JPH0638272B2 - Automatic sewing machine - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sewing device for overlapping and sewing objects having regular patterns, and more particularly to an automatic sewing device using image processing technology.

[Background of the Invention]

Conventionally, in an apparel maker, pattern cutting is performed by human judgment in laminating and cutting fabrics having patterns (patterns) and sewing such as attaching pockets.

However, in recent years, attempts have been made to automatically perform such sewing using image processing technology.

Therefore, using this image processing technique, for example, when a pocket is sewn on a cloth, it is necessary to place the pocket on the cloth, perform image processing on the mounting state, and correct the positional deviation of the pocket. . A major problem has been how to easily and satisfactorily detect the positional displacement amount of the pocket.

[Object of the Invention]

The object of the present invention is to have the same regular pattern (pattern) 2
An object of the present invention is to provide an automatic sewing device capable of automating sewing when two objects are overlapped and sewn.

[Outline of Invention]

In order to achieve the above-mentioned object, the present invention is to sew first and second objects having the same regular pattern in an overlapped manner, in which the second object is moved to a preset setting position. A moving means for superimposing on the first object, an image input means for inputting an image of the first object and an image of the second object moved by the moving means, and A first storage unit for storing an image of the first object input by the image input unit, a second storage unit for storing an image of the second object input by the image input unit, and An OR circuit that outputs a logical OR of the stored images of the first and second objects, and a pattern of the first object and a pattern of the second object based on the output of the OR circuit. A calculation means for calculating the deviation amount is provided and calculated by the calculation means. The said moving unit based on the amount of deviation is obtained so as to correct the position of the second object.

Example of Invention

The process and configuration of the automatic sewing device of the present invention will be described below with reference to the drawings.

FIG. 1 to FIG. 5 are views showing an embodiment in which a pocket having the same pattern as that of the regular pattern is superposed on a shirt having the regular pattern. First, FIG. 1 shows a case in which a pocket 2 is attached to a dress shirt 1 having a striped pattern (striped pattern). An image sensor 4 (ITV camera or the like) captures the images of the wipe 1 and the pocket 2 of the wipe 1 illuminated by the illumination device 3, and the images are processed by the image processing device 5, and both the wipe 1 and the pocket 2 are processed. The tilt amount (Δθ) and the shift amount (Δx, Δy) of the handle are obtained, and the tilt amount and the shift amount are transmitted to the material handling robots 6, 6'as control signals, and the robots 6, 6'are used as the control signals. Based on this, the pocket 2 is rotated and moved. After the patterns of both parties (Waishitsu and Pocket) match, XY-θ
The table 7 is moved and sewing is performed by the automatic machine 8. When operating the pocket 2 with the material handling robots 6, 6 ', the wiper 1 is fixed on the XX-.theta. Table 7 by, for example, temporary adhesion or vacuum suction.

Here, the image sensor 4 recognizes the wiper 1 and the pocket 2 separately based on the patterns attached to them, and in the case of particularly recognizing the pocket, it is performed as described below. . That is, as shown in FIG. 2, let us consider a case where, when pattern matching of the checkered pattern 10 is performed, the upper object 12 which is a pocket is superposed on the lower object 11 which is a waiter. Since the set position of the upper object 12 has been instructed to the material handling robot 6 in advance, it does not largely deviate from the target position, and the pattern (pattern) is slightly displaced. Therefore, the visual field of the easy sensor 4 is set to be inside the contour line 14 of the upper object 12 as shown by the broken line 13. Humans can also make this setting.

In addition to the above, provisionally, as shown by the alternate long and short dash line 15 for setting the visual field of the image sensor 4, the contour line 14 of the upper object 12,
Furthermore, when the contour line of the lower object 11 is set to be included, the lighting method is devised so that the contour line does not appear as an image, or a process of removing the contour line by some image processing method. Alternatively, it is necessary to recognize the contour line and distinguish it from the pattern.

The necessity of distinguishing the contour line from the pattern will be described below.

FIG. 3 shows the binarization of two images of the image of the main body 1 of the Waishatsu and the image of the pocket 2 placed on the main body 1 when the visual field of the image sensor 4 is set inside the contour line 14. The image resulting from the inter-image operation (logical sum operation) after processing is shown. From this image, the amount of deviation between the patterns of the two objects (Δx, Δ
It is shown by x ′, Δy, Δy ′. ) Is easy to find. Here, the smaller one of Δx and Δx ′ is considered as the shift amount in the x-axis direction, and the smaller one of Δy and Δy ′ is considered as the shift amount in the y-axis direction. This is because if the larger displacement amount is selected, the pocket will be displaced from its original position.

In this way, the outline of the pocket 2 appearing on the image surface will not be confused with the wipes and the pocket pattern, and the pocket can be positioned with respect to the waist only by the pattern.

FIG. 4 shows an image of the waist 1 and two images when the pocket 2 is placed on the waist 1 when the visual field of the image sensor 4 is set so as to include the outside or a part of the contour line 14 of the pocket 2. An image obtained as a result of performing inter-image calculation (logical sum calculation) after binarizing the image is shown (this image also includes the handle of the shirt 1 shown by the dotted line in the contour line 14). Note that, for simplicity, when handling an image obtained by binarizing only the image when the pocket 2 is placed on the dresser 1, the pattern portion indicated by the dotted line does not appear as an image. In both images, the contour line 14 becomes an obstacle as shown in FIG.
Axial and y-axis deviations Δx, Δx ′ and Δy, Δ
It will be a difficult process to obtain each y '.

FIG. 5 is a block diagram showing an embodiment of the above-mentioned image processing. In the figure, there are an image sensor 21 for recognizing a pattern of a shirt and an image sensor 22 for recognizing a pattern of a pocket. The outputs of these are input to the binarization processing devices 23 and 24, respectively, and their outputs are respectively. It is stored in the storage devices 25 and 26. The stored data is stored in the storage device 28 via the OR circuit 27. The stored data of the storage device 28 is C
It is output to the RT 29 and also input to the arithmetic circuit 30. The arithmetic circuit 30 calculates Δx and Δy shown in FIG. 3, and drives the arm 31 of the robot based on the calculated values.

According to the present embodiment, the image of the pattern of the shirt and the image of the pattern of the pocket are stored respectively, and since these ors are taken, the pattern of the shirt and the pattern of the pocket hidden when the pocket is placed on the shirt. It is possible to satisfactorily detect the amount of deviation from

〔The invention's effect〕

According to the present invention, the amount of deviation between the pattern of the lower object and the pattern of the upper object that are hidden by the superposition is satisfactorily detected, and the position of the object is corrected by this misalignment so that the superposition is performed. Since the sewing is performed in this state, the sewing device for overlapping and sewing two objects can be automated.

[Brief description of drawings]

FIG. 1 is a diagram showing an embodiment of a visual field setting method for an image sensor according to the present invention. FIG. 1 is a block diagram showing an image processing system to which the method is applied. FIG. 2 is a visual field setting method for an image sensor according to the present invention. FIG. 3 is an explanatory view showing a method for correcting a displacement of an object in the image processing system, and FIG. 4 is an explanation showing an example of a conventional visual field setting method of an image sensor. 5 and 5 are circuit diagrams showing an embodiment of a processing circuit in the image processing system. DESCRIPTION OF SYMBOLS 1 ... Waishatsu, 2 ... Pocket, 3 ... Illumination device, 4 ... Image sensor, 5 ... Image processing device, 6 ... Material handling robot, 7 ... XY-theta table, 8 ... Automatic machine (automatic sewing machine), 10 ... Lattice Pattern, 11 ... Lower object, 12 ... Upper object, 13 ... Image sensor field of view, 14 ... Contour line of upper object.

Claims (1)

[Claims] 1. A first and a second having the same regular pattern.
In the sewing of overlapping objects, the second object is moved to a preset position designated in advance,
Moving means for superposing on the first object, image input means for inputting the image of the first object and image of the second object moved by the moving means, and the image input A first storage unit for storing an image of the first target object input by the means; a second storage unit for storing an image of the second target object input by the image input unit; An OR circuit that outputs a logical sum of the images of the first and second objects, and a deviation between the pattern of the first object and the pattern of the second object based on the output of the OR circuit. An automatic sewing device, characterized in that a calculating means for calculating the amount is provided, and the moving means corrects the position of the second object based on the shift amount calculated by the calculating means.