JP2014016162A - Method for detecting corner vertex of sheet-like rubber material - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for detecting a corner vertex of a sheet-like rubber material which can accurately detect a position of the corner vertex of the sheet-like rubber material conveyed by a conveyor.SOLUTION: Provided is a method for detecting at least one corner vertex 18 of a sheet-like rubber material 2 by using a computer. The method includes: an imaging step for storing image data D of an edge area including a corner vertex 18 in the computer; a first step in which the computer identifies a corner area that includes the corner vertex 18 and is smaller than the edge area of the image data D by performing image processing; and a second step in which the computer detects the corner vertex 18 in the corner area. Thus, even when deformation is generated around an actual corner vertex of the sheet-like rubber material 2, it is possible to accurately detect a position of the corner vertex 18 of the sheet-like rubber material 2.

Description

  The present invention relates to a corner vertex detection method capable of accurately detecting, for example, a corner vertex of a sheet-like rubber material placed on a conveyor.

  Conventionally, a long belt ply used for manufacturing a tire is formed by sequentially joining sheet-like rubber materials which are a plurality of parallelogram-shaped ply pieces conveyed on a conveyor. As shown in FIG. 12, a forming apparatus a that performs such joining includes a conveyor c that sequentially conveys a plurality of sheet-like rubber materials b, an imaging means d provided above the conveyor c, and this imaging means. a computer e connected to d.

  In the forming apparatus a, the front edge h of the upstream sheet-like rubber material b1 conveyed by the conveyor c and the rear edge h of the downstream sheet-like rubber material b2 are photographed by the photographing means d. The Thereby, for example, the image data i of the area indicated by the virtual line is stored in the computer e.

  The computer e processes the image data i, detects a plurality of points on the contour line of the pair of side edges f, g extending in the longitudinal direction of the upstream sheet-like rubber material b1, and based on these, the side edges f, An approximate straight line that can be regarded as g is calculated. Similarly, the computer e detects a plurality of points on the contour line of the edge h, and calculates an approximate straight line that can be regarded as the edge h based on these. Then, the computer e calculates the coordinates of the intersection where the two approximate lines intersect on the two-dimensional plane. The coordinates of this intersection are the positions of the corner vertices fh and gh of the sheet-like rubber material b1. Similarly to this, the computer e detects each corner vertex of the rear edge h of the sheet-like rubber material b2 on the downstream side.

  In the conventional method, the approximate straight lines of the side edges f and g and the edge h are calculated based on the image data i of a relatively wide area of the sheet-like rubber material b. For this reason, when the deformation | transformation has generate | occur | produced in corner vicinity fh, gh of the sheet-like rubber material b, in a conventional method, there exists a possibility that a position greatly different from actual corner vertices fh, gh may be detected.

  The present invention has been made in view of the above problems, and provides a corner vertex detection method for a sheet-like rubber material that can accurately detect the position of the corner vertex of a sheet-like rubber material placed on a conveyor. .

  According to the first aspect of the present invention, when the sheet-like rubber material having a pair of side edges extending in the longitudinal direction and an end edge connecting between the side edges is viewed in a plane, the side edge and the end edge are seen. A corner apex detection method for a sheet-like rubber material that detects at least one corner apex that intersects with a computer using a computer, and images an end region including the corner apex of the sheet-like rubber material in a plan view. An imaging process for storing the image data in the computer, and the computer processes the image data to identify a corner region that includes the corner vertex and is smaller than the end region. And a second step in which the computer detects the corner vertex from the corner area.

  According to a second aspect of the present invention, the first step includes a step of detecting two spaced apart first points P1 and P2 on the outline of the side edge of the image data, and the first point. A step of rotating the image data so that a straight line passing through P1 and the second point P2 is parallel to a predetermined reference straight line; and one third point P3 on the contour of the edge of the image data. The corner vertex detection method for a sheet-like rubber material according to claim 1, comprising a step of detecting, and a step of specifying the corner region based on a relative positional relationship between the straight line and the third point P3.

  According to a third aspect of the present invention, the second step includes a step of identifying a side edge straight line portion in which the contour of the side edge can be regarded as a straight line in the corner region, and a contour of the edge in the corner region. The sheet form according to claim 1 or 2, comprising a step of specifying an edge straight line portion that can be regarded as a straight line, and a step of detecting a point where the side edge straight line portion and the edge straight line portion intersect as the corner apex. This is a method for detecting the corner apex of a rubber material.

  The invention according to claim 4 is the corner vertex detection method for sheet rubber material according to claim 3, wherein the side edge straight line portion and the edge straight line portion have a length of 10 to 50 mm.

  The invention according to claim 5 is the corner apex detection method for a sheet-like rubber material according to any one of claims 1 to 4, wherein the imaging step is performed by illuminating the sheet-like rubber material from the side. is there.

  According to a sixth aspect of the present invention, the color of the sheet-like rubber material is black, and the imaging step is performed by placing the sheet-like rubber material on a white-colored conveyor. 5. A corner apex detection method for a sheet-like rubber material according to any one of 5 above.

  According to the present invention, the computer processes the image data obtained by photographing the edge region including the corner vertex of the sheet-like rubber material, thereby identifying the corner region which is the image region including the corner vertex and smaller than the image data. A first step is performed. After this first step, a second step is performed in which the computer detects a corner vertex from the corner area. In this second step, a corner vertex is detected from a corner area smaller than the image data acquired first. Therefore, even when there is a deformation near the actual corner vertex of the sheet-like rubber material, the position of the corner vertex of the sheet-like rubber material can be detected with high accuracy.

It is a top view of the corner vertex detection apparatus which shows embodiment of this invention. (A) is a top view of the joined sheet-like rubber material, (b) is a side view of the same sheet-like rubber material. (A) is a top view which shows an imaging | photography means, (b) is AA sectional drawing of Fig.3 (a). It is a top view which shows an example of the image data image | photographed with the imaging | photography means. It is a top view which shows the other example of the image data image | photographed with the imaging | photography means. It is a flowchart of the corner vertex detection method of this embodiment. It is a flowchart of the 1st process of FIG. (A) is explanatory drawing which shows the image processing of step S12 of FIG. 7, (b) is explanatory drawing which shows the image processing of step S14 of FIG. (A) is explanatory drawing which shows the image processing of step S16 of FIG. 7, (b) is explanatory drawing which shows the image processing of step S18 of FIG. It is a flowchart of the 2nd process of FIG. (A) is explanatory drawing which shows the image processing of step S21 of FIG. 10, (b) is explanatory drawing which shows the image processing of step S22 of FIG. It is the schematic which shows the formation apparatus containing the conventional corner vertex detection apparatus.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
As shown in the plan view of FIG. 1, the corner vertex detection device 1 for the sheet-like rubber material 2 according to the present embodiment is indicated by a conveyor 3 that continuously conveys the plurality of sheet-like rubber materials 2 and a virtual line. As described above, the photographing means 5 for photographing the sheet-like rubber material 2 is provided. A joining means 4 for joining the sheet-like rubber material 2 is provided on the downstream side of the photographing means 5.

  The sheet-like rubber material 2 is, for example, a belt ply piece of a tire. This sheet-like rubber material 2 is formed by sequentially cutting a long belt-like body 12 in which an array of tire cords 11 aligned in parallel is covered with rubber. The long band-like body 12 is cut obliquely at a predetermined angle θ with respect to the tire cord 11 by, for example, a cutter 13. Thereby, the sheet-like rubber material 2 has a parallelogram shape. The cut sheet rubber material 2 is conveyed along the conveyance direction X of the conveyor 3.

  In this specification, as shown in FIG. 2 (a), in the sheet-like rubber material 2, of the side edges extending along the longitudinal direction, the left side toward the downstream side in the transport direction X (the front side in the drawing). The left side edge 6 is the left side edge 6 and the right side edge (illustrated side) is the right side edge 7. Further, an end edge that connects the ends of both side edges 6 and 7 on the downstream side in the transport direction X is a front end edge 8, and an upstream edge in the transport direction X is a rear end edge 9.

  The joining means 4 includes a gripping device that grips the sheet-like rubber material 2 (not shown). As shown in FIGS. 2A and 2B, this gripping device positions the front edge 8 of the upstream sheet-like rubber material 2a and the rear edge 9 of the downstream sheet-like rubber material 2b. Can be joined together. For this alignment, it is necessary to detect the corner apex of the sheet-like rubber material 2. As a joining method, for example, a zipper joint in which the front end edge 8 and the rear end edge 9 are abutted and joined is used. The sheet-like rubber material 2 joined by a zipper joint is used for manufacturing a tire as a long belt ply.

  As the sheet-like rubber material 2 of the present embodiment, a black rubber material used for a general tire is used. For this reason, the color of the sheet-like rubber material 2 is black.

  As shown in FIG. 3A or 3B, the photographing means 5 includes a pair of lighting devices 15 arranged on the left and right sides of the conveyor 3 and a camera 16 installed above the conveyor 3. . Further, the computer 10 is connected to the camera 16 as shown in FIG.

  The illuminating device 15 irradiates the sheet-like rubber material 2 placed on the conveyor 3 with light from the side, not directly above. For this reason, even if the light irradiated from the illuminating device 15 is reflected on the conveyance surface of the conveyor 3 and the outer surface of the sheet-like rubber material 2, the reflected light is suppressed from being input to the camera 16. Therefore, the sheet-like rubber material 2 is clearly photographed by the camera 16. In addition, the illuminating device 15 should just irradiate light to the sheet-like rubber material 2 from the side, for example, should just be distribute | arranged to either right or left.

  For the camera 16, for example, a well-known photographing device such as a digital camera or a CCD camera is used. In the present embodiment, the front end edge 8 of the upstream sheet-like rubber material 2a and the rear end edge 9 of the downstream sheet-like rubber material 2b are individually photographed by one camera 16. Note that the front end edge 8 of the upstream sheet-like rubber material 2a and the rear end edge 9 of the downstream sheet-like rubber material 2b may be photographed simultaneously by one camera. The photographing means 5 includes two cameras. The front edge 8 of the upstream sheet-like rubber material 2a is photographed by one camera, and the rear end of the downstream sheet-like rubber material 2b is photographed by the other camera. The edge 9 may be photographed.

  In the corner vertex detection method of the present embodiment, the camera 16 photographs the front end edge 8 of the upstream sheet-like rubber material 2a or the rear end edge 9 of the downstream sheet-like rubber material 2b, and this image data D Is stored in the computer 10.

  In FIG. 3A, an example of a shooting range P by the camera 16 is indicated by a virtual line. 4 shows a plan view of the image data D in which the front edge 8 of the sheet-like rubber material 2a is photographed and FIG. 5 shows the rear edge 9 of the sheet-like rubber material 2b. Note that the image data D is used in two-dimensional coordinates xy where x is the conveying direction of the conveyor and y is the direction perpendicular thereto.

  In the image data D, in order to clarify the boundary between the sheet-like rubber material 2 and the conveyor 3, the color of the conveying surface of the conveyor 3 is preferably a color having a large contrast with the sheet-like rubber material 2. For example, when the sheet-like rubber material 2 is black as in the present embodiment, the upper surface of the conveyor 3 is preferably white. On the other hand, when the sheet-like rubber material 2 is white, the upper surface of the conveyor 3 is preferably black.

  As shown in FIG. 4, image data D obtained by photographing the front end edge 8 of the upstream sheet-like rubber material 2 a includes a first corner vertex 18 a formed by the front end edge 8 and the left side edge 6, and the front end edge 8. And a second corner vertex 18 b formed by the right edge 7. As shown in FIG. 5, the image data D obtained by photographing the downstream sheet-like rubber material 2 b includes a third corner vertex 18 c formed by the rear edge 9 and the left edge 6, and the rear edge 9 and the right edge. And a fourth corner vertex 18d formed at the edge 7.

  Next, processing by the computer 10 of the corner vertex detection method of the present embodiment will be described in detail. In the following description, detection of the first corner vertex 18a shown in FIG. 4 will be described as an example, but other vertexes can be detected by the same processing.

  An example of the processing procedure by the computer 10 is shown in FIG. As shown in FIG. 6, image data D is input from the camera 16 to the computer 10. The input image data is subjected to, for example, noise cut and binarization processing as preprocessing. As a first step S1, the computer 10 processes the image data D to identify a corner area R that is an area including the corner vertex 18 and is smaller than the image data D.

  FIG. 7 shows a flowchart of a more specific processing procedure of the first step S1 for specifying the corner region R. First, the computer 10 detects the first point P1 (step S11). As shown in FIG. 8A, the first point P1 is a point on the contour line of the left edge 6 of the sheet-like rubber material 2a. The first point P1 is set based on the difference in brightness between the sheet-like rubber material 2a and the conveyor 3 in the image data D. That is, the first point P1 is detected as a binary information change point by scanning a pixel having an arbitrary x coordinate value in the y direction.

  Next, the computer 10 detects the second point P2 (step S12). The second point P2 is a point separated from the first point P1 by a predetermined distance on the outline of the left edge 6 of the sheet-like rubber material 2a. Similarly to the first point P1, the second point P2 can be detected as a binary information change point by scanning a pixel having an arbitrary x coordinate value in the image data D in the y direction.

  In the present embodiment, the imaging process is performed in a state where the shooting angle of the sheet-like rubber material 2 is fixed in advance. Therefore, in the image data D, by adjusting the value of the x coordinate within a certain range, both the first point P1 and the second point P2 are adjusted to be detected on the left edge 6 without fail. .

  Next, as shown in FIG. 8B, the computer 10 calculates a straight line L1 passing through the first point P1 and the second point P2 (step S13). The straight line L1 can be calculated using the coordinates of the detected first point P1 and second point P2.

  Next, the computer 10 rotates the image data D so that the straight line L1 obtained in step S13 and the reference straight line L set in advance on a two-dimensional plane are parallel to each other (step S14).

  Next, as shown in FIG. 9A, the computer 10 moves the image data D in the y direction, and superimposes the straight line L1 and the reference straight line L (step S15).

  When the reference straight line L and the straight line L1 are overlapped in step S15, the computer 10 detects the third point P3 (step S16). The third point P3 is a point on the contour line of the front edge 8, and a pixel having an arbitrary y coordinate value is scanned in the x direction, and is detected as a change point of the binary information.

  The third point P3 can also be detected on the front edge 8 by restricting the y-coordinate value within a certain range with the reference straight line L as a reference.

  Next, as shown in FIG. 9B, the computer 10 moves the image data D in the x direction along the reference straight line L (step S17). At this time, the image data D is moved so that the coordinates of the third point P3 become a preset value.

  Next, the computer 10 specifies the corner area R based on the relative positional relationship between the reference straight line L and the third point P3 (step S18). The corner region R is, for example, a region including the upper and lower regions of the reference straight line L and the left and right regions of the third point P3 on a two-dimensional plane. This region is geometrically set as a range sufficiently including the corner apex 18a by using the angle θ, the reference straight line L, and the third point P3 of the sheet-like rubber material. Further, the corner region R of the present embodiment is a rectangle whose longitudinal direction is along the x direction.

  Next, as shown in FIG. 6, the computer 10 executes the second step S <b> 2 for the corner region R, and detects the corner vertex 18 a from the corner region R (step S <b> 2). A flowchart of a specific processing procedure of the second step S2 is shown in FIG.

  In the second step S2, as shown in FIG. 10 and FIG. 11A, first, a plurality of arbitrary points Ni on the contour line of the left edge 6 of the sheet-like rubber material 2a are detected (step S21). In order to improve the detection accuracy, the number of the points is preferably 3 or more, particularly 5 or more. In the present embodiment, there are five points N1 to N5. The arbitrary points N1 to N5 are respectively detected at positions separated by a predetermined distance set in advance. These points N1 to N5 are also detected as changing points of binary information of pixels in the y direction. In step S21, the left edge straight line portion L2 that can be regarded as a straight line is specified based on the five points N1 to N5 on the outline of the left edge 6.

  Next, as shown in FIG. 11B, the computer 10 detects a plurality of arbitrary points Ki on the contour line of the front edge 8 of the sheet-like rubber material 2a (step S22). In order to improve the detection accuracy, the number of the points is preferably 3 or more, particularly 5 or more. In the present embodiment, there are five points K1 to K5. The arbitrary points K1 to K5 are detected at positions separated from each other by a predetermined distance. Further, these points K1 to K5 are detected as changing points of binary information of pixels in the x direction. In step S22, an edge straight line portion L3 that can be regarded as a straight line is specified based on the five points K1 to K5 on the contour line of the front edge 8.

  In addition, it is desirable that the side edge straight line portion L2 and the end edge straight line portion L3 detected in step S21 and step S22 are defined to have a length of, for example, 10 to 50 mm in the range of the corner region R. When the lengths of the side edge straight line portion L2 and the end edge straight line portion L3 are smaller than 10 mm, if the curl or the like is deformed in the contour line of the left edge 6 and the contour line of the front edge 8, the side edge There is a possibility that the straight line portion L2 and the edge straight line portion L3 may not be accurately specified. In addition, when the length of the side edge straight line portion L2 and the end edge straight line portion L3 is larger than 50 mm, if the deformation other than the curl occurs in the contour line of the left edge 6 and the contour line of the front edge 8, There is a possibility that the side edge straight line portion L2 and the end edge straight line portion L3 may not be accurately specified.

  Next, the computer 10 calculates the coordinate position of the intersection of the identified left-side edge straight line portion L2 and the front end edge straight-line portion L3 (step S23). This intersection is calculated, for example, as the intersection of the approximate straight line calculated using the points N1 to N5 of the left edge straight line portion L2 and the approximate straight line calculated using the points K1 to K5 of the edge straight line portion L3. The

  As a result, the computer 10 detects the calculated coordinate position of the intersection as the position of the corner vertex 18a (step S24).

  The computer 10 can detect the position of the second corner vertex 18b of the sheet-like rubber material 2a on the upstream side by executing the same processing as S11 to S18 and S21 to S24. Further, the computer 10 can detect the third corner vertex 18c and the fourth corner vertex 18d of the downstream sheet-like rubber material 2b by using the image data D shown in FIG.

  As described above, in the present embodiment, the corner region R in a range smaller than the image data D is specified in the first step S1. Thereafter, a corner vertex 18 is detected based on the corner region R in the second step S2. For this reason, even when there is distortion or deformation on the apex side of the sheet-like rubber material 2, the position of the corner apex 18 of the sheet-like rubber material 2 can be detected with high accuracy.

  In this embodiment, the computer 10 inputs the detected positions of the first to fourth corner vertices 18a to 18d to the joining means 4. In the joining means 4, the upstream sheet-like rubber material 2 a is gripped based on the input position of each corner vertex 18, and the front edge 8 of the upstream sheet-like rubber material 2 a is connected to the downstream sheet-like rubber material. The material 2b can be accurately aligned and joined to the rear edge 9 of the material 2b. About this, a well-known means is used suitably.

  As mentioned above, although the corner vertex detection apparatus 1 of this embodiment was explained in full detail, this invention is not limited to this embodiment, It can deform | transform and implement in a various aspect.

In order to confirm the effect of the present invention, a long belt ply was prototyped using the corner apex detection device shown in FIG. The specifications of the corner vertex detection device are as follows.
Computer: Visual sensor controller FZ3-H900 (made by OMRON Corporation)
Camera: 20 million pixel color digital camera Lens: 8mm lens Illumination device: LED lighting for image processing Illumination power supply: Pulse width dimming power supply Camera installation height: 800mm above conveyor top surface Shooting range P: 680mm x 490mm

  As the sheet-like rubber material, a black rubber material having a length of 1790 to 1830 mm parallel to the transport direction and a width of 150 to 180 mm perpendicular to the transport direction was used. Moreover, the conveyance surface of the conveyor is painted white. The conveyor surface was tack-free with fluororesin for the purpose of improving non-viscosity. Further, a general joining means for connecting the front edge and the rear edge of each sheet-like rubber material is arranged on the downstream side in the transport direction of the camera installation position.

  As a result of the test, when the joint portion of the joined sheet-like rubber material was examined with the naked eye, the amount of positional deviation at each corner apex was 1 mm at the maximum. On the other hand, in the case of the corner apex detection method used in the conventional forming apparatus, the positional deviation amount of each corner apex of the joined sheet-like rubber material was 3 to 4 mm.

DESCRIPTION OF SYMBOLS 1 Corner vertex detection apparatus 2 Sheet-like rubber material 3 Conveyor 4 Joining means 5 Image | photographing means 6 Left edge 7 Right edge 8 Front edge 9 Rear edge 10 Computer 15 Illumination device 16 Camera 18 Corner vertex X Conveyance direction S1 1st process S2 1st 2 processes P photographing range D image data L reference straight line L1 straight line L2 side edge straight line part L3 edge straight line part R corner area

Claims (6)

When a sheet-like rubber material having a pair of side edges extending in the longitudinal direction and an end edge connecting between the side edges is viewed in a plane, at least one corner apex where the side edges and the end edges intersect is calculated by a computer. A method for detecting a corner vertex of a sheet-like rubber material to be detected using:
In a plan view, an image capturing step of capturing an end region including the corner apex of the sheet-like rubber material and storing the image data in the computer;
A first step of identifying a corner area that is an image area that includes the corner vertex and is smaller than the edge area by performing image processing on the image data;
A corner vertex detection method for a sheet-like rubber material, wherein the computer includes a second step of detecting the corner vertex from the corner region. The first step includes
Detecting two separated first points P1 and P2 on the contour of the side edge of the image data;
Rotating the image data so that a straight line passing through the first point P1 and the second point P2 is parallel to a predetermined reference straight line;
Detecting one third point P3 on the edge contour of the image data;
The corner vertex detection method for a sheet-like rubber material according to claim 1, further comprising a step of specifying the corner region based on a relative positional relationship between the straight line and the third point P3. The second step includes
In the corner area, a step of identifying a side edge straight line portion where the outline of the side edge can be regarded as a straight line;
Identifying an edge straight line portion in which the contour of the edge can be regarded as a straight line in the corner region;
The corner vertex detection method of the sheet-like rubber material according to claim 1 or 2 including the step of detecting the point where said side edge straight line part and said edge straight line part intersect as said corner vertex.   4. The corner apex detection method for a sheet-like rubber material according to claim 3, wherein the side edge straight line portion and the end edge straight line portion have a length of 10 to 50 mm.   The corner detection method for a corner of a sheet-like rubber material according to any one of claims 1 to 4, wherein the imaging step is performed by illuminating the sheet-like rubber material from the side. The color of the rubber sheet material is black,
6. The corner apex detection method for a sheet-like rubber material according to claim 1, wherein the imaging step is performed by placing a sheet-like rubber material on a white-colored conveyor.

Images