Manufacturing apparatus and manufacturing method

Abstract

This technology provides advantages for evaluating the state of liquids placed on a substrate. [Solution] A manufacturing apparatus for manufacturing a display is provided, comprising: a dispensing unit that dispenses liquid and places it on a substrate; an imaging unit that images the substrate on which the liquid is placed and acquires an image including image irregularities; and an evaluation unit that inputs the image acquired by the imaging unit into a machine learning model and evaluates the state of the liquid placed on the substrate based on the output of the learning model obtained.

Description

Technical Field

[0001] The present invention relates to a manufacturing apparatus and a manufacturing method.

Background Art

[0002] In a manufacturing apparatus for manufacturing a display, a liquid ejection device (printing device) such as an inkjet drawing device is used. The liquid ejection device includes a stage that holds and moves a substrate, and a nozzle that ejects ink (liquid such as functional ink) onto the substrate.

[0003] In the liquid ejection device, it is necessary to accurately eject ink at a predetermined position on the substrate, for example, in a pixel area. By controlling the movement of the stage (substrate) and the ejection timing of the nozzle, it is possible to continuously eject ink onto the pixel area on the substrate. The pixel area is partitioned by a partition wall called a bank. If the ink deviates from the pixel area, it will cause a deterioration in the quality of the display. Therefore, it is necessary to inspect the ejection state of the ink and maintain the ejection performance of the nozzle with high precision. Therefore, regarding the inspection of the ejection state of the ink, a technique has been proposed in which an image is obtained by imaging the substrate immediately after the ink ejected from the nozzle is placed, and the quality of the ink ejection result (print quality) is determined from such an image (see Patent Document 1).

[0004] The ink placed on the substrate in the liquid ejection device forms a film (functional film) having a uniform thickness (film thickness) through a drying process in a drying device. In the period from the arrangement process of arranging the ink on the substrate to the transition to the drying process (that is, transporting the substrate to the drying device), if natural drying of the ink progresses, the flatness of the surface of the film formed on the substrate is impaired, leading to a deterioration in the quality of the display. Therefore, when the ink is placed on the substrate in the liquid ejection device, it is necessary to transport such a substrate to the drying device in a short time. In addition, techniques for determining defects in images (image data) printed on a printing medium such as paper or displayed on a display have also been proposed (see Patent Documents 2 and 3).

Prior Art Documents

[0005] [Patent Document 1] Japanese Patent Publication No. 2008-225249 [Patent Document 2] Japanese Patent Publication No. 2023-129970 [Patent Document 3] Japanese Patent Publication No. 2019-144245 [Overview of the project] [Problems that the invention aims to solve]

[0006] However, when attempting to inspect the ink ejection state using conventional technology in manufacturing equipment for display devices, the area to be inspected is large, making it difficult to meet the inspection time constraints. For example, when inspecting a substrate on which ink is placed, using high-resolution images results in a large amount of data, requiring a considerable amount of inspection time.

[0007] This invention has been made in view of the problems of the prior art, and its exemplary objective is to provide a technique that is advantageous for evaluating the state of a liquid placed on a substrate. [Means for solving the problem]

[0008] To achieve the above objective, a manufacturing apparatus as one aspect of the present invention is a manufacturing apparatus for manufacturing a display, characterized by comprising: a dispensing unit that dispenses a liquid and places it on a substrate; an imaging unit that images the substrate on which the liquid is placed and acquires an image including image irregularities; and an evaluation unit that inputs the image acquired by the imaging unit into a machine learning model and evaluates the state of the liquid placed on the substrate based on the output of the learning model obtained.

[0009] Further objects or other aspects of the present invention will be revealed by embodiments described below with reference to the accompanying drawings. [Effects of the Invention]

[0010] According to the present invention, for example, it is possible to provide a technique that is advantageous for evaluating the state of a liquid placed on a substrate. [Brief explanation of the drawing]

[0011] [Figure 1] This is a schematic diagram showing the basic configuration of a manufacturing apparatus as one aspect of the present invention. [Figure 2] This is a schematic diagram showing the basic configuration of the imaging unit. [Figure 3] This diagram illustrates the relationship between the state of the liquid placed in the pixel area and the image acquired by the imaging unit. [Figure 4] This diagram illustrates the relationship between the state of the liquid placed in the pixel area and the image acquired by the imaging unit. [Figure 5] This is a flowchart to explain the operation of the manufacturing equipment. [Modes for carrying out the invention]

[0012] The embodiments will be described in detail below with reference to the attached drawings. Note that the following embodiments do not limit the invention as defined in the claims. While the embodiments describe multiple features, not all of these features are essential to the invention, and the features may be combined in any way. Furthermore, in the attached drawings, identical or similar configurations are given the same reference numerals, and redundant descriptions are omitted.

[0013] In this specification and the accompanying drawings, direction is defined by an XYZ coordinate system in which the plane parallel to the surface on which the substrate is placed is the XY plane. The X, Y, and Z axes are orthogonal to each other. The direction along the X axis is called the X direction, the direction along the Y axis is called the Y direction, and the direction along the Z axis is called the Z direction. The Z direction is also the vertical direction, and the X and Y directions are also the horizontal directions.

[0014] In this embodiment, "liquid" refers to a liquid (ink) used to form patterns or films on an object such as a substrate. Furthermore, the components of the "liquid" are not particularly limited, and for example, a liquid containing a solute and a solvent for forming functional elements such as organic EL elements (functional materials such as organic EL materials or quantum dot materials) can be used.

[0015] In this embodiment, the act of dispensing a liquid and placing (applying or supplying) it to an object may be described as "printing." "Printing" also includes, for example, placing a liquid on an object in order to manufacture articles such as functional thin films, functional elements, and three-dimensional objects. The object to which the liquid is placed includes components that serve as a base material for manufacturing articles such as functional thin films, functional elements, and three-dimensional objects, such as substrates.

[0016] Figure 1 is a schematic diagram showing the basic configuration of the manufacturing apparatus 1 as one aspect of the present invention. In this embodiment, the manufacturing apparatus 1 functions as an inkjet drawing apparatus (printing apparatus) that ejects liquid 4, and is embodied as a manufacturing apparatus for manufacturing displays (display panels). The manufacturing apparatus 1 includes a substrate stage 3, a liquid ejection head 5, a structure 6, an imaging unit 10, a processing unit 11, and a control unit 12.

[0017] The substrate stage 3 is a stage that holds and moves the substrate 2 of the display panel. The substrate 2 is appropriately selected from glass substrates, plastic substrates, etc., depending on the product (article) to be manufactured. The substrate 2 is typically a plate-shaped component, but its form is not limited as long as it functions as a base material. For example, the substrate 2 may be a deformable film or a circular substrate. The substrate 2 is provided with pixel areas 7 for forming a large number of display pixels by applying a liquid.

[0018] The liquid ejection head 5 is supported by the structure 6. The liquid ejection head 5 includes a plurality of nozzles (discharge ports) and is a discharge unit that ejects the liquid 4 as droplets toward a predetermined position on the substrate 2. The liquid ejection heads 5 are arranged in each of the X direction and the Y direction. Under the control of the control unit 12, by individually controlling the ejection of each liquid 4 from the liquid ejection head 5, the liquid 4 can be arranged in a predetermined distribution in the pixel area 7 of the substrate 2. Further, the manufacturing apparatus 1 also has a recovery unit (not shown) that performs a cleaning process or the like on the liquid ejection head 5 to recover the ejection characteristics of the liquid ejection head 5.

[0019] The imaging unit 10 is supported by the structure 6 and images the substrate 2 on which the liquid 4 is arranged to acquire an image. The image acquired by the imaging unit 10 includes image unevenness according to the state of the liquid 4 arranged on the substrate 2. As shown in FIG. 2, the imaging unit 10 includes an irradiation unit 101 and an imaging element 102. FIG. 2 is a schematic diagram showing the basic configuration of the imaging unit 10. The irradiation unit 101 and the imaging element 102 may be arranged in plurality according to the size of the substrate 2 that is the imaging target. Further, in FIG. 2, the irradiation unit 101 and the imaging element 102 are arranged on the minus side in the Y direction with respect to the liquid ejection head 5, but may be arranged on the plus side in the Y direction.

[0020] The irradiation unit 101 is arranged facing the surface of the substrate 2 and irradiates the substrate 2 on which the liquid 4 is arranged with light. The imaging element 102 is arranged facing the surface of the substrate 2 and detects light from the substrate 2 on which the liquid 4 is arranged. In the present embodiment, the irradiation unit 101 and the imaging element 102 are arranged such that the light from the irradiation unit 101 illuminates the substrate 2 and the reflected light thereof is detected by the imaging element 102. The light that the irradiation unit 101 irradiates the substrate 2 with may be red light or white light. Further, the imaging element 102 is arranged to detect scattered light generated on the surface of the substrate 2 instead of regular reflected light among the reflected light from the substrate 2. The imaging element 102 is arranged such that the angle θ1 formed by the optical axis of the imaging element 102 and the surface of the substrate 2 is an acute angle.

[0021] The image sensor 102 is an element that includes an imaging surface in which a plurality of pixels are arranged, and is composed of, for example, an area camera or a line camera. In this embodiment, the resolution of the image sensor 102 is coarse (low) with respect to the pixel area 7 of the substrate 2, and is at least equal to or greater than the pixel size of the pixel area 7 of the substrate 2. In other words, the size of each pixel area 7 on the substrate corresponding to each pixel of the display that is projected onto the imaging surface of the image sensor 102 is smaller than the size of one pixel on the imaging surface. Specifically, at least two or more pixel areas 7 on the substrate, preferably three or more pixel areas 7, are projected onto one pixel on the imaging surface of the image sensor 102.

[0022] The processing unit 11 has the function of processing the substrate 2 on which the liquid 4 is placed. The processing unit 11 may be implemented, for example, as a drying device that dries the liquid 4 placed on the substrate 2 under controlled conditions, or as a firing device that sintersects the liquid 4 placed on the substrate 2 under controlled conditions. By drying and sintering the liquid 4 placed on the substrate 2, it becomes possible to control the shape of the surface of the liquid 4. The processing unit 11 may also be implemented as a curing device that irradiates the liquid 4 placed on the substrate 2 with light such as ultraviolet light to cure the liquid 4. Note that the processing unit 11 may not be part of the manufacturing device 1, but may be provided outside the manufacturing device 1 as an external processing device. In this case, the substrate 2 on which the liquid 4 is placed is transported out of the manufacturing device 1 via a transport robot and brought into the external processing device.

[0023] The control unit 12 is composed of a computer (information processing device) including, for example, a CPU and memory. The control unit 12 comprehensively controls each part of the manufacturing apparatus 1 according to a program stored in a memory unit or the like to operate the manufacturing apparatus 1. The control unit 12 controls various processes related to the process of discharging liquid 4 from the liquid discharge head 5 and placing it on the substrate 2, and the process of processing the substrate 2 with the liquid 4 on it in the processing unit 11 to manufacture a display.

[0024] In this embodiment, the control unit 12 functions as an evaluation unit that evaluates the state of the liquid 4 placed on the substrate by the liquid ejection head 5 from the image acquired by the imaging unit 10. As part of the evaluation of the state of the liquid 4 placed on the substrate, the control unit 12 determines, for example, whether the state of the liquid 4 placed on the substrate is good or bad. The control unit 12 may also control the amount of irradiation light, irradiation time and irradiation timing in the irradiation unit 101, and the exposure time and exposure timing in the image sensor 102 when acquiring an image for evaluating the state of the liquid 4 placed on the substrate. Hereinafter, the state of the liquid 4 placed on the substrate may also be referred to as print quality.

[0025] The control unit 12 configures an AI filter that determines the quality of the print from the image acquired by the imaging unit 10. The AI ​​filter configured in the control unit 12 includes a learning model generated by machine learning, using pre-prepared images corresponding to the quality of the print as training data. Therefore, in this embodiment, the control unit 12 determines the quality of the print based on the output of the learning model obtained by inputting the image acquired by the imaging unit 10 into the machine learning learning model. Note that the machine learning calculation unit that determines the parameters of the learning model included in the AI ​​filter may be provided separately from the control unit 12.

[0026] The following describes the determination process for determining the quality of the print, which is performed under the control of the control unit 12 in the manufacturing apparatus 1. As shown in Figure 2, when the liquid 4 discharged from the liquid discharge head 5 is placed in the pixel area 7 of the substrate 2, the substrate stage 3 is moved to position the substrate 2 with the liquid 4 in place at the focal point of the imaging unit 10. The focal point of the imaging unit 10 is determined by the known optical specifications and component arrangement of the image sensor 102. Alternatively, the focal point of the imaging unit 10 may be calculated from the contrast information of the image acquired by the imaging unit 10, and the amount of movement of the substrate stage 3 may be adjusted. For example, the substrate stage 3 may be moved in the Z direction so that the amount of gradation change in brightness obtained by differentiating the brightness value of the image between adjacent pixels is large. In this embodiment, in order to image the entire substrate 2, the substrate stage 3 is moved to the negative side in the Y direction while the imaging unit 10 continuously images the substrate 2. At this time, the control unit 12 controls the irradiation timing in the irradiation unit 101 and the exposure time in the image sensor 102 according to the position of the substrate stage 3 in the Y direction. Furthermore, the area of ​​the substrate 2 imaged by the imaging unit 10 (the area used to determine the quality of the print) may be limited to a portion of the substrate 2, rather than the entire substrate 2.

[0027] Here, we will explain the relationship between the state of the liquid 4 placed in the pixel area 7 of the substrate 2 and the image acquired by the imaging unit 10. Figure 3(a) shows the liquid 4 placed in the pixel area 7 of the substrate 2, the light (irradiated light) 202 irradiated onto the substrate 2 from the irradiation unit 101, and the light (reflected light) 203 reflected from the substrate 2. The illuminated light 202 is monochromatic light. The reflected light 203 includes surface reflected light 203A reflected from the surface of the liquid 4 and back reflected light 203B that has passed through the liquid 4 and been reflected from the back surface. When the surface reflected light 203A and the back reflected light 203B coincide on the surface of the liquid 4, constructive interference of light occurs. Since the interference between the surface reflected light 203A and the back reflected light 203B depends on the thickness of the liquid 4, it is possible to observe minute irregularities on the surface of the liquid 4 as variations in the intensity of the reflected light 203. In this embodiment, the state of the liquid 4 (surface state) is evaluated by detecting reflected light 203 (i.e., interference light between surface reflected light 203A and back reflected light 203B) and acquiring it as an image. Also, as shown in Figure 3(b), when the angle of the irradiating light 202 with respect to the liquid 4 placed in the pixel area 7 of the substrate 2 is shallow, the range of intensity of the reflected light 203 (interference light between surface reflected light 203A and back reflected light 203B) becomes wider. Therefore, it is preferable to position the irradiating unit 101 such that the angle θ2 between the optical axis of the irradiating unit 101 and the surface of the substrate 2 including the pixel area 7 is acute, and position the image sensor 102 such that the angle θ1 between the optical axis of the image sensor 102 and the surface of the substrate 2 is acute. This improves the accuracy of evaluating the state of the liquid 4 placed in the pixel area 7 of the substrate 2, that is, determining the quality of the print. Furthermore, if the liquid 4 has curing properties with respect to ultraviolet light, the curing reaction of the liquid 4 placed in the pixel area 7 can be suppressed by using red light, for example, as the irradiating light 202. In other words, in order to suppress the hardening of the liquid 4 placed in the pixel area 7 when the imaging unit 10 images the substrate 2 and acquires an image, the irradiation unit 101 should irradiate light 202 with a wavelength band different from the wavelength band in which the liquid 4 hardens.

[0028] Referring to Figures 4(a) and 4(b), the case in which white light is used as the light irradiated onto the substrate 2 by the irradiation unit 101 will be explained. Figure 4(a) shows the liquid 4 placed in the pixel area 7 of the substrate 2, the light irradiated onto the substrate 2 from the irradiation unit 101 (irradiated light) 302, and the light reflected from the substrate 2 (reflected light) 303. When the irradiation light 302 is white light, in addition to the phenomena explained with reference to Figures 3(a) and 3(b), the effects of refraction of the irradiation light 302 at each wavelength are superimposed. As shown in Figure 4(a), when the irradiation light 302 enters the interior of the liquid 4, it becomes different light rays 302A and 302B at each wavelength, and each of the light rays 302A and 302B passes through the liquid 4, is reflected on the back surface, and reaches the surface of the liquid 4. These interfere with the light reflected from the surface of the liquid 4 and appear as variations in the intensity of the reflected light 303. As a result, color unevenness (light interference at each wavelength) corresponding to the thickness of the liquid 4 placed in the pixel area 7 of the substrate 2 is detected. Furthermore, as shown in Figure 4(b), it can be seen that the detection sensitivity of color unevenness is high when the angle of the irradiating light 302 to the liquid 4 placed in the pixel area 7 of the substrate 2 is shallow.

[0029] Furthermore, from the viewpoint of more efficiently detecting the state of the liquid 4 placed in the pixel area 7 of the substrate 2, it is preferable that the irradiation unit 101 includes, for example, a diffuser plate that diffuses light, so that the light irradiated onto the substrate 2 is diffused light. By making the light irradiated onto the substrate 2 diffused light and dispersing the orientation angle, it becomes possible to reliably capture unevenness (image unevenness, interference fringes) in the image acquired by the imaging unit 10, as described with reference to Figures 3(a), 3(b), 4(a), and 4(b). Note that multiple irradiation units 101 may be arranged for one imaging area on the substrate (an area to be captured as a single image). This allows light to be irradiated onto the liquid 4 placed in the pixel area 7 of the substrate 2 at multiple angles, enabling efficient detection of unevenness.

[0030] Thus, minute shape changes in the surface state of the substrate 2, particularly the state of the liquid 4 placed in the pixel area 7 of the substrate 2, are detected as visible irregularities when viewed from a macroscopic perspective. Therefore, instead of directly detecting the state of the minute liquid 4 for each pixel in the pixel area 7 of the substrate 2, it is possible to determine the quality of the print by detecting these irregularities. As a result, the imaging unit 10 does not need a high resolution corresponding to the pixel size of the pixel area 7 of the substrate 2, but only needs to have a resolution that can detect irregularities that appear on the substrate at a visually visible period. Consequently, the amount of image data acquired by the imaging unit 10 imaging the entire surface of the pixel area 7 of the substrate 2 is suppressed, and the time required for image processing and determining the quality of the print can be reduced. In addition, the number of imaging units 10 required to image the entire substrate 2 can be reduced. These contribute to space saving, cost saving, cycle time reduction, and improved print quality in the manufacturing equipment 1 for manufacturing displays. Moreover, these effects become more pronounced as the size of the substrate 2 increases, making them a significant advantage.

[0031] The images acquired in this manner are input to a learning model configured as an AI filter in the control unit 12, and the quality of the print is determined based on the output of this learning model. The parameters of the learning model are determined by prior machine learning. As training data, image data containing image irregularities, i.e., poor print quality, and image data without image irregularities, i.e., good print quality, are prepared, and machine learning is performed using these as input data.

[0032] In this embodiment, the amount of data in the image acquired by the imaging unit 10 is kept low, and the image irregularities contained in such images have low reproducibility and do not necessarily have the same shape. When determining the quality of print from such images, it is difficult to set a threshold for determining the quality of print using conventional techniques, such as numerical contrast analysis of image analysis. On the other hand, in this embodiment, as described above, by using an AI filter that includes a machine learning model, it is possible to extract the feature quantities of the image irregularities contained in the image acquired by the imaging unit 10 and determine the quality of print from those feature quantities. Furthermore, the AI ​​filter that includes a machine learning model extracts the image irregularities contained in the image as feature quantities of their shape. Therefore, it is possible to detect image irregularities caused by printing defects without being affected by the electrical wiring patterns or background patterns formed on the substrate 2 (i.e., without treating these patterns as image irregularities).

[0033] Referring to Figure 5, the operation of the manufacturing apparatus 1 will be explained. Here, we will focus on the judgment process that evaluates the state of the liquid 4 placed on the substrate 2, specifically the process that determines whether the print quality is good or bad, and explain the operation of the manufacturing apparatus 1.

[0034] In S501, the liquid 4 is dispensed from the liquid dispensing head 5 and placed in the pixel area 7 of the substrate 2. In S502, the substrate 2 with the liquid 4 placed in the pixel area 7 is imaged by the imaging unit 10 to acquire an image. In this embodiment, the imaging unit 10 images the substrate 2 with the liquid 4 placed in it during the period from when the liquid dispensing head 5 dispenses the liquid 4 and places it in the pixel area 7 of the substrate 2 until the substrate 2 is transported to the processing unit 11. This is a necessary requirement for evaluating the state of the liquid 4 immediately after it is placed on the substrate 2.

[0035] In S503, a determination process is performed to evaluate the state of the liquid 4 placed in the pixel area 7 of the substrate 2 based on the image acquired in S502, that is, to determine whether the print quality is good or bad, and to determine whether the print is good or bad. Specifically, the image acquired in S502 is input to a learning model included in the AI ​​filter configured in the control unit 12, and the quality of the print, i.e., whether the print is good or bad, is determined based on the output of this learning model. If the print is good, the process proceeds to S504; if the print is not good, i.e., if the print is poor, the process proceeds to S505.

[0036] In S504, the substrate 2 with liquid 4 placed in the pixel area 7 is transported to the processing unit 11, where the processing unit 11 processes the substrate 2. The processing of the substrate 2 includes, as described above, a process to dry the liquid 4 placed on the substrate 2, a process to bake the liquid 4 placed on the substrate 2, and a process to harden the liquid 4 placed on the substrate 2.

[0037] In S505, based on the results of the judgment process performed in S503, abnormal areas where image unevenness exists are identified in the image acquired in S502. In S506, a precise judgment is performed on the area of ​​substrate 2 corresponding to the abnormal area identified in S505. However, the area on which the precise judgment is performed is not limited to the area of ​​substrate 2 corresponding to the abnormal area, but may be a part of the area of ​​substrate 2 that includes at least the area of ​​substrate 2 corresponding to the abnormal area, or it may be the entire substrate 2. In the precise inspection, the quality of the print is inspected (judged) for each pixel area 7 (each pixel of the display). For the precise inspection, an image acquired by the imaging unit 10 may be used, or an image acquired by a high-resolution imaging unit other than the imaging unit 10 may be used. Alternatively, a length measuring sensor that measures the height of the object to be measured may be used for the precise inspection.

[0038] In S507, a defective nozzle is identified from among the multiple nozzles of the liquid ejection head 5. Since each pixel area 7 of the substrate 2 and each nozzle of the liquid ejection head 5 are associated with a pre-calculated ejection pattern, a defective nozzle can be identified from the inspection results of the print quality for each pixel area 7.

[0039] In S508, correction processing is performed according to the number of defective nozzles identified in S507. For example, if the number of defective nozzles is small, the correction processing involves changing the discharge pattern (rewriting the discharge data) so that other nozzles are used instead of the defective nozzles. On the other hand, if the number of defective nozzles is large, the correction processing involves replacing the liquid discharge head 5 with a new liquid discharge head. If the number of defective nozzles is neither large nor small, the correction processing involves cleaning the liquid discharge head 5 to restore its discharge characteristics.

[0040] In this embodiment, an imaging unit 10 used to efficiently capture printing defects as unevenness (image unevenness) in relation to evaluating the state of the liquid 4 placed on the substrate 2 was described. Furthermore, an AI filter (learning model) that evaluates the state of the liquid 4 placed on the substrate 2 from images with a small amount of data acquired by the imaging unit 10, specifically determining the quality of the printing, was also described. As a result, in this embodiment, a manufacturing apparatus 1 can be realized that can determine the quality of the state of the liquid 4 placed on the substrate 2 in a manufacturing line (inline) without degrading the quality of the liquid 4.

[0041] Furthermore, according to this embodiment, printing defects caused by poor ejection of the liquid 4 ejected from the liquid ejection head 5 can be accurately determined immediately after the liquid 4 is placed on the substrate 2, that is, before the liquid 4 placed on the substrate 2 dries naturally. The thickness of the liquid 4 placed on the substrate 2 decreases to about 1 / 10 of its original thickness upon drying, making measurement itself difficult. Also, factor analysis becomes difficult after going through each process. Therefore, it is preferable to evaluate the state of the liquid 4 immediately after it is placed on the substrate 2. However, this embodiment may also be applied when the state of the liquid 4 on the substrate is evaluated after the liquid 4 has been placed on the substrate 2 and gone through each process.

[0042] The manufacturing method according to an embodiment of the present invention is suitable for manufacturing a display, such as an organic light-emitting diode (OLED) panel, using a manufacturing apparatus 1. The manufacturing method of this embodiment includes a step of placing a liquid on a substrate using the manufacturing apparatus 1 (placement step). The manufacturing method also includes a step of processing the substrate on which the liquid has been placed (processing step), for example, by drying and firing the liquid, or by curing the liquid, to obtain a substrate on which a film has been formed. Furthermore, the manufacturing method includes other well-known steps for manufacturing a display from the processed substrate (substrate on which a film has been formed) (cooling, dehumidification, dry cleaning, electrode formation, sealing film formation, etc.). The manufacturing method of this embodiment is advantageous over conventional methods in at least one of the performance, quality, productivity, and production cost of the display.

[0043] The disclosures herein include the following manufacturing apparatus and manufacturing methods.

[0044] (Item 1) A manufacturing apparatus for producing displays, A dispensing unit that dispenses liquid and places it on the substrate, An imaging unit that images a substrate on which the liquid is placed and acquires an image including image irregularities, An evaluation unit evaluates the state of the liquid placed on the substrate based on the output of a machine learning model obtained by inputting the image acquired by the imaging unit into the machine learning model, A manufacturing apparatus characterized by having

[0045] (Item 2) The imaging unit includes an imaging surface on which a plurality of pixels are arranged, The size of each pixel area on the substrate corresponding to each pixel of the display projected onto the imaging surface is smaller than the size of a single pixel on the imaging surface. The manufacturing apparatus described in item 1, characterized by the following:

[0046] (Item 3) The manufacturing apparatus according to item 2, characterized in that at least two or more pixel areas on the substrate are projected onto one pixel of the imaging surface.

[0047] (Item 4) The manufacturing apparatus according to any one of items 1 to 3, characterized in that the evaluation unit determines whether the state of the liquid disposed on the substrate is good or bad.

[0048] (Item 5) The manufacturing apparatus according to any one of items 1 to 4, characterized in that the imaging unit acquires the image by detecting interference light between light reflected from the surface of the liquid placed on the substrate and light transmitted through the liquid and reflected from the back surface.

[0049] (Item 6) The system further includes a processing unit that processes the substrate on which the liquid is placed, The imaging unit images the substrate on which the liquid is placed during the period from when the dispensing unit dispenses the liquid and places it on the substrate until the substrate is transported to the processing unit. A manufacturing apparatus as described in any one of items 1 to 5, characterized by the above.

[0050] (Item 7) The manufacturing apparatus according to item 6, characterized in that the processing unit performs a process of drying the liquid disposed on the substrate.

[0051] (Item 8) The manufacturing apparatus according to item 6 or 7, characterized in that the processing unit performs a process of firing the liquid placed on the substrate.

[0052] (Item 9) The manufacturing apparatus according to item 6, characterized in that the processing unit performs a process to harden the liquid disposed on the substrate.

[0053] (Item 10) The imaging unit includes an illumination unit that irradiates light onto the substrate and an image sensor that detects light from the substrate. The angle between the optical axis of the irradiation unit and the substrate, and the angle between the optical axis of the image sensor and the substrate are acute angles. A manufacturing apparatus as described in any one of items 1 to 9, characterized by the above.

[0054] (Item 11) The manufacturing apparatus according to item 10, characterized in that the irradiation unit includes a diffuser plate for diffusing the light irradiated onto the substrate.

[0055] (Item 12) The manufacturing apparatus according to item 10 or 11, characterized in that the irradiation unit irradiates the substrate with light of a wavelength band different from the wavelength band in which the liquid hardens.

[0056] (Item 13) The manufacturing apparatus according to item 10 or 11, characterized in that the irradiation unit irradiates the substrate with white light.

[0057] (Item 14) The manufacturing apparatus according to any one of items 1 to 13, characterized in that the imaging unit acquires the image by detecting scattered light scattered by the substrate from the light emitted from the substrate.

[0058] (Item 15) A step of placing liquid on a substrate using a manufacturing apparatus described in any one of items 1 to 14, A step of processing the substrate on which the liquid is placed, A process for manufacturing a display from the processed substrate, A manufacturing method characterized by having the following:

[0059] The invention is not limited to the embodiments described above, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, claims are attached to disclose the scope of the invention. [Explanation of Symbols]

[0060] 1: Manufacturing equipment 2: Substrate 4: Liquid 5: Liquid discharge head 7: Pixel area 10: Imaging unit 12: Control unit

Claims

1. A manufacturing apparatus for producing displays, A dispensing unit that dispenses liquid and places it on the substrate, An imaging unit that images a substrate on which the liquid is placed and acquires an image including image irregularities, An evaluation unit evaluates the state of the liquid placed on the substrate based on the output of a machine learning model obtained by inputting the image acquired by the imaging unit into the machine learning model, A manufacturing apparatus characterized by having

2. The imaging unit includes an imaging surface on which a plurality of pixels are arranged, The size of each pixel area on the substrate corresponding to each pixel of the display projected onto the imaging surface is smaller than the size of a single pixel on the imaging surface. The manufacturing apparatus according to feature 1.

3. The manufacturing apparatus according to claim 2, characterized in that at least two or more pixel areas on the substrate are projected onto one pixel of the imaging surface.

4. The manufacturing apparatus according to claim 1, characterized in that the evaluation unit determines whether the state of the liquid disposed on the substrate is good or bad.

5. The manufacturing apparatus according to claim 1, characterized in that the imaging unit acquires the image by detecting interference light between light reflected from the surface of the liquid placed on the substrate and light transmitted through the liquid and reflected from the back surface.

6. The system further includes a processing unit that processes the substrate on which the liquid is placed, The imaging unit images the substrate on which the liquid is placed during the period from when the dispensing unit dispenses the liquid and places it on the substrate until the substrate is transported to the processing unit. The manufacturing apparatus according to feature 1.

7. The manufacturing apparatus according to claim 6, characterized in that the processing unit performs a process of drying the liquid disposed on the substrate.

8. The manufacturing apparatus according to claim 6, characterized in that the processing unit performs a process of firing the liquid disposed on the substrate.

9. The manufacturing apparatus according to claim 6, characterized in that the processing unit performs a process to harden the liquid disposed on the substrate.

10. The imaging unit includes an illumination unit that irradiates light onto the substrate and an image sensor that detects light from the substrate. The angle between the optical axis of the irradiation unit and the substrate, and the angle between the optical axis of the image sensor and the substrate are acute angles. The manufacturing apparatus according to feature 1.

11. The manufacturing apparatus according to claim 10, characterized in that the irradiation unit includes a diffuser plate for diffusing the light irradiated onto the substrate.

12. The manufacturing apparatus according to claim 10, characterized in that the irradiation unit irradiates the substrate with light of a wavelength band different from the wavelength band in which the liquid hardens.

13. The manufacturing apparatus according to claim 10, characterized in that the irradiation unit irradiates the substrate with white light.

14. The manufacturing apparatus according to claim 1, characterized in that the imaging unit acquires the image by detecting scattered light scattered by the substrate from the light emitted from the substrate.

15. A step of placing a liquid on a substrate using the manufacturing apparatus described in any one of claims 1 to 14, A step of processing the substrate on which the liquid is placed, A process for manufacturing a display from the processed substrate, A manufacturing method characterized by having the following:

Images