JP4138943B2 - Plate shape recognition device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a plate shape recognition device that is applied to, for example, a laser processing machine and recognizes the time measurement of the uppermost plate material from among several stacked plate materials.
[0002]
[Prior art]
In general, a laser processing machine uses a computer to nest a workpiece having a shape corresponding to the thickness of a product, and cuts the plate material by laser processing based on this nesting data. Yes.
[0003]
The product obtained by laser processing and the remaining material from which the product is separated are handled separately, and the remaining material is often stacked on a pallet or the like, for example.
[0004]
Since products obtained by such a laser processing machine have various sizes and shapes, the size and shape of the remaining portion of the remaining material from which the product is separated are also various.
[0005]
Therefore, there are cases where the necessary product can be obtained by using the remaining part of the remaining material. In fact, as a plate material that cuts out the product because the unit price of the product is low or the material order is not in time. The remaining material is used.
[0006]
[Problems to be solved by the invention]
However, in order to cut the product using the remaining material, it is necessary to nest the target product with respect to the remaining material, but conventionally, the size of the remaining portion of the remaining material stacked on a pallet or the like, There is a problem that it is difficult to automatically measure the shape in a stacked state.
[0007]
This invention solves the problems of the above-mentioned conventional ones, and can detect the outer shape of the uppermost plate material from a plurality of stacked plate materials and the hole shape when there is a hole. The purpose is to submit the shape recognition device.
[0008]
[Means for Solving the Problems]
In order to achieve the above object, the present invention is an apparatus for recognizing the shape of the uppermost plate member from among a plurality of stacked plate members, and is based on an image pickup device that picks up the plate member from above and obtained image information. The uppermost surface detection unit that detects the surface area of the uppermost plate material, the edge detection unit that detects the position of the edge of the plate material of each part based on the image information, and the region where the edge exists From the edge trace part that makes a round to the trace start point while always selecting a certain direction at the branch point, the round direction, and the selection direction at the branch point, And a determination unit for determining whether the uppermost plate has an outer shape or a hole shape.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described based on examples with reference to the drawings.
[0010]
FIG. 1 is a block diagram showing an embodiment of a plate shape recognition device according to the present invention. This shape recognition device recognizes the shape of the uppermost plate material from a plurality of stacked plate materials. The imaging device 1, the uppermost surface detection unit 3, the edge detection unit 5, the edge trace unit 7, the determination unit 9, and the control unit 11 that controls them are provided.
[0011]
The imaging device 1 captures a plurality of overlapping plate materials from above, and is configured using, for example, a device such as a line sensor or an area sensor using a CCD camera.
[0012]
Such imaging is performed by scanning a required imaging region, for example, above a plate placed on a pallet in order to input a plurality of overlapped plates as image information. The captured image information is spatially sampled and arranged on the memory coordinates as a luminance signal representing the brightness of each pixel.
[0013]
Although not shown, this image information is subjected to preprocessing such as density correction and image quality improvement as necessary so as to be suitable for subsequent image processing.
[0014]
The uppermost surface detection unit 3 detects the surface area of the uppermost plate material based on the image information obtained by the imaging device 1. Such detection of the uppermost surface is obtained by the imaging device 1. In the obtained image, the fact that the surface of the uppermost plate is brighter than the other portions is used.
[0015]
Therefore, by applying a predetermined threshold value relating to light and darkness to the image information obtained by the imaging device 1, the surface area of the uppermost plate material can be detected separately from the other portions. The other parts include a part where the surface of the second sheet from the top is partially exposed and a part where the surface of the third sheet or less is partially exposed. There is no need to distinguish between them.
[0016]
The edge detection unit 5 detects the position of the edge of the plate material of each part based on the image information obtained by the imaging device 1, and the detection of such an edge exists where the edge changes drastically. It uses the fact that it does.
[0017]
Therefore, the edge detection unit 5 includes an appropriate edge enhancement filter using a high-pass filter, and a Sobel filter, a Laptian filter, or the like can be used as the edge enhancement filter. Then, by using the edge enhancement filter, the edge detection unit 5 can detect not only the edge located at the periphery of the surface area of the uppermost plate material, but also the edge of the exposed portion of the second plate material from the top, The edge of the exposed portion of the plate material below the eye can also be detected.
[0018]
As described above, the top surface detection unit 3 is configured such that the surface area of the top plate material is exposed from the top surface of the second, third, and lower plate materials, or the pallet that is not covered by the plate material. On the other hand, the edge detection device 5 detects not only the edge of the uppermost plate material but also the edges of the second, third, and lower plate materials from the top as similar edges. It is supposed to be.
[0019]
Then, in order to detect only the edge of the uppermost plate material, the edge detection unit 5 separates from the edge detection unit 15 that detects the position of the edge of the plate material of each part, and the highest level obtained by the uppermost surface detection unit 3. The uppermost edge detector 17 is configured to detect only the edge of the uppermost plate from the surface area data of the plate and the edge position data obtained by the edge detector 5.
[0020]
The edge trace unit 7 traces the region where the edge exists along the edge from the uppermost surface, and makes a round to the trace start point while always selecting a certain direction at the branch point. The edge area is traced using a filter.
[0021]
2, x (i, j) indicates a peripheral pixel arrangement used in the spatial filter output _{calculations, C (i n, j n} ) center _{pixel, A 1 (i n-1} , j n) ~ A 8 (I _n-1 , j _n-1 ) are pixels arranged around the center C in the clockwise direction. These may literally be the pixels themselves, but the edge width is larger than the pixels, so if the pixel of 3 rows and 3 columns cannot indicate the edge and its neighborhood, it will fit the edge width. As described above, it is possible to configure a pixel aggregate composed of m rows and m columns of pixels in 3 rows and 3 columns. In this specification, such a pixel aggregate is also referred to as a pixel.
[0022]
When tracing an edge region using a spatial filter having a pixel arrangement as shown in FIG. 2, the edges are A ₁ (i _n−1 , j _n ) to A ₈ (i _n−1 , j _n−1 ). enter from anywhere, the center C (i _n, j _n) through, check the Where do go out of _{a 1 (i n-1,} j n) ~A 8 (i n-1, j n-1) The table is shown in FIG.
[0023]
In FIG. 3, B (B _{1 to} B ₈ ) indicates what enters from A _{1 to} A ₈ . For example, those coming from the A ₂ a (A ₂ → C) and B _2,
B ₂ = C−A ₂
= {(I _n −i _n−1 ), (j _n −j _{n + 1} )}
Here, if i _n → i, j _n → j,
B ₂ = [{i− (i−1)}, {j− (j + 1)}]
= (1, -1)
It becomes.
[0024]
On the other hand, in FIG. 3, F (F _{1 to} F ₈ ) indicates what goes out of A _{1 to} A ₈ . For example, if those leaving the A ₆ a (C → A ₆₎ and F _6,
F ₆ = A ₆ -C
= {(I _{n + 1} −i _n ), (j _n−1 −j _n )}
Here, if i _n → i, j _n → j,
F ₆ = [{(i + 1) −i}, {(j−1) −j}]
= (1, -1)
It becomes.
[0025]
FIG. 3 shows the value D (i, j) of B + F. For example, when entering from A ₂ and exiting from A ₆ (A ₂ → C → A ₆ ), the edge of the portion extends linearly. At this time,
D ₂₆ = B ₂ (1, -1) + F ₆ (1, -1)
= (2, -2)
It is.
[0026]
When exiting from A ₅ (A ₂ → C → A ₅ ), the edge turns 45 ° (L2 / 1〓R) to the left at that portion and exits from A ₄ (A ₂ → C → A ₄ ) If the edge turns 90 ° (L 左 R) to the left at that part and exits from A ₃ (A ₂ → C → A ₃ ), the edge is 135 ° to the left at that part (L 3 / L 2 〓R) Turn.
[0027]
On the other hand, when exiting from A ₇ (A ₂ → C → A ₇ ), the edge bends 45 ° (R2 / 1〓R) to the right at that portion and exits from A ₈ (A ₂ → C → A ₈ ) When the edge turns 90 ° (R〓R) to the right at that part and exits from A ₁ (A ₂ → C → A ₁ ), the edge turns 135 ° to the right at that part (R3 / R 2 〓R) Turn.
[0028]
FIG. 3 shows D values for all combinations of B (B _{1 to} B ₈ ) + F (F _{1 to} F ₈ ).
[0029]
FIG. 4 shows the D values in FIG. 3 rearranged based on the bending direction of the edge. In FIG. 4, when B (i, j) is Bi, Bj, it is divided into a case where either Bi or Bj is 0 or a case where neither is 0, and details are as follows.
[0030]
( ₁ ) When Bj = 0 (B ₁ , B ₅ )
If (sign of Bi) = (sign of Dj) → L
If (sign of Bi) ≠ (sign of Dj) → R
(2) When Bj = 0 (B ₃ , B ₇ )
If (sign of Bj) ≠ (sign of Di) → L
If (sign of Bj) = (sign of Di) → R
(3) When Bi * Bj <0 (B ₂ , B ₆ ),
If | Di |> | Dj | → L
If | Di | <| Dj | → R
( ₄ ) When Bi * Bj> 0 (B ₄ , B ₈ )
If | Di | <| Dj |
If | Di |> | Dj | → R
Therefore, the edge tracing device can trace while detecting the bending direction and the bending angle of the edge from the combination of Bi, Bj and Di, Dj. Therefore, the tracing is performed while always selecting a certain direction at the branch point. It can be done. Then, when one round is made to the trace start point, the larger one is detected as the round direction from the total left turn amount and the right turn amount in the trace.
[0031]
In summary, as shown in FIG. 1, the edge trace unit 7 includes a front detection unit 21 that detects B (B _{1 to} B ₈ ) and a rear detection unit 23 that detects F (F _{1 to} F ₈ ). The bending detection means 25 for detecting the bending of the edge from D (B + F), the direction detection means 27 for detecting the direction of the edge from B and D, and the direction to be selected at the branch point are determined to be a predetermined direction in advance. branch selection means 29 became comprises a total L _T of the left curve amount being traced round, from the right bending amount of total R _T, and a round direction detecting means 31 for detecting the round direction.
[0032]
The determination unit 9 determines the edge of the plate having the highest edge from the round direction obtained by the round direction detection unit 31 of the edge trace unit 7 and the direction of selection at the branch point obtained by the branch selection unit 29. It is determined whether the shape is an outer shape or a hole shape.
[0033]
As shown in FIG. 5, when the round direction and the selection direction at the branch point are the same, the determination unit 9 determines that the rounded edge is the outer shape of the uppermost plate material. ing.
[0034]
On the other hand, as shown in FIG. 6, when the round direction and the direction of selection at the branch point are opposite, the determination unit 9 determines that the rounded edge is the hole shape of the uppermost plate material. It has become.
[0035]
Next, the operation of this embodiment will be described with reference to the flowchart shown in FIG.
[0036]
As shown in FIG. 7, first, a plurality of overlapping plate members are used as objects, and the object is imaged from above by the imaging device 1 (step ST1). The captured image information is spatially sampled and arranged on the memory coordinates as a luminance signal representing the brightness of each pixel.
[0037]
Next, based on the image information obtained by the imaging device 1, the uppermost surface detection device 3 detects the surface area of the uppermost plate material (step ST2). For this reason, the surface area of the uppermost plate material is detected separately from the exposed surface portion of the second, third and lower plate materials from the top, and the pallet surface not covered by the plate material.
[0038]
On the other hand, based on the image information obtained by the imaging device 1, the edge detection unit 15 of the edge detection device 5 detects the position of the edge of the plate material of each unit (step ST3). Therefore, not only the edge of the uppermost plate material but also the edges of the second, third, and lower plate materials from the top are detected as similar edges.
[0039]
Then, the uppermost edge detection means 17 of the edge detection device 5 uses the surface area data of the uppermost plate material obtained by the uppermost surface detection unit 3 and the edge position data obtained by the edge detection means 15. Only the edge of the uppermost plate material is detected (step ST4).
[0040]
Then, the edge trace device 7 starts tracing along the edge from an arbitrary point on the edge (uppermost edge) of the uppermost plate material (step ST5).
[0041]
If there is a branching point in the middle of the trace (step ST6), a predetermined fixed direction is selected (step ST7). Then, the same steps (steps ST6, 7, 8, 9) are repeated until the trace start point is reached (step ST8) and the trace is continued.
[0042]
After one round, the determination unit 9 determines from the round direction and the direction of selection at the branch point (step ST10) that, in the same case, the rounded edge is the outer shape of the uppermost plate (step ST10). ST11) In the opposite case, it is determined that the rounded edge is the hole shape of the uppermost plate (step ST12).
[0043]
After that, if there is another uppermost edge (step ST13), the same steps (steps ST5 to ST12) are repeated, and the process ends when there is no uppermost edge.
[0044]
As described above, the outer shape of the uppermost plate material from among the stacked plate materials, and the hole shape when there is a hole can be automatically detected, and the shape of the plate material according to the present invention The recognition device has various advantages as follows.
[0045]
In general, when a product is processed and taken out from a workpiece by a processing machine, the objects handled by the processing machine are processed from the workpiece before processing, the processed product, the remaining material from which the product has been removed, and the remaining portion of the remaining material. There are secondary products.
[0046]
When the workpiece is a plate material, a plurality of workpieces before processing are generally stacked at a standby position. In order to suck up the uppermost one of the stacked plate materials with an appropriate apparatus and carry it into the machining area, it is necessary to obtain the center of gravity of the uppermost workpiece. The plate shape recognition apparatus according to the present invention can be applied to such applications.
[0047]
In the case of a plate material, the processed products are stacked on a pallet, for example, and sent to the next process. Therefore, in the next process, it is necessary to obtain the center of gravity as in the case of the workpiece. Further, when there are several types of such products, in order to find the target product, it is necessary to confirm whether or not the outer shape matches the target product shape. The plate shape recognition apparatus according to the present invention can be applied to such applications.
[0048]
In addition, a product extraction hole is formed in the remaining material from which the product has been removed. By comparing this hole shape with the target product shape, it can be confirmed whether or not the target product has been processed. it can. The plate shape recognition apparatus according to the present invention can be applied to such applications.
[0049]
【The invention's effect】
Since the present invention is configured as described above, it is possible to detect the outer shape of the uppermost plate material from a plurality of stacked plate materials, and the hole shape when there is a hole. When applied to a processing machine, the size and shape of the remaining part of the remaining material stacked on a pallet, etc. can be automatically measured in the stacked state. There is an effect that the product can be easily nested and the target product can be cut out from the remaining material.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating an embodiment of a plate shape recognition apparatus.
FIG. 2 is a diagram illustrating a peripheral pixel arrangement used for calculating a spatial filter output of x (i, j).
FIG. 3 is a diagram illustrating a relationship between front detection B, rear detection F, and bending detection D;
FIG. 4 is a diagram showing D values rearranged based on edge bending directions;
FIG. 5 is an explanatory view of a round of the outer shape of the uppermost plate member.
FIG. 6 is an explanatory view of making a round of the hole shape of the uppermost plate member.
FIG. 7 is a flowchart showing the operation of the control device.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Imaging device 3 ... Top surface detection part 5 ... Edge detection part 7 ... Edge trace part 9 ... Determination part 11 ... Control part 15 ... Edge detection means 17 ... Top edge detection means 21 ... Front detection means 23 ... Back detection Means 25: Bending detection means 27 ... Direction detection means 29 ... Branch selection means 31 ... Round direction detection means

Claims (3)

A device for recognizing the shape of the uppermost plate material from a plurality of stacked plate materials,
An imaging device for imaging the plate material from above;
Based on the obtained image information, a top surface detector that detects the surface area of the top plate,
An edge detection unit that detects the position of the edge of the plate material of each part based on the image information;
An edge trace section that traces the region where the edge exists along the edge from the top surface, and makes a round to the trace start point while always selecting a certain direction at the branch point,
A determination unit that determines that the edge that has made a round is the outer shape or the hole shape of the uppermost plate material from the round direction and the direction of selection at the branch point,
A shape recognition apparatus for a plate material, comprising: 2. The determination unit according to claim 1, wherein when the one-round direction and the direction of selection at the branch point are the same, the determination unit determines that the one-round edge is an outer shape of the uppermost plate member. Plate shape recognition device. 2. The determination unit according to claim 1, wherein when the one-round direction and the direction of selection at the branching point are opposite, the determination unit determines that the one-round edge is a hole shape of the uppermost plate material. Plate shape recognition device.

Images