is stored in the frame memory for loop simulation, and a pseudo image of the knitting is generated. For displaying a pseudo image on the monitor, the loop simulation means, a false image which is generated in the loop simulation means,
A knit design system provided with mesh mapping means for performing mesh mapping on an image representing a mannequin and artificially trying on a pseudo image of a knitted fabric on the mannequin. 2. At least two frame memories for knitting design data are provided, design data relating to a knitting structure is stored in a first frame memory, and design data for other knitting methods is stored in a second frame memory. The knit design system according to claim 1, wherein the design data of each frame memory can be independently corrected. 3. The unit according to claim 1, wherein the loop simulation means is provided with means for correcting the pseudo image stored in the frame memory for loop simulation according to the aspect ratio of the loop. Design system. 4. The knit design system according to claim 1, wherein a knitting machine is connected to the automatic control means.