JPH0157185B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a cutting device that automatically cuts clothing fabric.
The present invention relates to an automatic cutting device that corrects the cutting position depending on the position of the pattern and can always cut the pattern at a constant position.

[Conventional technology]

FIG. 6 shows a block diagram of a conventional automatic cutting device. In the figure, 1 is a reversal section that moves the cloth to the cutting place, 2 is a cloth cutting section, 3 is an image input section that inputs the design of the cloth as a digital image, and 4 is the detection of the position of the design from the digital image data. The symbol position detection section 5 is a cutting control section that corrects the cutting position based on the detected symbol position and controls the cutting section 2 so that the symbol is always cut at a constant position.

FIG. 7 is a detailed block diagram of the pattern position detecting section 4 in the conventional automatic cutting apparatus shown in FIG. 1, and the same reference numerals as in FIG. 6 indicate the same parts.
In the figure, 6 is an image storage unit that stores input images, 7 is a characteristic image storage unit that stores image data of characteristic images of designs in advance, and 8 is an image storage unit 6 and a characteristic image storage unit 7. an image correlation calculation unit that calculates the correlation between the images respectively stored in the image correlation calculation unit;
Reference numeral 9 denotes a reference position detection unit that detects the reference position of the pattern by finding the position with the highest correlation based on the calculated correlation value. The automatic cutting device moves the cloth to a cutting place by means of cloth section 1, cuts the cloth by means of cutting section 2, and after the cutting is finished, supplies new cloth to the cutting place again by means of sheet cloth section 1. , and cut by the cutting section 2. This device is capable of automatically cutting a large amount by repeating this process repeatedly.

Due to mechanical errors, the pattern on the cloth moved to the cutting place by the fabric part 1 is not always set at a constant position at the cutting place, resulting in positional deviation. The image input section 3 uses the image data of the cloth pattern to obtain information necessary for the positional deviation in the pattern position detection section 4, and based on this, the cutting control section 5 corrects the cutting position, so that the pattern is constantly adjusted. The cutting section 2 is controlled so that the cutting can be performed at a fixed position.

This pattern position detection section 4 receives characteristic image data of the cloth pattern stored in advance and an image input section 2.
The reference position of the pattern was detected by calculating the correlation value with the input image data. This operation will be explained with reference to FIG. Image data a _ij inputted as a digital image by the image input section 2 is stored in the image storage section 6 . Here, a _ij
represents the density level value of each pixel, and i and j represent the positions of the image in the horizontal and vertical directions, respectively. In addition to this, the characteristic image storage unit 7
The characteristic image b _ij of the design stored in and the correlation value
Find _Rk . The reference position detection unit 9 detects the position with the highest correlation as the reference position. FIGS. 8 to 10 are explanatory diagrams of this operation.

FIG. 8 shows an example of image data a _ij input from the image input section 2 and stored in the image storage section 6. 9th
The figure shows the characteristic image data b _ij stored in the characteristic image storage section 7 and the reference position. In this example, the image correlation calculation unit 8 calculates that j shown in FIG. 9 is from J ₀ to (J ₀ +M).
image a _ij with a width of , and the characteristic image data shown in Fig. 10.
The correlation value R _k between b _ij is calculated using the following equation.

R _k = _N 〓 ⁱ⁼¹ _M 〓 ^j=1 ｜a _ln −b _ij ｜ where l=i+(k-N/2), k≧N/2 m=j+J _p The results for Y are shown in Figure 10. . Reference position detection section 9
Then, find the position k where the correlation of this image is the highest, that is, where R _k is the minimum, and use this as the reference position.

[Problem that the invention seeks to solve]

Conventional automatic cutting devices use the image correlation between the characteristic image data and the input image data to detect the position of the pattern as described above, but the image correlation includes image data in the horizontal and vertical directions. Since the calculation is performed using two-dimensional data, the correlation value may not be calculated correctly in the case of a positional shift that includes rotation. Further, since such rotational deviation was not detected, there was a problem that when rotational deviation occurred when cutting a large cloth width, the cutting position could not be controlled to be constant with respect to the pattern.

This invention was made to solve these problems, and it is possible to detect the position and rotational deviation of the reference stripes, thereby making it possible to automatically cut the pattern so that the cutting position is always constant with respect to the pattern. The purpose is to obtain a cutting device.

[Means to solve the problem]

In the automatic cutting device according to the present invention, the pattern position detecting device includes a first storage device for storing digital image data, a stripe width pattern detection device for detecting a stripe width pattern of the pattern, and a stripe width pattern detecting device for storing the stripe width pattern. between the second storage means for storing in advance the characteristic stripe width pattern of the cloth design, and the respective stripe width patterns stored in the second and third storage means; a means for detecting the position of the reference stripe from the correlation value; a means for detecting the position of the reference stripe in small areas above and below or to the left and right of the detected reference stripe; A means for calculating rotational deviation from the position of is provided.

[Effect]

The automatic cutting device of the present invention detects the stripe width pattern of input image data by the stripe width pattern detection means, and stores the stripe width pattern in the second storage means. The memorized stripe width pattern and
The correlation calculation means calculates the correlation with the characteristic stripe width pattern of the design stored in advance in the third storage means, and the highest correlation is determined by the correlation value calculated by the means for detecting the position of the reference stripes. The position of the reference stripe is detected by detecting the high position of the reference stripe. The position of the reference stripe in small areas above and below or to the left and right of the detected reference stripe is detected, and rotational deviation is detected from that position. In this way, the position of the reference stripes and rotational deviation can be detected not by correlation using images, but by correlation of the stripe width pattern, which has little effect on rotational deviation.This allows the cutting position to be corrected for rotational deviation, which could not be corrected in the past. Correction can be made.

〔Example〕

FIG. 1 is a block diagram showing an automatic cutting device according to an embodiment of the present invention, and FIG. 2 is a detailed block diagram of the pattern position detecting section 4 shown in FIG. Note that the same reference numerals as in FIG. 6 indicate the same or corresponding parts, and detailed explanation thereof will be omitted. The block configuration diagram shown in FIG. 1 is the same as the conventional device shown in FIG. 6, but the difference from the conventional device is the symbol position detection section 4. Therefore, FIG. 2 showing the detailed block configuration will be explained. In the figure, 10 is a stripe width pattern detection section that detects a stripe width pattern, 12 is a characteristic stripe width pattern storage section that stores a characteristic stripe width pattern of a pattern in advance, and 13 is a stripe width pattern storage section 11 and the characteristic stripe width pattern. A stripe width pattern correlation calculating section 14 calculates the correlation between the respective stripe width patterns stored in the stripe width pattern storage section 12, and a reference stripe position extracting section 14 determines the position of the reference stripe from the correlation value. 15 is a small region reference stripe position detection unit that detects the position of the reference stripe in a small region above and below or to the left and right of the detected reference stripe;
Reference numeral 16 denotes a rotation calculation unit that calculates the rotation angle of rotational deviation from the position of the reference stripes in the determined small area.

3, 4, and 5 are explanatory diagrams of the operation, and FIG. 3 is an example of image data input from the image input unit 2.Hereafter, the operation will be explained according to an example of processing this image data. explain. FIG. 4 shows a density distribution diagram in the horizontal direction at the center m in the vertical direction of the image in FIG. The stripe width pattern detection unit 10 compares a constant density threshold with its density distribution, detects a stripe width pattern based on the magnitude thereof, and detects a stripe width pattern {w ₀ , w ₁ , w ₂ , . . . …w ₆ } is detected. The detected stripe width pattern {w _i } is stored in the stripe width pattern storage section 11 . The characteristic stripe width pattern {W _i } of the design is stored in advance in the characteristic stripe width pattern storage section 12, and the correlation between this and the stripe width pattern {Wi} stored in the manuscript width pattern storage section 11 is calculated. γ _k is determined by the stripe width pattern correlation calculation unit 13 using the following equation.

γ _k =Σ|W _ik −W _i | The characteristic stripe width pattern is shown in FIG. 4 {W ₀ , W ₁ ,
W ₂ }, the obtained correlation value is the fifth
The result is shown in the figure. The correlation between the stripe width pattern {w _i } and {W _i } is highest when γ _k is minimum. The reference fringe position detection unit 14 calculates the value of k when γ _k is the minimum, and stores it in the characteristic fringe width pattern storage unit 12.
The position of the reference fringe in the input image is determined from the position of the reference fringe stored in advance in the image. In the examples of FIGS. 4 and 5, the correlation value γ _k is k=1
When the minimum characteristic stripe width pattern {W ₀ , W ₁ , W ₂ }
Since W ₂ is the reference width, from the correspondence between {w _i } and {wj}, when k=1, w ₃ corresponds to the width of the reference fringe, so 2 〓 ⁱ⁼⁰ wi + w ₃₂ ) is the reference fringe. Detected as the position of In the case of vertical stripes, move the small area centered around the position of the found reference stripes up and down, in the case of horizontal stripes, to the left and right, and in the case of checkered manuscripts, move up and down around the reference positions of the detected vertical stripes and horizontal drafts. , left and right, and check the image data stored in the image storage unit 6 in that area, and determine the position of each reference document by the small area reference stripe position detection unit 15. In the case of the example shown in Figure 5, the image data in the upper and lower small areas are averaged in the vertical direction, the stripe width pattern of the reference stripes is determined by comparing the magnitude of a specific threshold value, and the center position of the reference stripes is determined. Find it in each small area.

The rotation angle calculation unit 16 calculates the rotational deviation of the reference stripes from each of the determined center positions. In this way, the cutting position is corrected by the cutting control unit 5 based on the position of the reference stripes detected by the pattern position detection unit 4 and the rotational deviation angle, so that the cutting position is always at a constant position with respect to the pattern. The cutting section 2 is controlled accordingly.

〔Effect of the invention〕

As explained above, the present invention uses one-dimensional calculation in an automatic cutting device to determine the position of the reference stripe by correlating stripe width patterns that are less affected by rotational deviation, and then determines the position of the reference stripe by using one-dimensional calculation in an automatic cutting device. By determining the center position of the reference stripes, it is possible to quickly and accurately determine the angle of rotational deviation, which has the excellent effect of making it possible to correct rotational deviation, which was previously impossible. It is something to play.

[Brief explanation of drawings]

FIG. 1 is a block configuration diagram showing an image processing apparatus which is an embodiment of the present invention, FIG. 2 is a detailed block configuration diagram of the symbol position detection section in FIG. 1, and FIG.
Figures 4 and 5 are for explaining the operation of the symbol position detection section in Figure 2. Figure 3 shows the input image data, Figure 4 shows the density distribution and stripe width pattern, and Figure 5 shows the pattern position detection section.
The figure shows the stripe width pattern correlation results. Figure 6 is a block diagram showing a conventional automatic cutting device;
The figure is a detailed block configuration diagram of the symbol position detecting section 4 in FIG. 6, and FIGS. 8, 9, and 10 are for explaining the operation of the symbol position detecting section 4 in FIG. The figure shows input image data, FIG. 9 shows characteristic image data, and FIG. 10 shows correlation value detection results and reference position detection results. In the figure, 1... spreading part, 2... cutting part, 3
...Image input section, 4...Design position detection section, 5...
Cutting control unit, 6... Image storage unit, 7... Feature image storage unit, 8... Image correlation calculation unit, 9... Reference position detection unit, 10... Stripe width pattern detection unit, 11...
Stripe width pattern storage unit, 12... Characteristic stripe width parine storage unit, 13... Stripe width pattern correlation calculation unit, 14
. . . Reference stripe position detection section, 15 . . . Small area reference stripe position detection section, 16 . . . Rotation angle calculation section. In addition,
In each figure, the same reference numerals indicate the same or corresponding parts.

Claims (1)

[Claims] 1 In an automatic cutting device that automatically cuts cloth, a cloth cutting means that moves the cloth to a cutting place and sets the cloth at the cutting place, a cutting means that cuts the cloth, and a pattern of the cloth are input as digital image data. a detection means for detecting the position and rotational deviation of the pattern from the digital image, and a cutting position that is corrected based on the detected position and rotational deviation of the pattern, so that the pattern can always be cut at a constant position. In particular, in the position and rotational deviation detection means of the symbol, a first storage means for storing digital image data and a stripe width pattern for detecting the stripe width pattern of the symbol are provided. a width pattern detection means, a second storage means for storing the stripe width pattern, and a third storage means for storing the characteristic stripe width pattern of the cloth pattern in advance;
Correlation calculation means for calculating the correlation between the respective stripe width patterns stored in the second and third storage means; means for detecting the position of the reference stripe from the correlation value; or comprising means for detecting the position of the reference document in the left and right small areas, and means for calculating the rotational deviation from the position of the reference strip,
The paper width pattern is detected from the input digital image, and the correlation with the characteristic stripe width pattern of the pre-stored design is determined, thereby making it possible to detect the position and rotational deviation of the reference stripe. Features automatic cutting equipment.