Inspection system, method and program

Abstract

To provide an inspection system, its method, and program that can improve inspection efficiency.SOLUTION: An inspection system 60 for moldings formed by laminating powder includes: an acquisition unit 61 to acquire images of a surface of each layer; a specifying unit 63 that specifies a defective part (concavity or protrusion) and a position of the defective part on the surface of a powder layer based on the acquired image; and a determination unit 64 that determines an abnormality when multiple defects occur in a same position in multiple layers in succession.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present disclosure relates to an inspection system, a method, and a program therefor. [Background technology]

[0002] AM (Additive Manufacturing) products are manufactured by layering materials, and non-destructive testing such as X-ray inspections may be used to ensure the quality of AM products.

[0003] In particular, in PBF (Powder Bed Fusion), inspection may be performed during modeling by checking the state of powder placement (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6708792 Summary of the Invention [Problem to be solved by the invention]

[0005] If non-destructive testing such as X-rays is used to inspect AM products, this could be a factor in increasing costs. If images recording the powder application state during the manufacturing process are judged by a human, the number of images required for each manufactured object can reach 20,000, making it difficult to inspect efficiently.

[0006] When making a judgment using image data of the build surface as the powder laying state, as in Patent Document 1, if a defect is detected on only one build surface layer and an abnormality is detected, the number of layers that need to be judged as abnormal may become enormous, which may increase the burden of the confirmation work.

[0007] The present disclosure has been made in consideration of the above circumstances, and aims to provide an inspection system, a method thereof, and a program that can improve the efficiency of inspections. [Means for solving the problem]

[0008] An inspection system according to one aspect of some embodiments of the present disclosure is an inspection system for a shaped object formed by stacking materials, and includes an acquisition unit that acquires an image of the surface of each layer, an identification unit that identifies defects and the positions of the defects on the surface of the layer based on the images, and a determination unit that determines an abnormality when multiple consecutive defects occur in the same position in multiple layers.

[0009] An inspection method according to one aspect of some embodiments of the present disclosure is a method for inspecting a shaped object formed by stacking materials, and includes the steps of acquiring an image of the surface of each layer, identifying defects and positions of the defects on the surface of the layer based on the images, and determining that an abnormality has occurred when multiple consecutive defects occur in the same position in multiple layers.

[0010] An inspection program according to one aspect of some embodiments of the present disclosure is an inspection program for a shaped object formed by stacking materials, and causes a computer to perform the following processes: acquiring an image of the surface of each layer; identifying defects and the positions of the defects on the surface of the layer based on the images; and determining that an abnormality has occurred when multiple consecutive defects occur in the same position in multiple layers. [Effects of the Invention]

[0011] According to the present disclosure, an effect is achieved in that the efficiency of the examination can be improved. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating a schematic configuration of a molding apparatus according to an embodiment of the present disclosure. [Figure 2]1A and 1B are diagrams illustrating an example of forming a modeling layer according to an embodiment of the present disclosure. [Figure 3] 1A and 1B are diagrams illustrating an example of forming a modeling layer according to an embodiment of the present disclosure. [Figure 4] 1A and 1B are diagrams illustrating an example of forming a modeling layer according to an embodiment of the present disclosure. [Figure 5] FIG. 1 is a diagram illustrating an example of a hardware configuration of an inspection system according to an embodiment of the present disclosure. [Figure 6] FIG. 2 is a functional block diagram illustrating functions of an inspection system according to an embodiment of the present disclosure. [Figure 7] FIG. 10 is a diagram illustrating an example of image processing according to an embodiment of the present disclosure. [Figure 8] FIG. 10 is a diagram illustrating an example of an image that has undergone image processing according to an embodiment of the present disclosure. [Figure 9] 10A and 10B are diagrams illustrating images of an example of an abnormality that becomes a defective portion according to an embodiment of the present disclosure. [Figure 10] 10 is a flowchart illustrating an example of a procedure for an abnormality determination process according to an embodiment of the present disclosure. [Figure 11] FIG. 10 is a diagram illustrating an example of a first process according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013] An embodiment of an inspection system, method, and program according to the present disclosure will be described below with reference to the drawings.

[0014] FIG. 1 is a diagram illustrating a schematic configuration of a molding apparatus 10 according to an embodiment of the present disclosure. The surface plate 22 and the pillars 26 are fixed to the base 21. The base 21 is installed so that its upper surface is horizontal. The upper surface of the surface plate 22 is also horizontal. The upper surface of this surface plate 22 serves as a stage, on which powder material is spread to form a modeling layer. A flange-like convex portion 22a is formed around the entire periphery of the upper surface of the surface plate 22. The outer peripheral surface of the convex portion 22a is in contact with the inner surface of the modeling tank 23, and the powder can be held in the space surrounded by the upper surface of the surface plate 22 and the inner surface of the modeling tank 23. The powder material is, for example, a metal material or a ceramic material. In other words, the modeling apparatus 10 forms a model by Powder Bed Fusion (PBF).

[0015] Powder is laid on the forming section 35 of the forming tank 23 to form a powder layer. Then, by irradiating the powder layer with a laser 32, the powder is melted and solidified to form a forming layer (hardened layer). In this embodiment, the laser 32 is described as the heat source that solidifies the powder. The forming tank 23 is movable in the vertical direction (stacking direction) and is raised a certain amount to form and stack the forming layers. Specifically, the support section 24 supports the lower surface of the flange section 23a of the forming tank 23 so that the upper surface of the flange section 23a is horizontal. The support section 24 is connected to the drive section 25 and moves the forming tank 23 in the vertical direction. The drive section 25 is fixed to a pillar 26 and includes, for example, a motor.

[0016] The laser scanner 28 irradiates the powder layer formed in the modeling unit 35 with a laser 32. The laser scanner 28 is supported by a support unit 27 and can scan the laser 32 on a horizontal surface. In other words, it is possible to selectively heat and solidify the powder at any location on the horizontal surface. The laser 32 is generated by a laser oscillator and introduced into the laser scanner 28 via an optical fiber.

[0017] In the modeling unit 35, powder is supplied from the forming unit 41. For example, pipes 51 and 48 are connected to the supply unit 46 via a branch 47, and a pressure reducer 49 reduces the pressure in the supply unit 46 via the pipe 48, and powder is supplied from the storage unit 50 via the pipe 51 to the supply unit 46. The powder is then supplied from the supply unit 46 via the pipe 45 to the distributor 44. The distributor 44 measures the powder and supplies a predetermined amount of powder to the gap 42 in the modeling unit 41. The forming unit 41 then moves horizontally (a direction perpendicular to the vertical direction) to form a powder layer in the modeling unit 35.

[0018] Specifically, when forming the object 37, the drive unit 25 moves the modeling tank 23 upward, which creates a step between the already formed modeling surface and the upper surface of the flange portion 23a of the modeling tank 23. Then, the forming unit 41 moves horizontally, forming a new powder layer on top of the previously formed, solidified modeling layer. At this time, the height of the uppermost powder layer becomes equal to the height of the upper surface of the flange portion 23a. Then, the laser 32 is irradiated to a predetermined area of ​​the powder layer, selectively heating and solidifying it. In this way, modeling layers are formed. By moving the modeling tank 23 upward and repeating the same operation, each modeling layer is stacked.

[0019] The modeling apparatus 10 is further provided with a camera 55. The camera 55 is supported by, for example, the laser scanner 28 or the support unit 27, and is provided near the laser scanner 28. The camera 55 captures an image of the modeling unit 35 to generate an image of the formed modeling surface. The modeling surface is the surface of the layer to be modeled. The captured image is used in the inspection system 60, which will be described later. The specifications of the camera 55 are set based on the desired detection resolution for the defect D. The captured image is, for example, a black and white image. That is, for each pixel of the image, a value ranging from 0 to an upper limit is set as a brightness value. The upper limit is 255 in 8 bits (256 gradations). The closer the brightness value is to 0, the darker the image is displayed, and the closer the brightness value is to the upper limit, the whiter the image is displayed. The defect D may appear white or black.

[0020] Next, an example of forming each modeling layer will be described with reference to FIG. For example, as shown in FIG. 2, once the modeling layers A1, A2, and A3 have been formed, modeling of modeling layer A4 begins, and powder is laid on top of modeling layer A3. This forms a powder layer corresponding to modeling layer A4. Then, beam scanning is performed, and the laid powder solidifies to form modeling layer A4. Then, modeling of the next layer, modeling layer A5, begins. Similarly, powder is laid to form a powder layer for modeling layer A5, and beam scanning solidifies the powder to form modeling layer A5. In this way, modeling layers are stacked to form a model.

[0021] Here, the laser 32 can fix powder within a penetration depth range in the stacking direction. The penetration depth of the laser 32 is the distance (depth) in the stacking direction at which the laser (heat source) 32 can melt and fix the powder. The penetration depth is a predetermined distance L from the surface of the top layer. In this way, the laser 32 can not only fix the powder in the top layer, but also fix powder at deeper positions. For example, the penetration depth (i.e., the predetermined distance L) is several times larger than the thickness of one layer.

[0022] Next, the inspection system 60 in the modeling apparatus 10 will be described. The inspection system 60 inspects the molded object. For example, as shown in FIG. 3, after the formation of the molded layer B1, the molded layer B2, and the molded layer B3, the molding of the molded layer B4 begins, and powder is laid on top of the molded layer B3. Assume that defect D occurs at this time. Defect D is a depression in the surface of the molded surface. Normally, the height of the molded surface is nearly uniform. However, if the surface is abnormal, the height of a portion of the surface may be lower (i.e., a depression). This depression is designated as defect D (abnormal portion). For example, a portion that is lower than the normal layer surface (predetermined height) by a predetermined value or more may be designated as defect D. Specific examples of such an abnormality include powder depression and powder short. Then, when the beam is scanned, defect D becomes a surface defect and remains on the surface of the molded layer. When powder is laid to form molded layer B5 in this state, the powder is laid on the molded layer B5 while filling the surface defect. When the beam is scanned, the penetration depth of the laser 32 is sufficiently deeper than the stacking height of the modeling layer B5, so the powder in the defective portion D of the modeling layer B4 is also fixed while the modeling layer B5 is formed. In other words, the powder defects no longer remain, and the impact on internal quality is suppressed.

[0023] On the other hand, as shown in FIG. 4, suppose that the forming layers C1, C2, and C3 are formed, and defect D occurs consecutively at the same position. Then, powder is laid to form the forming layer C4 above the forming layer C3, and defect D also occurs in the forming layer C4. Then, when beam scanning is performed, defect D becomes a surface defect and remains on the surface of the forming layer. In this state, when powder is laid to form the forming layer C5, the powder is laid on the forming layer C5 while filling the surface defect. Then, when beam scanning is performed, the powder in defect D also solidifies, forming the forming layer C5. However, the powder in defect D, which is not reached by the penetration depth of the laser 32, does not solidify and remains as an internal defect.

[0024] The inspection system 60 is capable of efficiently inspecting the molded object by taking into account the examples shown in Figures 3 and 4. In the examples shown in Figures 3 and 4, a defect in a recess is described as an example, but a defect (abnormal part) D may also be a protrusion. For example, a part that is higher than a predetermined value from the surface (predetermined height) of a normal layer is a protrusion defect D. Specific examples of this abnormality include powder falling and recoater interference. Even if the defect D is a protrusion, the amount of powder applied increases locally, and the part that the penetration depth does not reach may become an internal defect.

[0025] FIG. 5 is a diagram showing an example of the hardware configuration of an inspection system 60 according to this embodiment. 5, the inspection system 60 is a computer system including, for example, a CPU 110, a ROM (Read Only Memory) 120 for storing programs and the like executed by the CPU 110, a RAM (Random Access Memory) 130 that functions as a work area when each program is executed, a hard disk drive (HDD) 140 as a large-capacity storage device, and a communication unit 150 for connecting to a network or the like. Note that a solid-state drive (SSD) may also be used as the large-capacity storage device. These units are connected via a bus 180.

[0026] The inspection system 60 may also include an input unit consisting of a keyboard, a mouse, etc., and a display unit consisting of a liquid crystal display device or the like for displaying data.

[0027] The storage medium for storing the programs executed by the CPU 110 is not limited to the ROM 120. For example, it may be other auxiliary storage devices such as a magnetic disk, a magneto-optical disk, or a semiconductor memory.

[0028] A series of processing steps for realizing the various functions described below are recorded in the form of a program on the hard disk drive 140 or the like, and the CPU 110 reads this program into the RAM 130 or the like and executes information processing and arithmetic operations to realize the various functions described below. The program may be pre-installed in the ROM 120 or other storage medium, provided in a state stored in a computer-readable storage medium, or distributed via wired or wireless communication means. Examples of computer-readable storage media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, and semiconductor memories.

[0029] 6 is a functional block diagram showing functions of the inspection system 60. As shown in FIG. 6, the inspection system 60 includes an acquisition unit 61, a processing unit 62, an identification unit 63, and a determination unit 64.

[0030] The acquisition unit 61 acquires images of the surface of each layer. Specifically, the acquisition unit 61 acquires images of the surface (modeling surface) of the modeling layer being modeled, layer by layer. In other words, inspection is performed on all (or part of) the modeling layers that make up the model. In particular, the images record the powder application state. The image capture timing is preferably performed after the powder is laid but before it solidifies (i.e., before the beam irradiation). This is because, while defects in recesses can be identified both before and after they solidify, defects in protrusions can be identified before they solidify but may be difficult to identify after they solidify because they are smoothed (to a uniform height) by the beam irradiation. In other words, whether the defect is a recess or a protrusion, capturing an image before it solidifies (before the beam irradiation) allows for effective defect identification. However, as long as the powder application state is known, the image may be taken either before or after the powder is solidified by the beam irradiation.

[0031] The processing unit 62 performs image processing on the acquired image. Specifically, the processing unit 62 performs image processing (feature portion extraction) on the acquired image to emphasize the defect portion D. The image processing is at least one of trapezoidal correction, trimming, contrast adjustment, binarization, first differentiation, second differentiation, contour extraction, and noise removal. Other image processing may also be used as long as it emphasizes the defect portion D (makes the defect portion D easier to determine).

[0032] FIG. 7 is a diagram showing an example of image processing. In FIG. 7, image processing proceeds in the direction of the arrow. First, the input image is a captured image. When capturing an image of the printing surface with camera 55, the shooting direction of camera 55 may be tilted relative to the vertical direction. In such cases, the printing surface that appears in the image becomes a trapezoid as shown in FIG. 7. For this reason, trimming and keystone correction are performed as process P1. Trimming is a process that removes parts other than the printing surface from the image. Keystone correction corrects the printing surface that appears trapezoidal to a square (as if it were captured from directly in front).

[0033] Next, contrast adjustment is performed in process P2, and binarization (which may be first or second derivative) is performed in process P3. This makes the defect D stand out, and then contour extraction is performed in process P4. Since noise is also contained in the image in this state, noise removal is performed in process P5. In noise removal, defect D that is smaller than a set size (for example, area) may also be removed. In this way, defect D is emphasized.

[0034] Fig. 8 shows an example of an image in which the defect D is emphasized after image processing. In Fig. 8, the coordinates on the printing surface (i.e., the horizontal plane) are represented by the x-axis and y-axis. In this way, the defect D and its position (coordinates) can be read from the image.

[0035] Fig. 9 shows an image of an example of an abnormality that becomes a defect D. As shown in Fig. 9, each abnormality such as recoater interference, powder falling, powder short circuit, and powder depression appears as a defect D in the image.

[0036] The identifying unit 63 identifies the defects D and the positions of the defects D on the surface of the layer based on the image (the image that has been subjected to image processing). As shown in FIG. 8, the defects D appear in the image. Therefore, the identifying unit 63 identifies each defect D and the coordinates on the printing surface that are the positions of each defect D based on the image. Since the defects D have a range, it is preferable to identify coordinates that fall within this range.

[0037] In this embodiment, image processing is performed on the captured image, but if the defect D and the position of the defect D can be identified without image processing, it is not necessary to perform image processing.

[0038] The determination unit 64 determines an abnormality when multiple consecutive defects D occur at the same position in multiple layers. As mentioned above, even if a defect occurs in a single layer as shown in Figure 3, the defect may be repaired during the formation of the next layer, and therefore, determining an abnormality and performing confirmation work on a single layer would be a heavy workload. On the other hand, when consecutive defects occur, as shown in Figure 4, the penetration depth of the laser 32 may not be enough to cover them, and internal defects may remain. For this reason, the determination unit 64 determines an abnormality when defect D occurs at the same position for more than a predetermined number of layers. In other words, based on the positions of defect D in each layer identified by the identification unit 63, an abnormality is determined when defect D occurs at positions with the same coordinates in more than a predetermined number of layers.

[0039] Specifically, as shown in FIG. 4, if the penetration depth of the laser 32 is a predetermined distance L, and there is unadhered powder in a position that cannot be reached by this predetermined distance L, the area of ​​this powder will become a defect. For this reason, the predetermined number of layers is set so that the stacking distance (distance in the stacking direction) is equal to or greater than the predetermined distance L. In other words, the distance in the stacking direction of the predetermined number of modeling layers is equal to or greater than the predetermined distance L. Then, the determination unit 64 determines that an abnormality may occur if a defect D occurs in the same position for more than the predetermined number of consecutive layers. If it is determined that an abnormality may occur, a warning may be issued to an inspector, etc.

[0040] In the example of FIG. 4, the penetration depth of the laser 32 is a predetermined distance L, and the number of layers stacked at or above the predetermined distance L is four. Therefore, for example, the predetermined number of layers is set to four. If defects occur in the same position on four consecutive layers as shown in FIG. 4, an internal defect will occur, and an abnormality will be detected. Note that even if an internal defect occurs, the predetermined number of layers may be set to correspond to a distance greater than the predetermined distance L, as long as the size is within the allowable range.

[0041] In this way, by performing an abnormality determination based on the defect portions D of multiple layers, it is possible to suppress the determination of an abnormality for defects that will be repaired during the lamination process, thereby making the inspection more efficient. For example, after an abnormality is determined, non-destructive testing may be performed on the required range. In such a case, the inspection range of the non-destructive testing can be suppressed.

[0042] Next, an example of an abnormality determination process performed by the inspection system 60 described above will be described with reference to FIG. 10. FIG. 10 is a flowchart showing an example of the procedure for the abnormality determination process according to this embodiment. The flow shown in FIG. 10 is performed, for example, every time a modeling layer is formed and an image is acquired. Instead of every time an image is acquired, images corresponding to each modeling layer may be accumulated and the process may be performed after modeling, for example.

[0043] First, an image is captured (S101). Next, trimming and keystone correction are performed (S102). Next, contrast adjustment is performed (S103). Next, binarization (first or second differentiation may be performed) is performed (S104). Then, noise removal is performed (S105). In other words, steps S102 to S105 constitute image processing.

[0044] Next, it is determined whether or not a defect D has been detected (S106). For example, if the area of ​​a region in the image that is assumed to be a defect D is equal to or larger than a threshold, this region is determined to be a defect D. Note that the determination method is not limited as long as the defect D can be determined based on the image.

[0045] If no defect D is detected (NO in S106), the image is determined to be normal (S107). If a defect D is detected (YES in S106), the coordinates of the defect D are recorded (S108).

[0046] Next, it is determined whether or not a defect D has occurred at the same coordinates for a predetermined number of consecutive layers (S109). In S109, the determination is made by referring to the record of the coordinates of the defect D for the lower modeling layer for which the determination has been made (processing of S108).

[0047] If no defect D occurs at the same coordinates for a predetermined number of consecutive layers (NO in S109), the determination of an abnormality is suspended (S110).

[0048] If a defect D occurs at the same coordinates for a predetermined number of consecutive layers (YES in S109), it is determined to be abnormal (S111).

[0049] This process can reduce the inspection burden while suppressing deterioration of quality assurance.

[0050] Although the image processing performed by the processing unit 62 has been described above, other image processing may also be performed. For example, by using an image that does not contain the defect portion D as the reference image, the influence of components originally contained in the image can be suppressed. The reference image is, for example, acquired in advance as an image of a normal printing surface that is assumed not to contain the defect portion D. The reference image may simply be an image of the printing unit 35. Below, first processing, second processing, and third processing are described regarding processing using the reference image. These processing are performed, for example, between S102 and S103 in FIG. 10, but can also be performed before S102.

[0051] First, the first process using the reference image will be described. In the first process, the processing unit 62 performs a process of superimposing a black-and-white inverted image of a reference image on a captured image. An example of the first process is shown in FIG. 11. As shown in FIG. 11, a normal image is subjected to trimming and trapezoidal correction to obtain a reference image. Then, a black-and-white inverted image of the reference image is generated. This black-and-white inverted image is superimposed (added) on, for example, the image on which process S102 in FIG. 10 has been executed, to obtain an evaluation image. Process S103 and subsequent processes in FIG. 10 are executed using this evaluation image.

[0052] In this way, by using the black and white inverted image of the reference image, the original brightness variations can be averaged out and the defect D can be emphasized. Note that when the defect D appears black (low brightness), adding the black and white inverted images of the reference image together may result in a brightness value that exceeds the upper limit. In this case, it is possible to display the brightness values ​​of the parts (pixels) that are equal to or greater than the upper limit after synthesis as the upper limit. In this way, the defect D that is displayed in black is emphasized. Note that the parts that are likely to have brightness values ​​that exceed the upper limit after synthesis are parts that correspond to the normal printing surface in the evaluation image, excluding the defect D.

[0053] Next, the second process using the reference image will be described. In the second process, the processing unit 62 performs a process of subtracting the reference image from the captured image. Specifically, the reference image is subtracted from the image on which process S102 has been performed to obtain an evaluation image. Process S103 and subsequent processes in FIG. 10 are performed using this evaluation image.

[0054] In this way, by using the reference image, it is possible to suppress variations in brightness originally contained in the image and emphasize the defect D. Note that when the defect D appears white (high brightness), there is a possibility that the brightness value will be 0 or less when the reference image is subtracted. In this case, it is possible to display the brightness value of parts (pixels) with a brightness value of 0 or less after compositing as 0 (lower limit value). In this way, the defect D displayed as white is emphasized. Note that the parts that may have a brightness value of 0 or less after compositing are parts that correspond to the normal printing surface excluding the defect D in the evaluation image.

[0055] Next, the third process using the reference image will be described. In the third process, both the first process and the second process are performed. That is, a first evaluation image is obtained by superimposing (adding) a black-and-white inverted image of the reference image onto the captured image. This first evaluation image is used to execute processes S103 and subsequent steps in FIG. 10. That is, the defect D is identified based on the first evaluation image. Note that when the defect D appears black (low brightness), adding the black-and-white inverted image of the reference image may result in a brightness value that exceeds the upper limit. In this case, it is preferable to display the brightness value of any portion (pixel) whose brightness value exceeds the upper limit after combination as the upper limit. Note that the portion whose brightness value may exceed the upper limit after combination is a portion of the evaluation image that corresponds to a normal printing surface excluding the defect D. Meanwhile, a second evaluation image is obtained by subtracting the reference image from the image on which process S102 has been executed. Process S103 and subsequent processes in FIG. 10 are executed using this second evaluation image. In this way, the defect D is also identified based on the second evaluation image. Note that if the defect D appears white (high brightness), subtracting the reference image may result in a brightness value of 0 or less. In this case, it is preferable to display the brightness value of any portion (pixel) with a brightness value of 0 or less after synthesis as 0. Note that the portion with a brightness value of 0 or less after synthesis is a portion of the evaluation image that corresponds to a normal modeling surface excluding the defect D.

[0056] Then, the defective portion D and the position of the defective portion D identified based on the first evaluation image are combined with the defective portion D and the position of the defective portion D identified based on the second evaluation image, and the positions (coordinates) of the defective portion D and the defective portion D relative to the originally captured image are recorded. In this way, the third process performs processing corresponding to both the first process and the second process, so that the defective portion D can be further emphasized.

[0057] Even if the image of the modeling surface is free of defects, it has a brightness distribution. The brightness distribution and average brightness may differ depending on the modeling device 10 (especially the camera 55). However, by performing processing using a reference image, it is possible to suppress device dependency and perform defect determination.

[0058] As described above, the inspection system, method, and program according to this embodiment identify the defect D and its position based on an image of the layer surface, and perform an abnormality determination when multiple defect Ds occur consecutively at the same position, thereby reducing the inspection burden for internal defects. For example, if an abnormality determination is performed for each layer, confirmation work is required for each abnormality. However, even if a defect D occurs in a certain layer, if no defect D is formed in the same position in the formation of the next layer, the defect D may be improved. In other words, it is thought that internal quality is affected when defect Ds occur in the same position in multiple consecutive layers. Therefore, if multiple defect Ds occur consecutively at the same position, it is determined to be an abnormality, thereby reducing the confirmation work burden and improving inspection efficiency.

[0059] The predetermined number of layers is set to a number of layers equal to or greater than the predetermined distance L, where L is the distance in the stacking direction at which the powder can be fixed by the heat source. If a defect D occurs consecutively in the same position for more than the predetermined number of layers, the judgment unit 64 judges it to be an abnormality, thereby enabling efficient inspection while suppressing deterioration of internal quality.

[0060] The present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the invention. It is also possible to combine the various embodiments.

[0061] The inspection system, method, and program described in each of the above-described embodiments can be understood, for example, as follows. An inspection system (60) according to a first aspect of the present disclosure is an inspection system for a shaped object formed by stacking materials (powder), and includes an acquisition unit (61) that acquires an image of the surface of each layer, an identification unit (63) that identifies a defect (D) on the surface of the layer and the position of the defect based on the image, and a determination unit (64) that determines that an abnormality has occurred when multiple consecutive defects occur in the same position in multiple layers.

[0062] According to the inspection system disclosed herein, defects and their locations are identified based on images of the surface of a layer, and an abnormality determination is performed when multiple consecutive defects occur at the same location, thereby reducing the inspection burden for internal defects. For example, if an abnormality determination is performed for each layer, confirmation work is required for each abnormality. However, even if a defect occurs in a layer, if the defect is not formed in the same location in the formation of the next layer, the defect may be improved. In other words, if defects occur in the same location in multiple consecutive layers, it is thought that internal quality will be affected. Therefore, if multiple consecutive defects occur in the same location, an abnormality determination can be performed, reducing the confirmation work burden and improving inspection efficiency.

[0063] In the inspection system according to the second aspect of the present disclosure, in the first aspect, the object is formed into layers by fixing powder with a heat source (32), and the judgment unit judges the object to be abnormal when the defective portion occurs consecutively at the same position for a predetermined number of layers or more, and the predetermined number of layers may be set to a number of layers equal to or greater than a predetermined distance (L) that is a distance in the stacking direction over which the powder can be fixed by the heat source.

[0064] According to the inspection system of the present disclosure, the predetermined number of layers is set to a number of layers equal to or greater than the predetermined distance in the stacking direction at which the powder can be solidified by a heat source, and the judgment unit judges that an abnormality has occurred when a defective portion occurs consecutively in the same position for more than the predetermined number of layers, thereby enabling efficient inspection while suppressing deterioration of internal quality.

[0065] The inspection system according to the third aspect of the present disclosure may be configured to include a processing unit (62) that performs image processing on the acquired image to highlight the defective portion in the first or second aspect, and the identification unit may identify the defective portion and its position based on the image that has been subjected to the image processing.

[0066] According to the inspection system according to the present disclosure, image processing is performed to highlight defective portions, thereby making it possible to make defective portion determination more efficient.

[0067] In the inspection system according to the fourth aspect of the present disclosure, in the third aspect, the image processing may include at least one of trapezoidal correction of the surface, trimming of the surface, contrast adjustment, binarization, first differentiation, second differentiation, contour extraction, and noise removal.

[0068] According to the inspection system of the present disclosure, defects can be highlighted by at least one of surface trapezoid correction, surface trimming, contrast adjustment, binarization, first differentiation, second differentiation, contour extraction, and noise removal.

[0069] In the inspection system according to the fifth aspect of the present disclosure, in the third or fourth aspect, the processing unit may use the image that does not include the defective portion as a reference image and perform a process of overlaying a black-and-white inverted image of the reference image on the image.

[0070] According to the inspection system of the present disclosure, an image that does not contain defects is used as a reference image, and a black-and-white inverted image of the reference image is overlaid on the reference image, thereby suppressing the influence of the black-and-white shading originally contained in the image and emphasizing the defects.

[0071] In the inspection system according to the sixth aspect of the present disclosure, in any one of the third to fifth aspects, the processing unit may be configured to use the image that does not include the defective portion as a reference image and perform a process of subtracting the reference image from the image.

[0072] According to the inspection system of the present disclosure, an image that does not contain defects is used as a reference image, and the reference image is subtracted from the image, thereby suppressing the influence of the black and white shading originally contained in the image and emphasizing the defects.

[0073] In the inspection system according to the seventh aspect of the present disclosure, in the third or fourth aspect, the processing unit may be configured to use the image that does not include the defective portion as a reference image, subtract the reference image from the image, set the brightness value of any portion where the brightness value is equal to or less than 0 to 0, add a black-and-white inverted image of the reference image to the image, and set the brightness value of any portion where the brightness value is equal to or greater than an upper limit to the upper limit.

[0074] According to the inspection system of the present disclosure, an image that does not contain defects is used as a reference image, and the reference image is subtracted from the image, and the brightness values ​​of areas where the brightness values ​​are 0 or less are set to 0, thereby enabling the defect areas that appear in a color close to white in the image (brightness values ​​close to the upper limit) to be highlighted.Then, a black-and-white inverted image of the reference image is added to the image, and the brightness values ​​of areas where the brightness values ​​are equal to or greater than the upper limit are set to the upper limit, thereby enabling the defect areas that appear in a color close to black in the image (brightness values ​​close to 0) to be highlighted.

[0075] An inspection method according to an eighth aspect of the present disclosure is a method for inspecting a shaped object formed by stacking materials, and includes the steps of acquiring an image of the surface of each layer, identifying defects and the positions of the defects on the surface of the layer based on the images, and determining that an abnormality has occurred when multiple consecutive defects occur in the same position in multiple layers.

[0076] An inspection program according to a ninth aspect of the present disclosure is an inspection program for a shaped object formed by stacking materials, and causes a computer to execute the following processes: acquiring an image of the surface of each layer; identifying defects and the positions of the defects on the surface of the layer based on the images; and determining that there is an abnormality when multiple consecutive defects occur in the same position in multiple layers. [Explanation of symbols]

[0077] 10: Molding equipment 21: Bass 22: Surface plate 22a: Convex part 23: Modeling tank 23a: Flange part 24: Support part 25: Drive unit 26: Pillar 27: Support part 28: Laser scanner 32: Laser (heat source) 35: Modeling Department 37: Modeled object 41:Formation part 42 :Void 44:Distributor 45: Piping 46: Supply section 47: Branch 48: Piping 49: Pressure reducer 50: Storage section 51: Piping 55: Camera 60: Inspection system 61: Acquisition part 62: Processing section 63: Specific part 64: Judgment section 110:CPU 120:ROM 130:RAM 140: Hard disk drive 150: Communications Department 180: Bus D: Defective part L: predetermined distance

Claims

1. An inspection system for a molded object formed by layering materials, comprising: an acquisition unit for acquiring an image of the surface of each layer; an identifying unit that identifies defects on the surface of the layer and the positions of the defects based on the image; a determination unit that determines that an abnormality has occurred when the defect portion occurs multiple times consecutively at the same position in multiple layers; a processing unit that performs image processing on the acquired image to emphasize the defective portion, the identifying unit identifies the defect and its position based on the image that has been subjected to the image processing; The processing unit performs a process of superimposing a black-and-white inverted image of the reference image on the image, the image not including the defect being a reference image.

2. An inspection system for a molded object formed by layering materials, comprising: an acquisition unit for acquiring an image of the surface of each layer; an identifying unit that identifies defects on the surface of the layer and the positions of the defects based on the image; a determination unit that determines that an abnormality has occurred when the defect portion occurs multiple times consecutively at the same position in multiple layers; a processing unit that performs image processing on the acquired image to emphasize the defective portion, the identifying unit identifies the defect and its position based on the image that has been subjected to the image processing; The processing unit performs processing in which the image that does not contain the defect is used as a reference image, the reference image is subtracted from the image, the brightness value of any portion where the brightness value is below 0 is set to 0, a black-and-white inverted image of the reference image is added to the image, and the brightness value of any portion where the brightness value is above an upper limit value is set to the upper limit value.

3. 1. A method for inspecting a shaped object formed by stacking materials, comprising: acquiring an image of the surface of each layer; identifying defects and locations of the defects in the surface of the layer based on the image; determining that a defect has occurred in a plurality of layers in succession at the same position; a processing step of performing image processing on the acquired image to emphasize the defective portion, the identifying step includes identifying the defect and a position of the defect based on the image that has been subjected to the image processing; The processing step is an inspection method in which the image not including the defect is used as a reference image, and a black-and-white inverted image of the reference image is superimposed on the image.

4. A method for inspecting a molded object formed by stacking materials, comprising: acquiring an image of the surface of each layer; identifying defects and locations of the defects in the surface of the layer based on the image; determining that a defect has occurred in a plurality of layers in succession at the same position; a processing step of performing image processing on the acquired image to emphasize the defective portion, the identifying step includes identifying the defect and a position of the defect based on the image that has been subjected to the image processing; The processing step is an inspection method in which the image that does not contain the defective portion is used as a reference image, the reference image is subtracted from the image, the brightness value of any portion where the brightness value is 0 or less is set to 0, a black-and-white inverted image of the reference image is added to the image, and the brightness value of any portion where the brightness value is equal to or greater than an upper limit is set to the upper limit.

5. An inspection program for a model formed by stacking materials, acquiring an image of the surface of each layer; identifying defects and locations of the defects on the surface of the layer based on the image; a process of determining that a defect has occurred in a plurality of layers in succession at the same position; and performing image processing on the acquired image to highlight the defective portion, the identifying process identifies the defect and its position based on the image that has been subjected to the image processing; The image processing is an inspection program in which the image not including the defect is used as a reference image, and a black-and-white inverted image of the reference image is superimposed on the image.

6. An inspection program for a molded object formed by stacking materials, comprising: acquiring an image of the surface of each layer; identifying defects and locations of the defects on the surface of the layer based on the image; a process of determining that a defect has occurred in a plurality of layers in succession at the same position; and performing image processing on the acquired image to highlight the defective portion, the identifying process identifies the defect and its position based on the image that has been subjected to the image processing; The image processing is an inspection program in which the image that does not contain the defective portion is used as a reference image, the reference image is subtracted from the image, the brightness value of any portion where the brightness value is below 0 is set to 0, a black-and-white inverted image of the reference image is added to the image, and the brightness value of any portion where the brightness value is above an upper limit value is set to the upper limit value.

Images