Large-area 3D printing quality diagnosis apparatus, diagnosis method, and 3D printer employing same

Abstract

The present invention relates to an apparatus and a method required to analyze and diagnose the quality of 3D printing having a large area in real time, and to a 3D printer employing the method. The 3D printer of the present invention comprises a plurality of imaging apparatuses for monitoring the lamination and solidification states and further comprises a projection apparatus that forms an image pattern including feature points for use when stitching a plurality of captured images on the upper surface of a job box to generate a stitching transformation parameter required when stitching images identified using the plurality of imaging apparatuses, or projects a specific pattern on the upper surface of the job box or a recoated surface.

Description

Large-area 3D printing quality diagnosis device, diagnosis method, and 3D printer applying the same

[0001] The present invention relates to a device and a method for diagnosing the quality of work in a large-area 3D printing operation, and more specifically, to a device and method necessary for analyzing and diagnosing the quality of a large-area 3D printing in real time, and a 3D printer applying the method.

[0002] 3D printing is a technology that creates a final 3D shape by repeatedly layering the material to be produced one thin layer at a time until the desired height is reached. Recently, as 3D printing has been gradually expanded from prototype production to mass production or serial production, the trend of larger 3D printers is accelerating.

[0003] Conventional 3D printers stack one layer at a time, and if a defect occurs, it results in the inclusion of defects inside the product to be manufactured. If a printing quality problem is identified after printing is complete, the entire output may have to be discarded. In addition, due to the nature of stacking equipment that prints layer by layer, analysis and diagnosis must be performed on a layer-by-layer basis for printing quality control. However, currently, only simple confirmations such as flatness diagnosis by simple sensing such as level sensors are performed, resulting in many stacking quality problems.

[0004] Meanwhile, the applicant of the present invention has proposed a method for identifying a layering quality problem in real time through Patent Publication No. 10-2023-0059214, but a new method needs to be developed for quality diagnosis during large-area 3D printing work, and additionally, a device and method for diagnosing the condition of the recoated surface for each layer and the condition quality of the shape-output surface in real time during large-area 3D printing work are needed.

[0005] The present invention proposes a device configuration and method for merging image information obtained from two or more overhead cameras or image sensors into a single high-quality image in a system that uses two or more overhead cameras or image sensors to diagnose the quality of layer-by-layer recoating or printing in a large-area 3D printer.

[0006] To be more specific, for large-area response, the use of two or more overhead cameras or image sensors is essential. In the stitching process of acquiring two or more images and merging them into a single image, feature points must be extracted from each image, the feature points must be matched, and then each image must be transformed such as rotated, translated, scaled, and tilted using the feature point matching in the overlapping area. In the case of 3D printers, the stitching reliability is not high because the obtained image is too simple for extracting a large number of feature points from the job box surface or the recoating surface. In a real-time printing quality diagnosis system for a large-area 3D printer, we propose a device configuration and method that enable highly reliable stitching.

[0007] In order to achieve the above-described object, one embodiment of the present invention proposes a 3D printer including a build plate capable of moving in a vertical direction and on which metal or non-metallic powder is solidified and stacked; a job box capable of receiving a stacked result while the build plate is lowered by a stacking height; and a recoating device for applying powder onto the build plate; wherein the 3D printer further includes a plurality of imaging devices for monitoring a stacking and solidification state, and a control unit for monitoring the application, stacking, or solidification state of the powder by stitching images recognized through the plurality of imaging devices.

[0008] The large-area 3D printing quality diagnosis device, diagnosis method, and 3D printer applying the same of the present invention configure a system capable of highly reliable stitching in the process of combining two or more images obtained from a large-area 3D printer equipped with two or more cameras or image sensors into one image, thereby enabling analysis and diagnosis of the stacking quality of each layer with high reliability.

[0009] In addition, during the 3D printing process, quality issues with the output can be identified in real time, and if the identified quality issues can be resolved immediately by the equipment itself, the quality of the final 3D printed output can be improved by controlling real-time automatic actions according to the issue.

[0010] FIG. 1 is a perspective view of a 3D printer according to an embodiment of the present invention.

[0011] FIG. 2 is a conceptual diagram showing a plurality of imaging devices arranged on the upper surface of a build plate according to one embodiment of the present invention.

[0012] FIG. 3 is a conceptual diagram illustrating a plurality of imaging devices capturing images of the upper surface of a build plate according to one embodiment of the present invention.

[0013] Figure 4 is a conceptual diagram showing an example of overlapping photographing areas on the upper surface of a build plate.

[0014] Figure 5 is a conceptual diagram showing that an image pattern including feature points is projected onto an overlap area through a projection device.

[0015] Figure 6 is a conceptual diagram showing an example of an image pattern formed on the upper surface of a jab box.

[0016] Figure 7 is a conceptual diagram showing an example of an image pattern projected onto the upper surface of a jab box or a recoated surface.

[0017] FIGS. 8 and 9 are conceptual diagrams showing examples of a pattern for image stitching being laminated and formed on the upper surface of a build plate according to another embodiment of the present invention.

[0018] FIG. 10 is a flowchart illustrating a method of performing a lamination process by stitching projected image patterns according to one embodiment of the present invention.

[0019] FIG. 11 is a flowchart showing a method of performing a lamination process by forming an image pattern of a pattern for image stitching formed on an upper surface of a build plate and then stitching the image pattern according to another embodiment of the present invention.

[0020] In order to achieve the above-described object, the present invention provides a 3D printer comprising: a build plate capable of moving in a vertical direction, on which metal or non-metallic powder is solidified and stacked; a job box capable of receiving a stacked result while the build plate is lowered by a stacking height; and a recoating device for applying powder onto the build plate; wherein the 3D printer further comprises a plurality of imaging devices for monitoring a stacking and solidification state, and a control unit for monitoring the application, stacking, or solidification state of the powder by stitching images recognized through the plurality of imaging devices.

[0021] In addition, the plurality of imaging devices are arranged so that their imaging areas overlap each other, and the control unit derives a stitching transformation parameter for stitching a plurality of images from an image pattern formed or projected in the overlapping area, and the stitching transformation parameter is characterized in that it is used in quality diagnosis for multiple stacking steps.

[0022] At this time, a projection device for projecting a specific pattern onto the upper surface or recoated surface of the above-mentioned box may be further included.

[0023] Additionally, the projection device projects an image pattern including a feature point onto the overlapping area.

[0024] At this time, the control unit can laminate a pattern for image stitching on the build plate to stitch images recognized through a plurality of imaging devices on a specific layer.

[0025] Additionally, the pattern for image stitching can be formed on a blank layer that does not contain an image to be output before printing begins or during the printing process.

[0026] Additionally, the pattern for image stitching may be formed in an area where the imaging areas of the plurality of imaging devices overlap each other.

[0027] In addition, the control unit is characterized in that it stores stitching transformation parameters required to combine images into one through image stitching work before or during printing, and periodically updates the stitching transformation parameters.

[0028] In addition, the control unit is characterized in that it stores the stitching conversion parameter values ​​in a storage within the equipment or a cloud storage, etc.

[0029] A method for diagnosing the quality of 3D printing according to one embodiment of the present invention relates to a method for diagnosing the printing quality of a 3D printer, comprising: a build plate capable of moving up and down, on which metal or non-metal powder is solidified and stacked; a job box capable of receiving a stacked result while the build plate is lowered by a stacking height; a recoating device for applying powder onto the build plate; a plurality of imaging devices for monitoring a stacking and solidification state; and a projection device for projecting a specific pattern onto an upper surface or a recoated surface of the job box; the method comprising: a step of arranging imaging areas of the plurality of imaging devices to overlap each other; a step of projecting an image pattern including a feature point onto the overlapping area through the projection device; a step of deriving a stitching transformation parameter for stitching a plurality of images recognized through the plurality of imaging devices from the image pattern projected onto the overlapping area; a step of acquiring an individual imaging image of each imaging area through the plurality of imaging devices; A step of stitching captured images obtained through the plurality of imaging devices for each stacked layer; and a step of monitoring the application, stacking, or solidification state of the powder through the stitched image.

[0030] In another embodiment, a method for diagnosing the printing quality of a 3D printer, comprising: a build plate that is movable in a vertical direction and on which a metallic or non-metallic powder is solidified and stacked; a job box that can receive a stacked result while the build plate is lowered by a stacking height; a recoating device for applying powder onto the build plate; and a plurality of imaging devices for monitoring a stacking and solidification state, the method comprising: a step of arranging imaging areas of the plurality of imaging devices to overlap each other; a step of forming a pattern for image stitching on the build plate; a step of deriving a stitching conversion parameter for stitching a plurality of images recognized through the plurality of imaging devices from an image pattern projected on the overlapping area; a step of acquiring individual captured images of each imaging area through the plurality of imaging devices; a step of stitching the captured images acquired through the plurality of imaging devices for each stacked layer; and a step of monitoring a state of applying, stacking, or solidifying the powder through the stitched image.

[0031] At this time, a step of storing stitching transformation parameters required to combine images into one through image stitching work before or during printing and periodically updating the stitching transformation parameters may be further included.

[0032] Additionally, the step of storing the stitching transformation parameters in a storage within the device or in a cloud storage may be further included.

[0033] Specific embodiments and features of the present invention are described in detail with reference to the attached drawings. However, the present invention is not limited to the drawings and specific embodiments, but can be implemented in various different forms. These embodiments are provided solely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention. The present invention is defined solely by the claims. Like reference numerals designate like elements throughout the specification.

[0034] When describing embodiments of the present invention, detailed descriptions of known functions or configurations will be omitted if they are deemed to unnecessarily obscure the gist of the invention. Furthermore, the terms described below are defined in light of their functions in the embodiments of the present invention and may vary depending on the intent or custom of the user or operator. Therefore, their definitions should be based on the overall content of this specification.

[0035]

[0036] Hereinafter, with reference to the attached drawings, a large-area 3D printing quality diagnosis device and diagnosis method according to an embodiment of the present invention and a 3D printer (1000) using the same will be described.

[0037]

[0038] First, FIG. 1 is a perspective view of a 3D printer (1000) according to an embodiment of the present invention. Referring to FIG. 1, the 3D printer (1000) according to an embodiment of the present invention includes a build plate (100) that can move up and down and on which metal or non-metallic powder is solidified and stacked; a job box (200) that can receive a stacked result when the build plate (100) is lowered by the stacking height; and a recoating device (300) for applying powder onto the build plate (100). The 3D printer (1000) is formed by further including a plurality of imaging devices (400) for monitoring the stacking and solidification states, and a control unit (600) for monitoring the application, stacking, or solidification states of the powder by stitching images recognized through the plurality of imaging devices (400).

[0039] A plurality of imaging devices (400) are arranged so that the imaging areas (410) overlap each other, and the control unit (600) derives stitching transformation parameters for stitching a plurality of images from the image pattern (510) formed or projected on the overlapping area (420), and the stitching transformation parameters are used in quality diagnosis for various stacking steps.

[0040] That is, a 3D printer (1000) according to an embodiment of the present invention is configured to include a build plate (100) that serves as a reference for a surface on which lamination is performed, a job box (200) that can contain a lamination result when the build plate (100) is lowered by the lamination height, and two or more imaging devices (400) installed to analyze and diagnose the quality of a large-area lamination surface. At this time, the imaging devices (400) may use various image extraction devices including, for example, an overhead camera or an image sensor.

[0041]

[0042] Meanwhile, another 3D printer (1000) according to one embodiment of the present invention may additionally include a projection device (500) for projecting a specific image pattern (510) onto the upper surface of the job box (200), and it is also possible to laminate and form a specific image pattern (520) onto the upper surface of the job box (200) instead of the projection device (500).

[0043]

[0044] FIG. 2 is a conceptual diagram illustrating a plurality of imaging devices (400) arranged on the upper surface (110) of a build plate according to an embodiment of the present invention, and FIG. 3 is a conceptual diagram illustrating a plurality of imaging devices (400) capturing an image of the upper surface (110) of a build plate according to an embodiment of the present invention. Referring to FIGS. 2 and 3, in the case of a large-area 3D printer (1000), since the stacked quality imaging area (410) is wide, two or more imaging devices (400) are required, and since images obtained from two or more cameras or image sensors must be combined into one image, a process of combining them (stitching) is required. In the present invention, in order to ensure good stitching, it is proposed to form a section in which the areas visible from two cameras or images overlap each other.

[0045]

[0046] FIG. 4 is a conceptual diagram showing an example of overlapping shooting areas on the upper surface (110) of the build plate. Referring to FIG. 4, when four cameras or image sensors are used, the overlapping areas appear in a cross shape in two-dimensional space. At this time, even if two or more cameras or image sensors are installed perfectly, a stitching process through conversion of each image is absolutely necessary to perfectly combine multiple images into one image. In order to increase the reliability of stitching multiple images, many feature points must be extracted from each image. However, the upper surface (110) of the build plate of the job box (200) or the recoating surface during lamination only obtains relatively simple images for feature point extraction, so there are limitations in highly reliable conversion and stitching.

[0047]

[0048] The present invention proposes a method of utilizing an image pattern (510, 520) having many feature points for highly reliable transformation and image stitching.

[0049] FIG. 5 is a conceptual diagram showing an image pattern (510) including feature points projected onto an overlap area (420) through a projection device (500) according to one embodiment of the present invention, and FIGS. 6 and 7 are conceptual diagrams showing examples in which an image pattern (520) is formed on the upper surface of a job box (200). Referring to FIGS. 5 to 7, in one embodiment of the present invention, in order to provide an image pattern (510) having feature points, a projection device (500) is arranged to project a specific pattern onto the upper surface or recoated surface of the job box (200) as in FIG. 5, and an image pattern (510) suitable for extracting many feature points is projected onto the upper surface or recoated surface of the job box (200).

[0050] FIG. 6 shows an example of a case where four imaging devices (400) are used, in which an image pattern (510) is projected onto four overlapping areas (420), and FIG. 7 shows an example of a case where two imaging devices (400) are used, in which an image pattern (510) for extracting two feature points is projected onto one overlapping area (420).

[0051]

[0052] To explain in more detail the method of adding an image pattern (510) to the upper surface of the job box (200), an image pattern (510, 520) suitable for extracting many feature points is formed on the upper surface of the job box (200), and stitching transformation parameters for stitching transformation are derived before stacking begins. At this time, the pattern is formed to be included in or span an overlapping area among the areas visible to multiple cameras or image sensors, as shown in FIG. 5.

[0053] The pattern is projected to be included in or span an overlapping area among the areas visible to multiple cameras or image sensors, and the projected image pattern (510) is used in a step of extracting image stitching transformation parameters through feature point extraction, and when performing image stitching for each layer in the subsequent stacking step, the derived stitching transformation parameters are used identically to stitch multiple images and convert them in real time. The stitching transformation parameters for stitching transformation are derived before or during stacking, and when performing image stitching for each layer in the subsequent stacking step, the derived stitching transformation parameters are used identically to stitch multiple images and convert them in real time.

[0054]

[0055] FIGS. 8 and 9 are conceptual diagrams illustrating an example in which an image pattern (520) for image stitching is laminated and formed on the upper surface (110) of a build plate according to another embodiment of the present invention. More specifically, in order to stitch images recognized through a plurality of imaging devices (400) in a specific layer, the image pattern (520) for stitching can be laminated and formed on the build plate (100), and stitching conversion parameters are derived through the laminated pattern. The image pattern (520) for stitching can be formed on an empty layer that does not contain an image to be output before printing starts or during the printing process, and can be formed by being spread out over the entire upper surface (110) of the build plate, as in FIG. 8, or the imaging areas (410) of the plurality of imaging devices (400) can be formed intensively in an overlapping area (420) with each other, as in FIG. 9. At this time, it is preferable that the image pattern (520) formed by lamination be configured to have various shapes.

[0056]

[0057] Meanwhile, before printing begins, stitching transformation parameters required for combining multiple images into one through stitching operations can be secured and stored, or stitching transformation parameters required for combining images into one through image stitching operations by projecting a pattern onto a recoating surface during printing can be secured and stored, and the stored stitching transformation parameters are applied identically during stitching to combine images into one to diagnose the quality of each layer during 3D printing, thereby converting the captured image in real time to diagnose the quality.

[0058] In addition, since the conversion stitching conversion parameters may change due to vibration or external factors while the 3D printer (1000) is in use, a process for periodically updating the stitching conversion parameters may be added, and it is desirable to store the stitching conversion parameter values ​​in storage within the equipment or cloud storage so that they are not lost even when the power is turned off or the program is terminated.

[0059]

[0060] FIG. 10 is a flowchart showing a method of performing a lamination process by stitching a projected image pattern (510) according to an embodiment of the present invention. Referring to FIG. 10, a 3D printing quality diagnosis method according to an embodiment of the present invention includes: a build plate (100) that is movable in an up-and-down direction and on which metal or non-metallic powder is solidified and laminated; a job box (200) that can receive a lamination result while the build plate (100) is lowered by a lamination height; a recoating device (300) for applying powder onto the build plate (100); a plurality of imaging devices (400) for monitoring a lamination and solidification state; And a projection device (500) for projecting a specific pattern onto the upper surface or the recoated surface of the job box (200); A method for diagnosing the printing quality of a 3D printer (1000), comprising: a step of arranging the imaging areas (410) of the plurality of imaging devices (400) to overlap each other; a step of projecting an image pattern (510) including a feature point onto the overlap area (420) through the projection device (500); a step of deriving a stitching transformation parameter for stitching a plurality of images recognized through the plurality of imaging devices (400) from the image pattern (510) projected onto the overlap area (420); a step of acquiring an individual captured image of each imaging area (410) through the plurality of imaging devices (400); a step of stitching the captured images acquired through the plurality of imaging devices (400) for each stacked layer; and a step of monitoring the application, lamination, or solidification state of the powder through the stitched image.

[0061]

[0062] FIG. 11 is a flowchart showing a method for performing a lamination process by forming an image pattern (520) of an image stitching pattern formed on an upper surface (110) of a build plate according to another embodiment of the present invention on an upper surface (110) of a build plate and then stitching the image, and with reference to FIG. 11, in another embodiment, a method for diagnosing the printing quality of a 3D printer (1000) including a build plate (100) that is movable in the up-and-down direction and on which a metal or non-metallic powder is solidified and laminated; a job box (200) that can receive a lamination result while the build plate (100) is lowered by the lamination height; a recoating device (300) for applying powder onto the build plate (100); and a plurality of imaging devices (400) for monitoring a lamination and solidification state, the method comprising: a step in which the imaging areas (410) of the plurality of imaging devices (400) are arranged to overlap each other; The method includes: forming an image pattern (520) for stitching on the build plate (100); deriving a stitching conversion parameter for stitching a plurality of images recognized through the plurality of imaging devices (400) from the image pattern (510) projected on the overlap area (420); obtaining individual captured images of each imaging area (410) through the plurality of imaging devices (400); stitching the captured images obtained through the plurality of imaging devices (400) for each laminated layer; and monitoring the application, lamination, or solidification state of the powder through the stitched image.

[0063]

[0064] Meanwhile, a step of storing stitching transformation parameters required to combine images into one through image stitching work before or during printing and periodically updating the stitching transformation parameters may be further included.

[0065] Additionally, the step of storing the stitching transformation parameters in a storage within the device or in a cloud storage may be further included.

[0066]

[0067] The present invention is not limited to the above-described embodiments, and the scope of application is diverse. It goes without saying that anyone with ordinary skill in the art can make various modifications without departing from the gist of the present invention as claimed in the claims.

[0068]

[0069] 1000: 3D printer

[0070] 100: Build Plate

[0071] 110: Build plate top surface

[0072] 200: Job Box

[0073] 300: Recoating device

[0074] 400: Camera

[0075] 410: Shooting area 420: Overlap area

[0076] 500: Projection device

[0077] 510, 520: Image pattern

[0078] 600: Control unit

[0079] The present invention relates to a 3D printer for additive manufacturing and therefore has industrial applicability.

Claims

1. A build plate that can move up and down and on which powder of metal or non-metal is solidified and stacked; A job box capable of accommodating the stacked results as the above build plate is lowered to the stacking height; In a 3D printer, including a recoating device for applying powder on the build plate; The above 3D printer, A plurality of imaging devices for monitoring the lamination and solidification state; and A 3D printer further comprising a control unit for monitoring the application, lamination, or solidification state of powder by stitching images recognized through the plurality of imaging devices.

2. In paragraph 1, The above plurality of imaging devices are arranged so that their imaging areas overlap each other, The above control unit, From the image pattern formed or projected in the above overlapping area, a stitching transformation parameter for stitching multiple images is derived, A 3D printer, characterized in that the above stitching conversion parameters are used for quality diagnosis for multiple stacking steps.

3. In paragraph 2, A 3D printer further comprising a projection device for projecting a specific pattern onto the upper surface of the above-described work box or the recoated surface.

4. In paragraph 3, A 3D printer, wherein the projection device projects an image pattern including feature points onto the overlapping area.

5. In paragraph 2, A 3D printer in which the control unit laminates a pattern for image stitching on the build plate to stitch images recognized through a plurality of imaging devices on a specific layer.

6. In paragraph 5, The pattern for the above image stitching is formed on a blank layer that does not contain an image to be output before printing begins or during the printing process, in a 3D printer.

7. In paragraph 6, A 3D printer in which the pattern for the above image stitching is formed in an area where the imaging areas of the plurality of imaging devices overlap each other.

8. In paragraph 3 or paragraph 5, The above control unit, A 3D printer characterized in that it stores stitching transformation parameters required to combine images into one through image stitching operation before or during printing, and periodically updates the stitching transformation parameters.

9. In paragraph 8, The above control unit, A 3D printer characterized in that the above stitching conversion parameters are stored in a storage within the device or in a cloud storage, etc.

10. A method for diagnosing the printing quality of a 3D printer, comprising: a build plate capable of moving up and down and on which metal or non-metallic powder is solidified and stacked; a job box capable of receiving a stacked result while the build plate is lowered by the stacking height; a recoating device for applying powder onto the build plate; a plurality of imaging devices for monitoring the stacking and solidification state; and a projection device for projecting a specific pattern onto the upper surface of the job box or the recoated surface; A step of arranging the imaging areas of the plurality of imaging devices to overlap each other; A step of projecting an image pattern including feature points onto the overlapping area through the projection device; A step of deriving stitching transformation parameters for stitching a plurality of images recognized through the plurality of imaging devices from an image pattern projected onto the above overlapping area; A step of acquiring individual capture images of each capture area through the plurality of capture devices; A step of stitching the captured images acquired through the plurality of imaging devices for each stacked layer; and A method for diagnosing 3D printing quality, comprising: a step of monitoring the application, lamination, or solidification state of powder through a stitched image; 11. A method for diagnosing the printing quality of a 3D printer, comprising: a build plate that can move up and down and on which metal or non-metal powder is solidified and stacked; a job box that can receive the stacked result while the build plate is lowered by the stacking height; a recoating device for applying powder on the build plate; and a plurality of imaging devices for monitoring the stacking and solidification state; A step of arranging the imaging areas of the plurality of imaging devices to overlap each other; A step of forming a pattern for image stitching on the above build plate; A step of deriving stitching transformation parameters for stitching a plurality of images recognized through the plurality of imaging devices from an image pattern projected onto the above overlapping area; A step of acquiring individual capture images of each capture area through the plurality of capture devices; A step of stitching the captured images acquired through the plurality of imaging devices for each stacked layer; and A method for diagnosing 3D printing quality, comprising: a step of monitoring the application, lamination, or solidification state of powder through a stitched image; 12. In paragraph 10 or 11, A method for diagnosing 3D printing quality, further comprising a step of storing stitching transformation parameters required to combine images into one through image stitching work before or during printing, and periodically updating the stitching transformation parameters.

13. In paragraph 12, A method for diagnosing 3D printing quality, further comprising a step of storing the above stitching conversion parameters in a storage within the equipment or a cloud storage, etc.

Images