Search

KR-20260065332-A - Real-time monitoring system for thin film thickness in a vacuum drying device using artificial intelligence and real-time monitoring method for thin film thickness using the same

KR20260065332AKR 20260065332 AKR20260065332 AKR 20260065332AKR-20260065332-A

Abstract

One embodiment of the present invention provides a system for real-time monitoring of thin film thickness in a vacuum drying device using artificial intelligence and a method for real-time monitoring of thin film thickness using the same. The method for real-time monitoring of thin film thickness using the system for real-time monitoring of thin film thickness in a vacuum drying device according to one embodiment of the present invention has the effect of detecting quality defects during production at an early stage by allowing real-time observation of abnormal changes in the thickness or shape of a thin film of a display being dried.

Inventors

  • 강경태
  • 신동열
  • 조관현

Assignees

  • 한국생산기술연구원

Dates

Publication Date
20260508
Application Date
20241101

Claims (10)

  1. A data collection unit installed on the upper part of an inkjet printer, which obtains a planar image of ink sprayed inside a subpixel by a first microscope imaging device and obtains three-dimensional surface shape data including ink volume and thin film thickness information by a three-dimensional surface imaging device; A thin film prediction model generation unit including a machine learning unit connected to an information processing device using an electronic circuit, which generates an algorithm by machine learning such that a planar image of ink obtained from the data collection unit is used as an input value, and a 3D surface shape image converted from 3D surface shape data is used as an output value, so that the planar image of ink and the 3D surface shape image correspond according to the amount of ink; and A real-time thin film thickness monitoring system within a vacuum drying device, characterized by including a monitoring unit that obtains a planar image of an ink thin film during drying by means of a second microscope imaging device connected to a vacuum drying device, inputs the planar image of the ink thin film to an information processing device connected to the thin film prediction model generation unit to predict the thickness of the ink thin film, and then outputs the result.
  2. In paragraph 1, The above data collection unit is, A real-time thin film thickness monitoring system in a vacuum drying device, characterized by further including a data storage device that stores a planar image of ink obtained by the first microscope imaging device and three-dimensional surface shape data obtained by a three-dimensional surface imaging device.
  3. In paragraph 2, In the above data collection unit, A real-time thin film thickness monitoring system in a vacuum drying device characterized by converting the above-mentioned 3D surface shape data into a 2D surface shape image having pixel values of 16 bits or more.
  4. In paragraph 1, A real-time thin film thickness monitoring system in a vacuum drying device, characterized in that the machine-learned models used in the thin film prediction model generation unit are CNN, U-net, and Pix2Pix.
  5. In paragraph 1, The above monitoring unit A real-time thin film thickness monitoring system within a vacuum drying device, characterized by further including a vacuum pump for controlling the drying speed of ink introduced into the vacuum drying device.
  6. In paragraph 1, The above monitoring unit A real-time thin film thickness monitoring system within a vacuum drying device, characterized by further including a vacuum pump for controlling the drying speed of ink introduced into the vacuum drying device.
  7. A step of obtaining a planar image of the ink while the ink is sprayed into the subpixel by the first microscope imaging device, and obtaining three-dimensional surface shape data including the ink volume and thin film thickness information by a three-dimensional surface imaging device; A step of obtaining a 3D surface shape image by correcting the brightness of the above 3D surface shape data and then imaging it using computer vision technology; A step of generating a thin film prediction model by machine learning such that the planar image of the ink obtained from the data collection unit is used as an input value, and the 3D surface shape image converted from 3D surface shape data is used as an output value, so that the planar image of the ink and the 3D surface shape image correspond according to the amount of ink; and A method for monitoring the thickness of a thin film in a vacuum drying device in real time, characterized by including the step of inputting a planar image of ink introduced into the vacuum drying device into an information processing device based on a value predicted through a constructed thin film prediction model, predicting the thickness of the ink thin film in real time, and then outputting it.
  8. In Paragraph 7, In the step of obtaining the above three-dimensional surface shape image, A step of spraying UV-curing ink onto a substrate patterned with the above subpixels using inkjet printing; A step of curing the printed ink using the above UV lamp; and A method for real-time thin film thickness monitoring in a vacuum drying device, characterized by including the step of measuring the thickness of subpixels that are not filled with ink and subpixels that are filled to correct the camera brightness value of the drying device or applying a correction value to the thin film thickness of the predicted result in an algorithm.
  9. In Paragraph 7, In the step of obtaining the above three-dimensional surface shape image, A method for real-time thin film thickness monitoring in a vacuum drying device, characterized by converting the above-mentioned 3D surface shape data into a 3D surface shape image having pixel values of 16 bits or more.
  10. In Paragraph 7, In the step of generating a thin film prediction model using the machine learning described above, A method for real-time thin film thickness monitoring in a vacuum drying device, characterized in that the machine-learned models are CNN, U-net, and Pix2Pix.

Description

Real-time monitoring system for thin film thickness in a vacuum drying device using artificial intelligence and real-time monitoring method for thin film thickness using the same The present invention relates to a system for real-time monitoring of thin film thickness in a vacuum drying device using artificial intelligence and a method for real-time monitoring of thin film thickness using the same. Referring to FIG. 1, a conventional inkjet pixel manufacturing process can fill an inkjet head with ink and then generate pressure inside the inkjet head for a short period of time to spray ink droplets into a sub-pixel. At this time, in order to form a thin film through the ink droplets sprayed into the subpixel, referring to FIG. 2, the pressure and temperature inside the drying device can be controlled to evaporate and dry the residual solvent inside the printed subpixel, thereby forming a thin film. In this context, observing changes in the thickness of OLED thin films (EML, common layer) or quantum dot display thin films (QD-LED) in the display field is important for improving the efficiency and quality of the manufacturing process, and the manufacturing/drying process can be optimized by observing changes in the thin film thickness profile. However, non-uniform thin film thickness can cause uneven brightness within subpixels, leading to a Mura phenomenon that resembles staining and resulting in imperfections in the display panel. To minimize such mura, uniform drying and deposition of ink across the entire panel is essential; however, this requires the cumbersome process of precisely controlling drying parameters (temperature and pressure), detecting real-time thickness deviations, and modifying drying conditions using a high-level monitoring and feedback system. To solve the aforementioned problems, contact measurement, microscopic measurement, or optical interference measurement methods have been applied as techniques for measuring thin film thickness profiles during the thin film formation process using conventional inkjet printer manufacturing processes and drying devices; however, these measurement methods cannot be used inside a vacuum drying device, and furthermore, there was a problem in that it was impossible to monitor the thin film thickness in real time. Therefore, many challenges still remain for the uniform drying and deposition of ink. Figure 1 is a schematic diagram showing a conventional inkjet pixel manufacturing process. FIG. 2 is a schematic diagram showing the form of a thin film formed by evaporating and drying the residual solvent in a subpixel formed by controlling the pressure and temperature inside a conventional drying device. Figure 3 is a schematic diagram showing the configuration of a real-time thin film thickness monitoring system within a vacuum drying device. Figure 4 is a schematic diagram showing the configuration of the output section of the real-time thin film thickness monitoring system in the vacuum drying device. Figure 5 is a flowchart illustrating a method for real-time thin film thickness monitoring in a vacuum drying device. Figure 6 is a schematic diagram illustrating the steps of generating a thin film prediction model by machine learning a method for monitoring the thickness of a thin film in real-time within a vacuum drying device. Figure 7 is a schematic diagram showing the correction of the brightness of three-dimensional surface shape data in a real-time thin film thickness monitoring method in a vacuum drying device. Figure 8 is an image showing top-view microscope images over time and 3D thin film thickness profile data predicted by artificial intelligence for subpixels filled with OLED ink. Figure 9 is a graph of the data for the average thickness of the ROI area inside the subpixel. Figure 10 shows top-view microscope images over time of subpixels filled with OLED ink and thin film profile data predicted by artificial intelligence. The present invention will be described below with reference to the attached drawings. However, the present invention may be implemented in various different forms and is therefore not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification have been given similar reference numerals. Throughout the specification, when it is stated that a part is "connected (connected, in contact, combined)" with another part, this includes not only cases where they are "directly connected," but also cases where they are "indirectly connected" with other members interposed between them. Furthermore, when it is stated that a part "includes" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but rather allows for the inclusion of additional components. The terms used herein are merely for describing specific embodime