Method for checking the calibration of an image processing system of a sheet metal working machine

Abstract

The invention relates to a method for testing an image processing system (18). A calibration plate (24) is arranged at different points in a removal region (14) of a sheet metal working machine (10). In each case, an image is captured and it is checked whether properties extracted from the image correspond with known reference values.

Description

[0001] Method for checking the calibration of an image processing system of a sheet metal working machine

[0002] Background of the invention

[0003] The invention relates to a method for checking the calibration of an image processing system of a sheet metal processing machine, wherein a camera of the image processing system detects a removal area for machined workpieces.

[0004] From WO 2018 / 073418 A1 or WO 2018 / 073419 A1, it is known to point a camera at the sorting table of a flatbed machine tool for sheet metal processing. After calibrating the camera to the sheet metal surface or the sorting table, an image processing algorithm can be used for part recognition. For example, an optimal sorting strategy can be suggested to an operator via a projection screen of a data source.

[0005] To assist operators in sorting processed workpieces from a removal area, the image processing system must be able to correctly identify the workpieces. If the image processing system is not calibrated with sufficient accuracy, this cannot be guaranteed.

[0006] WO 2016 / 005159 A2 describes a method for machining flat workpieces, in particular sheet metal, or three-dimensional workpieces on a processing machine. A live image of a workpiece to be machined is captured using an image capture device for capturing two-dimensional images. For this purpose, the two-dimensional live image of the image capture device must be calibrated to the three-dimensional machine coordinate system. By calibrating at least three live image coordinates (reference image points) to known machine reference points, a machining contour (e.g., a laser cut to be performed) can be displayed superimposed in the live image exactly where the contour would be machined. In this way, reference points in the machine workspace are clearly assigned to reference image points in the live image, thus calibrating the camera. Advantageously, one of the machine reference points can be represented by a movable machine component (e.g.,The image acquisition device can be formed by a laser processing head of a laser processing machine, which has been moved to a position known in the machine coordinate system before capturing the at least one reference live image. Alternatively, machine reference points can be added to a workpiece by machining operations, for example, by marking or cutting out hole circles. In this method, the image acquisition device is calibrated, in particular manually, by aligning a CAD representation with a displayed image of the workpiece.

[0007] It is an object of the invention to provide a simple way of checking the

[0008] To enable the functionality of an image processing system for workpiece recognition.

[0009] This object is achieved according to the invention by a method according to claim 1. The subclaims and the description indicate advantageous variants.

[0010] According to the invention, a method is provided for checking the calibration of an image processing system of a sheet metal processing machine. The sheet metal processing machine is, in particular, a flatbed machine. The sheet metal processing machine can be a punching machine. The sheet metal processing machine is preferably a laser cutting machine.

[0011] At least one camera of the image processing system captures a

[0012] Removal area for machined workpieces. The camera can be attached to a housing of the sheet metal processing machine. The removal area can be formed on a pallet changer of the sheet metal processing system. The pallet changer particularly simplifies the removal of machined workpieces and preferably also the insertion of blanks to be machined. The removal area is typically flat. The removal area can extend more than one meter, in particular more than two meters, in the plane of its extension.

[0013] According to the invention, the method comprises the following steps

[0014] A) Arranging a calibration plate of predefined shape in the sampling area,

[0015] B) Taking a picture of the calibration plate,

[0016] C) Check whether properties of the image in different test sections of the sampling area match the stored properties of the calibration plate and output the test result.

[0017] The predefined shape of the calibration plate creates a standard for comparison with the image processing system. The calibration plate can have a predefined geometry, particularly with regard to shape and / or dimensions. Alternatively or additionally, the calibration plate can have a predefined surface finish, particularly with regard to color and / or pattern.

[0018] Step A) enables the image processing system to capture an image of the calibration plate in step B). The image is then analyzed by the image processing system. In particular, properties of the test plate are extracted from the captured image. The exact characteristics of these properties depend on a previous calibration of the image processing system. By comparing the image with the stored properties of the calibration plate, it can be checked in step C) whether the image processing system is correctly calibrated so that it functions reliably and can be used, in particular, for workpiece recognition. In particular, it can be checked whether properties extracted from the image lie within a predefined tolerance range relative to predefined reference values ​​for the calibration plate. By examining different test sections in step C), the calibration can be advantageously checked for the entire sampling area.The test sections can be located, for example, at corners of the sampling area and in its center.

[0019] The test result is output, for example, stored in a memory of the processing device and / or transmitted to the operating personnel, preferably by displaying it on a display device. If the test in step C) shows that the image processing system is correctly calibrated, the operating personnel can assume that the image processing system is functioning as intended. The image processing system can then be used in sheet metal processing, in particular to assist the operating personnel in absorbing the processed workpieces.

[0020] A plurality of measuring segments of a predefined shape can be formed on the calibration plate. The measuring segments can each have a predefined geometry, in particular with regard to shape and / or dimensions. Alternatively or additionally, the measuring segments can each have a predefined surface quality, in particular with regard to color and / or pattern. In step C), it can be checked for the different test sections whether properties of the image match the stored properties of the measuring segment of the calibration plate assigned to one of the test sections. By using a calibration plate with multiple measuring segments, a larger sub-area of ​​the sampling area, preferably the entire sampling area, can be prepared for checking the calibration by arranging the calibration plate once in the sampling area.

[0021] Preferably, however, the calibration plate is positioned sequentially in the different test sections, and steps B) and C) are repeated for each of the different test sections. In other words, a single (smaller) calibration plate is used multiple times to check the calibration of the image processing system across the entire sampling area. This simplifies handling of the calibration plate and, in particular, can allow the calibration plate to be positioned manually (without mechanical assistance) in the respective test sections.

[0022] Preferably, the image processing system displays the test sections. This makes it easier to position the calibration plate in the different test sections.

[0023] The image processing system may include a display device for displaying the image captured by the camera. Preferably, the display device displays a transformation of the image into a plan view of the sampling area. This simplifies a comparison of the situation in the sampling area with the information displayed on the display device.

[0024] Particularly preferably, the image processing system displays the test sections (together with the image of the sampling area or the transformation of the image) on the display device. The test sections or their boundaries can be overlaid on the displayed image. Thus, by simply looking at the display device, it is possible to determine whether the calibration plate is correctly positioned.

[0025] Alternatively or additionally, the image processing system can display the test sections in the sampling area. For example, lines marking the boundaries of the test sections can be projected into the sampling area. This further simplifies the correct positioning of the calibration plate.

[0026] Typically, the image processing system is recalibrated if the test in step C) shows that the properties of the image do not match the stored properties of the calibration plate. In this respect, the invention also relates to a method for calibrating an image processing system of a sheet metal processing machine. Inadequate calibration is thus corrected. A routine for calibrating the image processing system can be stored in a control device of the sheet metal processing machine or the image processing system. The calibration of the image processing system can be carried out, for example, as described in WO 2016 / 005159 A2. In particular, CAD data of the calibration plate can be aligned with the captured image. This can be done manually by comparing it with a displayed image. Alternatively, this can be done automatically; displaying the image is then not necessary.

[0027] Preferably, the calibration plate is measured by the image processing system. In this process variant, the calibration plate or the measuring segments have predefined dimensions. To measure the calibration plate, edges can be extracted from the acquired image and their distance in the image determined. Accurate measurement of dimensions across the entire sampling area is particularly important for the functionality of the image processing system.

[0028] In an advantageous method variant, a brightness value is determined for each of the different test sections in the recorded image of the calibration plate. Preferably, a check is carried out to determine whether a maximum deviation of the brightness values ​​for the different test sections exceeds a predefined limit. The limit can be specified absolutely or, preferably, as a relative deviation (of, for example, a maximum of 10 percent) between the individual brightness values. The result of this test can be output, for example, stored in a memory of the processing device and / or transmitted to operating personnel, preferably by display on a display device. This method variant allows the lighting situation perceivable by the image processing system to be recorded and evaluated.In particular, due to the camera's location at the side of the sampling area, the effective brightness for the image processing system or camera may differ from the brightness measurable at individual points in the sampling area using conventional brightness measuring devices. Therefore, whether the lighting situation allows for reliable use of the image processing system often cannot be assessed using separate brightness measuring devices.

[0029] Particularly preferably, the lighting of the sampling area is changed when the maximum deviation of the brightness values ​​exceeds the limit value. In particular, the lighting is adjusted so that the deviations of the brightness values ​​perceived by the image processing system remain below the limit value. Thus, a homogeneous brightness can be established throughout the entire sampling area for the image processing system. This increases the reliability of the image processing system.

[0030] Further features and advantages of the invention will become apparent from the description, the claims, and the drawings. According to the invention, the above-mentioned and further-described features can be used individually or in combination in any convenient way. The embodiments shown and described are not intended to be exhaustive, but rather are exemplary in nature for describing the invention.

[0031] Detailed description of the invention and drawing

[0032] The invention is illustrated in the drawing and described using exemplary embodiments. They show:

[0033] Fig. 1 shows a sheet metal processing machine with an image processing system which detects a removal area for machined workpieces, in a schematic perspective view;

[0034] Fig. 2 shows a schematic flow diagram of a method according to the invention for checking the image processing system; Fig. 3 shows a schematic representation of an image of the removal area displayed on a display device of the image processing system, wherein a calibration plate is arranged in a first test section;

[0035] Fig. 4 the image of Figure 3, indicating that properties of the calibration plate were correctly detected;

[0036] Fig. 5 is a schematic representation of an image of the sampling area displayed on the display device of the image processing system, wherein the calibration plate is arranged in a second test section in which its properties were not correctly recognized;

[0037] Fig. 6 is a schematic representation of an image of the sampling area displayed on the display device of the image processing system, indicating that properties of a calibration plate were correctly detected in various test sections and that the brightness perceived by the image processing system in the test sections meets a predefined criterion.

[0038] Figure 1 shows a sheet metal processing machine 10, in this case a flatbed laser cutting machine. The sheet metal processing machine 10 has a pallet changer 12, on which a removal area 14 for machined workpieces 16 is formed. The workpieces 16 can be removed (absorbed) from the removal area 14, in particular manually, and fed for further processing or use.

[0039] The sheet metal processing machine 10 further comprises an image processing system 18. The image processing system 18 comprises at least one camera 20 and a display device 22. In addition, the image processing system 18 or the sheet metal processing machine 10 can comprise a control device (not shown in detail). The camera 20 is directed at the removal area 14, see the dashed lines emanating from the camera 20. Here, the camera 20 captures the removal area 14 obliquely from above. An image of the removal area 14 recorded by the camera 20 can be displayed on the display device 22, in particular wherein the image is transformed for display into a top view (bird's eye view) with a view of the removal area 14 from vertically above.

[0040] The image processing system 18 provides support for the manual sorting of the machined workpieces 16. For this purpose, the workpieces 16 are identified in the image captured by the camera 20, and instructions for handling the individual workpieces can be displayed on the display device 22. Workpiece recognition requires that the image processing system 18 is correctly calibrated.

[0041] A calibration plate 24 is used to check the image processing system 18, in particular to check its calibration. The calibration plate 24 has a predefined shape. Here, the calibration plate 24 is designed as a cuboid with known dimensions. Furthermore, the calibration plate 24 has a color and, if appropriate, a pattern on its surface (not shown in detail in Figure 1).

[0042] To check the calibration of the image processing system 18, the calibration plate 24 is arranged in a first test section of the removal area 14 in a step 102 (see Figure 2). The camera 20 captures an image of the removal area 14 with the calibration plate 24 (see step 104). The captured image can be displayed on the display device 22 in a step 105 (see Figure 3). Preferably, boundaries 26 of the test sections 28 are also superimposed into the displayed image.

[0043] In a step 106, a check is carried out, for example by the appropriately programmed control device, to determine whether properties of the calibration plate 24 extracted from the recorded image correspond to the stored reference values ​​of these properties within a predetermined tolerance. For example, the calibration plate 24 can be measured by image analysis, and the result of this measurement can be compared with the known dimensions of the calibration plate 24. In a step 108, the result of the test can be output. In particular, the result of the test can be displayed on the display device 22. In Figure 4, a (preferably green) check mark indicates that the properties of the calibration plate 24 were correctly identified in the first test section 28.

[0044] For the calibration plate 24 arranged in a respective test section 28, a brightness value can also be determined from the recorded image in a step 112.

[0045] After the image of the sampling area 14 has been captured and evaluated with the calibration plate 24 arranged in the first test section 28, the calibration plate 24 is arranged in a further test section 28. The previously described steps are repeated accordingly. In this way, all test sections 28 are processed one after the other.

[0046] For example, the test in step 106 may reveal for a second test section 28 that the properties extracted from the captured image do not correspond sufficiently accurately to reference values ​​for the calibration plate 24. This is output in step 108. In Figure 5, the incorrect recognition of properties of the calibration plate in the second test section 28 is indicated by a (preferably red) cross.

[0047] Since the test of the image processing system 18 in the second test section 28 detected inadequate function, the image processing system 18 is recalibrated in a step 110. After successful calibration, the properties of the calibration plate 24 are correctly recognized in all test sections 28. This is indicated in Figure 6 by check marks in the various test sections 28.

[0048] The brightness values ​​for the calibration plate 24 in the different

[0049] Test sections 28 can be checked for deviations from one another in a step 114. The result of this check is output in a step 116. For example, Figure 6 displays a brightness symbol 30 in the shape of a sun. The check mark indicates a sufficiently small deviation between the brightness values; an impermissibly large deviation, on the other hand, could be indicated by a cross. If the captured image is not sufficiently homogeneously illuminated across the entire sampling area 14, the illumination of the sampling area can be corrected in a step 118.

[0050] In summary, the invention relates to a method for testing an image processing system. A calibration plate is placed at different locations in a removal area of ​​a sheet metal processing machine. An image is captured at each location and checked to determine whether the properties extracted from the image match known reference values.

[0051]

[0052] Sheet metal working machine 10

[0053] Pallet changer 12

[0054] Withdrawal area 14

[0055] Workpieces 16

[0056] Image processing system 18

[0057] Camera 20

[0058] Display device 22

[0059] Calibration plate 24

[0060] Borders 26

[0061] Test section 28

[0062] Brightness symbol 30

[0063] Arranging 102 a calibration plate in a test section

[0064] Taking a picture 104

[0065] Show 105 of the image

[0066] Check 106 whether properties of the calibration plate extracted from the image match known reference values

[0067] Output 108 of the test result from step 106

[0068] Calibration 110 of the image processing system

[0069] Determining 112 a brightness value

[0070] Check 114 the brightness values ​​for deviations from each other

[0071] Output 116 of the test result from step 114

[0072] Changing 118 the lighting of the sampling area

Claims

Claims: Method for checking the calibration of an image processing system (18) of a sheet metal working machine (10), in particular a flatbed machine, preferably a laser cutting machine, wherein at least one camera (20) of the image processing system (18) detects a removal area (14) for machined workpieces (16), comprising the steps A) Arranging (102) a calibration plate (24) of a predefined shape in the sampling area (14), B) Taking (104) an image of the calibration plate (24), C) Checking (106) whether the properties of the image in different test sections (28) of the sampling area (14) correspond to stored properties of the calibration plate (24) and outputting (108) the result of the test. Method according to claim 1, characterized in that several measuring segments of a predefined shape are formed on the calibration plate (24), and that in step C) it is checked for the different test sections (28) whether the properties of the image correspond to stored properties of the measuring segment of the calibration plate (24) assigned to each of the test sections (28). Method according to claim 1, characterized in that the calibration plate (24) is arranged successively in the different test sections (28) and steps B) and C) are repeated for each of the different test sections (28).

4. Method according to one of the preceding claims, characterized in that the image processing system (18) displays the test sections (28).

5. Method according to one of the preceding claims, characterized in that the image processing system (18) is a Display device (22) for displaying the image captured by the camera (20).

6. Method according to claim 5, characterized in that a transformation of the image into a top view of the extraction area (14) is displayed on the display device (22).

7. Method according to claim 4 and according to claim 5 or 6, characterized in that the image processing system (18) displays the test sections (28) on the display device (22).

8. Method according to claim 4 or 7, characterized in that the image processing system (18) displays the inspection sections (28) in the sampling area (14).

9. Method according to one of the preceding claims, characterized in that the image processing system (18) is recalibrated if the test in step C) shows that the properties of the image do not match the stored properties of the calibration plate (24).

10. Method according to one of the preceding claims, characterized in that the calibration plate (24) is separated from the The image processing system (18) is measured. A method according to any of the preceding claims, characterized in that a brightness value is determined for each of the different test sections (28) in the recorded image of the calibration plate (24). A method according to claim 11, characterized in that it is checked whether a maximum deviation of the brightness values ​​for the different test sections (28) exceeds a predefined limit value, and that the result of the test is output. A method according to claim 12, characterized in that the The illumination of the sampling area (14) is changed when the maximum deviation of the brightness values ​​exceeds the limit value. A method according to one of the preceding claims, characterized in that the sampling area (14) is located at a Pallet changer (12) of the sheet metal processing plant (10) is trained.

Images