KR20230064511A - Apparatus and method for measuring dip coating shape using image analysis - Google Patents

Abstract

According to an embodiment of the present invention, a dip coating applied shape measurement device includes: a base member image acquisition unit which acquires a non-coated base material image, and generates a template having an area which is set from the base material image; a coating image acquisition unit which acquires a coating image with respect to the coated base material after the base material is coated; a base material position acquisition unit which acquires position information of the base material from the coating image via template matching between the base material image and the coating image; an image pre-processing unit which performs image preprocessing for removing components not coupled to the base material from the coating image; a coating layer image acquisition unit which removes the base material image from the coating image through the image preprocessing, using the position information of the base material, and acquiring a coating layer image; and a coating layer analysis unit which analyzes a coating layer shape from the coating layer image to acquire a shape indicator of the coating layer. The present invention can efficiently grasp the influence of a processing condition on the quality of a product.

Description

Dip coating application shape measuring device and method using image analysis {APPARATUS AND METHOD FOR MEASURING DIP COATING SHAPE USING IMAGE ANALYSIS}

The present invention relates to an apparatus and method for measuring a dip coating coating shape using image analysis.

In general, dip coating is used to apply a solution to a wide variety of substrates, from chocolate confectionery to automobile parts. Recently, the need to thinly and uniformly apply a solution having complex rheological properties to a fine substrate has emerged. In this case, it is necessary to quantitatively grasp the formation process of the coating layer according to various process conditions.

Conventionally, the coating thickness was determined by inspecting the coated sample in that situation. That is, conventionally, a method of observing a coated substrate under a microscope has been used to evaluate the coating layer thickness in finite depth dip coating.

However, the conventional method has a disadvantage in that a separate inspection process is required after the process and phenomena occurring during the process cannot be grasped.

In addition, the conventional method has a disadvantage in that an approach to observe the deformation of the coating layer in finite depth dip coating has not been made. Accordingly, in the conventional method, since the deformation of the coating layer over time cannot be confirmed, there is a disadvantage in that the resulting change in quality cannot be immediately grasped.

(Prior art literature)

(Patent Document 1) KR Patent Publication No. 10-2013-0060508 (published date: 2013.06.10)

(Patent Document 2) JP Unexamined Patent Publication No. 2002-086572 (published date: 2002.03.26)

(Patent Document 3) JP Patent Publication No. 2019-071356 (published date: 2019.05.09)

An embodiment of the present invention provides a dip coating application shape measuring device and method using image analysis capable of obtaining shape indicators such as coating layer thickness by analyzing a coating process image using computer vision.

According to an embodiment of the present invention, a substrate image acquisition unit that acquires a substrate image for an uncoated substrate and creates a template having a set area in the substrate image: After the substrate is coated, a substrate image for the coated substrate Coating image acquisition unit for obtaining a coating image; a substrate position acquisition unit acquiring location information of the substrate in the coating image through template matching between the substrate image and the coating image;

an image pre-processing unit that performs image pre-processing to remove components not connected to the substrate from the coated image; a coating layer image obtaining unit that obtains a coating layer image by removing the substrate image from the coating image through the image preprocessing using location information of the substrate; and a coating layer analyzer configured to analyze a shape of the coating layer in the coating layer image to obtain a shape index of the coating layer. A dip coating coating shape measuring device including a is proposed.

In addition, according to another embodiment of the present invention, a first step of obtaining a substrate image for an uncoated substrate and generating a template having a set area in the substrate image: After the substrate is coated, the coated substrate A second step of obtaining a coating image for; A third step of acquiring location information of the substrate in the coating image through template matching between the substrate image and the coating image; A fourth step of performing image pre-processing to remove components not connected to the substrate 11 from the coated image; A fifth step of obtaining a coating layer image by removing the substrate image from the coating image through the image preprocessing using location information of the substrate; and a sixth step of obtaining a coating layer shape index by analyzing the coating layer shape in the coating layer image. A dip coating application shape measurement method including a is proposed.

According to an embodiment of the present invention, it is possible to quantify a phenomenon in which the coating layer is flattened over time after application, such as acquiring shape indicators such as coating layer thickness by analyzing a coating process image using computer vision, and process conditions Since the coating thickness according to can be grasped immediately without a separate inspection process, the effect of process conditions on the quality of the product can be grasped efficiently.

According to the present invention, it can be applied to substrates of various shapes and various coating solutions, and can be applied to technical fields such as precision parts production processes such as battery materials using dip coating on three-dimensional substrates, and simple facilities It is possible to measure the change in shape over time of the coating layer applied to the three-dimensional substrate.

In addition, according to the present invention, since the shape of the coating layer can be directly quantified from images taken during the coating process, the shape of the coating layer can be quickly measured under various conditions, and the change in the coating layer over time can be measured during the process. It can be used to derive the optimal process time.

In addition, since the fluid that comes along with the substrate is redistributed flatly over time, the flatness of the coating layer and the time required for the process have a trade-off relationship. By quantifying, the optimal process time can be derived.

1 is an exemplary view of a dip coating application shape measuring device according to an embodiment of the present invention.
2 is an exemplary diagram of a coating analyzer according to an embodiment of the present invention.
3 is an exemplary view of a method for measuring a dip coating coating shape according to an embodiment of the present invention.
Figure 4 is a detailed view of Figure 3;
5 is an exemplary diagram of a substrate image and a coating image.
6 is an exemplary view of a substrate image and a template region.
7 is an exemplary view of a vertex detection process of a substrate.
8 is an exemplary view of a process of determining the breakage of a capillary bridge.
9 is a diagram illustrating the presence and absence of a capillary bridge.
10 is an exemplary view of calculating a cross-sectional area of a coating layer.
11 is an exemplary view of calculating the volume of a coating layer.
12 is a video analysis flow chart applied to all frames of a video.
13 is a graph of a change in cross-sectional area of a coating layer according to analysis results for all frames of a video.
14 is a graph of the change in volume of the coating layer according to the analysis results for all frames of the video.

Hereinafter, it should be understood that the present invention is not limited to the described embodiments, and various changes can be made without departing from the spirit and scope of the present invention.

In addition, in each embodiment of the present invention, the structure, shape, and numerical value described as an example are only examples to help the understanding of the technical details of the present invention, so they are not limited thereto, but the spirit and scope of the present invention It should be understood that various changes can be made without departing from it. Embodiments of the present invention can be combined with each other to make various new embodiments.

In addition, in the drawings referred to in the present invention, elements having substantially the same configuration and function in light of the overall content of the present invention will use the same reference numerals.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily practice the present invention.

1 is an exemplary view of a dip coating application shape measuring device according to an embodiment of the present invention.

Referring to FIG. 1 , the dip coating coating shape measuring device 10 may include a light source 14, a camera 15, and a coating shape measuring device 20 to analyze the coating layer of the substrate 11. .

The light source 14 may operate under the control of the coating shape measuring instrument 20 to irradiate the substrate 11 with light.

The camera 15 operates under the control of the coating shape measuring instrument 20, can be synchronized with the operation of the light source 14, and captures the uncoated substrate 11 or the coated substrate 11 to describe the substrate. An image or coating image may be provided to the coating shape measuring device 20 .

For example, the coating shape measuring device 20 may include a controller 21 and a coating analyzer 22 . The controller 21 may control the operation of the light source 14 and the camera 15 . The coating analyzer 22 may analyze the coating layer using a substrate image or a coating image input from the camera 15 .

Also, in FIG. 1, 11a is a coating layer of the substrate 11, 16 is a coating pool, and 16a is a coating solution.

For example, the substrate 11 applied to the present invention may be a cylindrical SiO2 substrate, and for example, the diameter of the substrate may be about 3 mm. The coating solution 16a may be silicone oil, the viscosity of the coating solution may be 10,000 cSt, and the specific gravity may be 0.975.

On the other hand, if the photographing of the substrate 11 using the camera 15 is described as an example, after taking an image of the substrate 11, the substrate 11 is placed in the coating pool 16 having a depth of 1 mm at 1 mm/s It was immersed at a speed of , and the substrate 11 was pulled out at a speed of 1 mm/s 2 seconds after the substrate 11 touched the bottom of the coating pool 16. At the same time, after filming the change of the coating layer (11a) for about 5 minutes as a video, the cross-sectional area and volume in pixel units of the coating layer (11a) of the base material (11) were calculated, and then the pixel unit volume was converted into an actual unit volume. After that, the length unit error rate was calculated by comparing the volume of the actually applied coating layer. A detailed description of this will be given later.

2 is an exemplary diagram of a coating analyzer according to an embodiment of the present invention.

Referring to FIG. 2 , the coating analyzer 22 includes a substrate image acquisition unit 100, a coating image acquisition unit 200, a substrate location acquisition unit 300, an image preprocessing unit 400, and a coating layer image acquisition unit 500. ), and a coating layer analysis unit 600.

The substrate image acquisition unit 100 obtains a substrate image M1 for the uncoated substrate 11, and generates a template TP having a set area in the substrate image to obtain the substrate position acquisition unit 300. can be output to

After the substrate 11 is coated, the coating image acquisition unit 200 may obtain a coating image M2 of the coated substrate and output the obtained coating image M2 to the substrate location acquisition unit 300 .

The substrate position acquisition unit 300 may obtain location information of the substrate 11 in the coating image M2 through template TP matching between the substrate image M1 and the coating image M2. there is.

The image pre-processing unit 400 may perform image pre-processing to remove components not connected to the substrate 11 from the coating image M2 .

The coating layer image acquisition unit 500 may obtain a coating layer image M6 by removing the substrate image M1 from the coating image M2 through the image preprocessing using the location information of the substrate.

The coating layer analyzer 600 may obtain coating layer shape indicators (eg, cross-sectional area, volume) by analyzing the coating layer shape in the coating layer image M6 .

For example, in the substrate image acquisition unit 100 , the template TP may include a coupling unit CP coupling the substrate 11 to the device 1 .

The substrate position acquisition unit 300 converts the substrate image M1 and the coating image M2 into black and white images by binarizing the substrate image M1 and the coating image M2 before matching the template TP, and The template TP matching may be performed based on each black and white image of the coating image M2.

The substrate position acquisition unit 300 may detect four vertexes corresponding to the substrate in the black and white substrate image M1 and obtain location information of the substrate using the four vertexes of the substrate.

The image pre-processing unit 400 determines whether there is a capillary bridge (CB) connected to a fluid between the coating image M2 and the coating pool 16, and the capillary bridge (CB) ) is disconnected, the image pre-processing may be performed.

In the image preprocessing unit 400, components not connected to the substrate 11 may include the coating pool.

In the coating layer analyzer 600, the coating layer shape index may include a cross-sectional area of the coating layer and a volume of the coating layer. For example, the coating layer analyzer 600 may calculate the cross-sectional area and volume of the coating layer by counting the number of points constituting the coating layer, and based on this, the cross-sectional area in units of pixels and the volume in units of pixels may be obtained.

For example, the substrate image acquisition unit 100, the coating image acquisition unit 200, the substrate location acquisition unit 300, the image pre-processing unit 400, the coating layer image acquisition unit 500, and the coating layer analysis unit 600 All may be implemented on a single processor, or may be implemented on separate individual processors.

In each of the substrate image acquisition unit 100, the coating image acquisition unit 200, the substrate location acquisition unit 300, the image preprocessing unit 400, the coating layer image acquisition unit 500, and the coating layer analysis unit 600 For operation, reference may be made to the following description.

For each drawing of the present invention, unnecessary redundant descriptions of the components having the same reference numerals and the same functions can be omitted, and possible differences can be described for each drawing.

The description of the dip coating coating shape measuring device according to the embodiment of the present invention described above and the description of the dip coating coating shape measuring method described below may be mutually applied to each other complementaryly, and thus possible overlapping descriptions will be omitted. can

3 is an exemplary view of a method for measuring a dip coating coating shape according to an embodiment of the present invention, and FIG. 4 is a detailed view of FIG. 3 . And, Figure 5 is an exemplary diagram of a substrate image and a coating image.

Referring to FIGS. 3 and 4 , the dip coating application shape measurement method according to an embodiment of the present invention includes a first step (S100) of obtaining a substrate image M1 and generating a template TP, and a coating image A second step (S200) of acquiring (M2), a third step (S300) of acquiring location information of the substrate 11, a fourth step (S400) of performing image preprocessing, and obtaining a coating layer image (M6) A fifth step ( S500 ) and a sixth step ( S600 ) of obtaining a coating layer shape index may be included.

Referring to FIG. 3 , in a first step ( S100 ), a substrate image M1 of an uncoated substrate 11 may be obtained, and a template TP having a set area may be generated in the substrate image. For example, the first step ( S100 ) may be performed by the substrate image obtaining unit 100 of FIG. 2 . For example, by taking a silhouette image of the substrate 11 before starting the coating process, the template TP may be created from the silhouette image.

In the second step (S200), after the substrate 11 is coated, a coating image M2 of the coated substrate may be acquired. For example, the second step (S200) may be performed by the coating image acquisition unit 200 of FIG. For example, the experimental image, which is the coating image M2 , may be obtained by taking a silhouette image of the coated substrate by the camera 15 after the substrate is coated with the coating solution of the coating pool.

In the third step (S300), through the template (TP) matching between the substrate image (M1) and the coating image (M2), it is possible to obtain the location information of the substrate 11 in the coating image (M2). . For example, the third step (S300) may be performed by the description position obtaining unit 300 of FIG. 2 .

For example, before determining the position of the substrate 11 in the experimental image M2 through the template matching, after the template image and the coating image are binarized in black and white, template matching is performed between the two template images and the coating image. can be used to find the position of the object. Here, the template matching is a technique of searching for a region having the highest similarity with the template image in the coating image. In addition, one of methods such as correlation coefficient matching (CCOEFF), cross correlation matching (CCORR), and square difference matching (SQDIFF) may be applied as a method of calculating similarity. Although the CCOEFF method is used in this example, it is not limited thereto, and any method may be used as long as the location of the template can be well calculated.

Meanwhile, referring to FIG. 5 , M1 is a substrate image, M2 is a coating image, M3 is a coating image in which a matching area (TMA) is displayed, M4 is a pre-processed coating image, and M5 is a matching area in a pre-processed coating image. This is an image of the removed initial coating layer. And, M6 is the final coating layer image post-processed from the initial coating layer image.

In the fourth step ( S400 ), image preprocessing may be performed to remove components (eg, coating paste) not connected to the substrate 11 from the coating image M2 . For example, the fourth step ( S400 ) may be performed by the image pre-processing unit 400 of FIG. 2 .

For example, in the image preprocessing, all regions such as the coating pool 16 other than the substrate image M3 to which the coating layer is applied may be removed. In FIG. 5, M4 shows a pre-processed coating image.

In the fifth step (S500), the substrate image M1 may be removed from the coating image M2 through the image preprocessing using the location information of the substrate, thereby obtaining the coating layer images M5 and M6. For example, the fifth step ( S500 ) may be performed by the coating layer image acquisition unit 500 of FIG. 2 . When the initial coating layer image M5 is post-processed, a final coating layer image M6 may be obtained.

For example, the coating layer image M6 may be obtained by removing the tablet matching area TMA from the coating image M4 to which the coating layer is applied using positional information of the substrate corresponding to the tablet matching area TMA. can In FIG. 5 , when M3 and M5 are compared, it can be confirmed that the template matching area MA of M3 has been removed.

In the sixth step (S600), the coating layer shape index (eg, cross-sectional area, volume) may be obtained by analyzing the coating layer shape in the coating layer image M6. For example, the sixth step ( S600 ) may be performed by the coating layer analyzer 600 of FIG. 2 .

For example, the final coating layer image M6 generated after the initial coating layer image M5 is post-processed may be analyzed. Here, in the image post-processing, the final coating layer image M6 may be obtained by removing unnecessary areas and noise that have not yet been removed from the initial coating layer image M5. Such image post-processing may be performed in a fifth step or a sixth step.

after. The coating layer shape index such as the coating layer thickness may be quantified using the final coating layer image M6.

Referring to FIG. 4, in the first step (S100), the substrate image (M1) photographed through the camera 15 is loaded, and in the second step (S200), the coating image (M3) photographed through the camera is loaded. can be loaded. The substrate image M1 and the coating image M3 may be previously captured by the camera 15 and stored in the internal memory of the coating analyzer (FIGS. 1 and 22).

Thereafter, in the third step (S300), as shown in FIG. 5 , the range where the substrate is actually located in the substrate image M1 and the area TMA to be used as a template may be set. In this case, the range of the substrate image M1 may be set wide horizontally to include both left and right ends of the expected coating layer in order to secure the coating layer image. Next, both the substrate image M1 and the coating image M2 may be binarized into black and white (S310 and S320). For example, the binarization of an image may use an adaptive thresholding technique that automatically recognizes a contrast difference of the entire image. Then, the four vertices of the substrate may be detected from the black and white image of the substrate, and the coordinates of the vertices may be stored in the internal memory (S330). For this, reference may be made to FIG. 7 . Then, the location of the substrate may be determined through template matching between the template matching area TMA and the coating image M2.

Then, in the fourth step (S400), the video (FIG. 9, V1, V2) taken through the camera (FIG. 1, 15) is taken at the time when the substrate starts to come out of the coating pool (FIG. 1, 16) As shown in V1 of FIG. 9, a capillary bridge (CB) to which the coating solution 16a is connected may exist between the lower end of the substrate 11 and the coating pool at the beginning. This capillary bridge CB becomes thinner and then breaks over time, and from that point on, the coating layer 11a of the substrate is completely separated from the coating pool 16, and then the coating layer analysis can be performed. Accordingly, in order to perform the coating layer analysis, it is necessary to check whether the capillary bridge is broken (S410). For this, reference may be made to FIG. 8 .

For example, if a capillary bridge is not broken, it may end without proceeding to the coating layer analysis. If the capillary bridge is broken, image pre-processing may be performed. For example, the process of removing the coating fluid connected to the bottom of the coating image may be performed through a connected component labeling technique. A connected component labeling technique is an image analysis technique that divides and recognizes connected objects and separated objects as a “chunk” in an image.

Thereafter, after the area of the coated substrate and the area of the coating solution are completely separated, an image preprocessing process may be performed. The midpoint of the substrate is selected based on the four vertices of the substrate detected in the previous step (S420), Since the midpoint of is always located inside the substrate, image preprocessing may be performed by removing components not connected to the corresponding vertex through a connected component labeling technique (S430).

Then, in the fifth step (S500), the image of the substrate is removed from the preprocessed image, and as shown in image M5 of FIG. 5, a coating layer image M5 from which the image of the substrate is removed can be obtained ( S510), and also, unnecessary areas may be removed by removing pixels outside the top, left, and right sides of the substrate range in the coating layer image M5 (S520). Here, pixels outside the bottom of the substrate image are not removed because the coating layer is included in the corresponding image area.

Then, in a sixth step (S600), a post-processing process may be performed on the coating layer image before analyzing the coating layer (S601, S602, and S603). Noise can be removed by removing pixels above the expected coating depth in the coating layer image M6. In addition, the expected coating depth is obtained from the coating layer depth of the last frame of the video, and the same value is applied to all frames of the video.

A cross-sectional area of the coating layer and a volume of the coating layer may be calculated from the coating layer image obtained through the above process. In this regard, reference may be made to FIG. 11,

6 is an exemplary view of a substrate image and a template region.

Referring to FIG. 6 , in the first step S100, the template TP based on the substrate image M1 (or the template image) is a coupling portion CP where the substrate 11 is coupled to the device 1. ).

For example, the template TP may include a coupling portion CP in which the substrate 11 is coupled to the device, and this template TP determines the location of the substrate in the coating image (or experimental image) through template matching. can be used to figure out In addition, the template TP should include a coupling part CP where the substrate is coupled to the device, which is to calculate the position of the substrate in the height direction based on the corresponding information.

In FIG. 6 , the lower rectangular shape corresponds to the substrate 11 and the upper flat part corresponds to a part of the device 1 to which the substrate 11 is coupled.

4, 5 and 6, in the third step (S300), before matching the template (TP), the substrate image (M1) and the coating image (M2) are binarized to obtain a black and white image. conversion, and the template TP matching may be performed based on each of the black and white images of the substrate image M1 and the coating image M2.

7 is an exemplary view of a vertex detection process of a substrate.

4 and 7, in the third step (S300), four vertices corresponding to the substrate are detected in the black-and-white substrate image M1, and location information of the substrate is used using the four vertices of the substrate. can be obtained,

For example, in the third step (S300), an outer boundary may be first extracted from the image M1 of the substrate (S331). As an example, this can be performed using the 'Cannyedge detection' technique, and variables required for this can be set in advance and applied to all frames of the video at once.

Thereafter, straight lines may be detected from the outer boundary line (S333). For example, this may be performed using a 'Hough line transformation' technique, and variables necessary for this may be set in advance and applied collectively to all frames of the video.

Then, vertical and horizontal straight lines are collected from the detected straight lines, and then classified into upper, left, lower, and right areas of the substrate image area, and the straight line that best matches the outer boundary line in each area is the four sides of the substrate. It can be (S335).

Then, the coordinates of the vertex of the substrate image may be obtained by calculating the coordinates of the vertex from the intersection of the four sides (S337).

For example, the coordinates of the vertex in the coating layer image may be calculated by adding the position of the substrate in the coating layer image to the coordinates of the vertex in the substrate image.

8 is an exemplary view of a process for determining that a capillary bridge is disconnected, and FIG. 9 is an example view of the existence and non-existence of a capillary bridge.

4, 8 and 9, in the fourth step (S400), a capillary bridge (CB) in which fluid is connected between the coating image M2 and the coating pool 16 It is determined whether there exists, and when the capillary bridge CB is broken, the image pre-processing may be performed. For example, components not connected to the substrate 11 may include the coating pool.

For example, referring to FIGS. 8 and 9, in the fourth step (S400), the coordinates of the lower two vertices at the base position are obtained (S411). Then, after setting the lower end of the coating image from the two points to the detection area (S412), it is calculated whether there is a column consisting of only black in the area, and if there is a column consisting only of black, the capillary bridge (CB) is not broken. It is determined that it is not (S413), otherwise it can be determined that the capillary bridge is broken.

10 is an exemplary view of calculating the cross-sectional area of the coating layer, and FIG. 11 is an exemplary view of calculating the volume of the coating layer.

For example, referring to FIG. 10 , in the sixth step (S600), the cross-sectional area may be determined by counting the total number of pixels of the image corresponding to the coating layer in the image from which only the coating layer is extracted (S610).

Referring to FIG. 11 , in the sixth step (S600), the center line of the substrate is obtained from the four vertexes of the substrate (S621). Since the substrate has a cylindrical shape and satisfies axial symmetry, a distance from each pixel of the coating layer image to the center line of the substrate is obtained (S622). Set this distance as a radius, calculate the circumference of the circle, and add it for all pixels. At this time, since the point on the left and the point on the right were calculated overlapping from the center line, if the sum is divided by 2, the coating layer according to the principle of quadrature The pixel unit volume of is calculated (S623). Afterwards, the actual distance per pixel is calculated from the pixel distance between the bottom two vertices of the substrate and the width of the substrate (S624), and based on this, the pixel unit volume of the coating layer can be converted into the actual unit volume of the coating layer (S625).

12 is a video analysis flow chart applied to all frames of a video.

Referring to FIG. 12, in the video analysis in step 6 (S600), if the video frame is analyzed by applying the analysis of the present invention, it is first determined whether all frames have been analyzed (S631), and all frames are analyzed. If so, this analysis process ends, and if not, it can be determined whether or not the description position has been transmitted in the previous frame (S632). If the substrate position is not transmitted in the previous frame, the template matching range is set to the entire frame (S633). If the description position is transmitted in the previous frame, the template matching range is set to ±10 pixels from the previous description position (S634). Thereafter, the next frame analysis is performed (S635), and the analysis result is recorded (S636).

Here, when analyzing the first frame, all areas were set as template matching ranges. The range obtained by adding a margin of ±10 pixels from the position of the substrate in the previous frame was limited to the range of template matching, which is to improve the analysis speed by reducing the number of pixels to be searched for during template matching.

13 is a graph of the change in cross-sectional area of the coating layer according to the analysis result for all frames of the video, and FIG. 14 is a graph of the change in volume of the coating layer according to the analysis result for all the frames of the video.

Referring to FIGS. 13 and 14 , right after the capillary bridge is broken, errors rapidly decrease in cross-sectional area and volume values, but both data show stable values thereafter.

Since it is difficult to calculate actual measured values, the cross-sectional area in units of pixels in FIG. 13 is used for the purpose of confirming only the qualitative trend of the data, and the volume in units of pixels in FIG. 14 is compared with the volume of the actually applied fluid to verify the error. The actual volume was calculated using the weight change of the substrate before and after coating and the specific gravity of the fluid. As a result of converting the volume unit error of the calculation result into the length unit error, it was confirmed that the thickness of the coating layer was observed with an accuracy of 10% to 20% in this embodiment.

As described above, the present invention is utilized as a general knowledge for theoretical analysis of the dip coating process to enable effective process design, and interest in coating layer control is continuously increasing and trending in the manufacturing market, which is being refined and miniaturized. The market to which the present invention is applied is also expected to expand.

On the other hand, the coating analyzer 22 of the dip coating coating shape measuring device according to an embodiment of the present invention is a processor (eg, a central processing unit (CPU), a graphic processing unit (GPU), a microprocessor, an application specific semiconductor (Application Specific Integrated Circuit (ASIC), Field Programmable Gate Arrays (FPGA), etc.), memory (e.g., volatile memory (e.g., RAM, etc.), non-volatile memory (e.g., ROM, flash memory, etc.), input device (e.g., : Keyboard, mouse, pen, voice input device, touch input device, infrared camera, video input device, etc.), output device (eg display, speaker, printer, etc.) and communication access device (eg modem, network interface card (NIC) ), integrated network interfaces, radio frequency transmitters/receivers, infrared ports, USB interfaces, etc.) interconnect each other (e.g. Peripheral Component Interconnect (PCI), USB, firmware (IEEE 1394), optical bus structures, networks, etc.) ) can be implemented as a computing environment.

The computing environment may be a personal computer, server computer, handheld or laptop device, mobile device (mobile phone, PDA, media player, etc.), multiprocessor system, consumer electronic device, mini computer, mainframe computer, any of the foregoing systems or It may be implemented in a distributed computing environment including devices, but is not limited thereto.

In the above, the present invention has been described as an embodiment, but the present invention is not limited to the above-described embodiments, and those skilled in the art in the field to which the invention pertains without departing from the gist of the present invention claimed in the claims Anyone can make various variations.

11: registration
100: substrate image acquisition unit
200: coating image acquisition unit
300: description location acquisition unit
400: image pre-processing unit
500: coating layer image acquisition unit
600: coating layer analysis unit

Claims (16)

A substrate image acquisition unit for obtaining a substrate image for an uncoated substrate and generating a template having a set area in the substrate image:
a coating image acquisition unit acquiring a coating image of the coated substrate after the substrate is coated;
a substrate position acquisition unit acquiring location information of the substrate in the coating image through template matching between the substrate image and the coating image;
an image pre-processing unit that performs image pre-processing to remove components not connected to the substrate from the coated image;
a coating layer image obtaining unit that obtains a coating layer image by removing the substrate image from the coating image through the image preprocessing using location information of the substrate; and
a coating layer analyzer for obtaining a coating layer shape index by analyzing the coating layer shape in the coating layer image;
Dip coating coating shape measuring device comprising a.
The method of claim 1, wherein the substrate image acquisition unit,
The template includes a coupling portion in which the substrate is coupled to the device
Dip coating coating shape measuring device.
The method of claim 1, wherein the description location acquisition unit,
Prior to the template matching, the base image and the coating image are binarized and converted into black and white images, and the template matching is performed based on the black and white images of each of the base image and the coating image
Dip coating coating shape measuring device.
The method of claim 3, wherein the description location acquisition unit,
Detecting four vertices corresponding to the substrate in the black and white substrate image, and obtaining location information of the substrate using the four vertices of the substrate
Dip coating coating shape measuring device.
The method of claim 1, wherein the image pre-processing unit,
Determining whether there is a capillary bridge connected to the fluid between the coating image and the coating pool, and proceeding with the image preprocessing when the capillary bridge is broken
Dip coating coating shape measuring device.
The method of claim 5, wherein the image pre-processing unit,
Components not connected to the substrate include the coating pool.
Dip coating coating shape measuring device.
The method of claim 1, wherein the coating layer analysis unit,
The coating layer shape index includes the cross-sectional area of the coating layer and the volume of the coating layer.
Dip coating coating shape measuring device.
The method of claim 7, wherein the coating layer analysis unit,
The cross-sectional area and volume of the coating layer are obtained by counting all the points constituting the coating layer, and obtaining the cross-sectional area and volume in pixel units based on this.
Dip coating coating shape measuring device.
A first step of acquiring a substrate image for an uncoated substrate and generating a template having a set area in the substrate image:
a second step of acquiring a coating image of the coated substrate after the substrate is coated;
A third step of acquiring location information of the substrate in the coating image through template matching between the substrate image and the coating image;
a fourth step of performing image preprocessing to remove components not connected to the substrate from the coated image;
A fifth step of obtaining a coating layer image by removing the substrate image from the coating image through the image preprocessing using location information of the substrate; and
A sixth step of obtaining a coating layer shape index by analyzing the coating layer shape in the coating layer image;
Dip coating coating shape measurement method comprising a.
The method of claim 9, wherein in the first step,
The template includes a coupling portion in which the substrate is coupled to the device
Dip coating coating shape measurement method.
The method of claim 9, wherein in the third step,
Prior to the template matching, the base image and the coating image are binarized and converted into black and white images, and the template matching is performed based on the black and white images of each of the base image and the coating image
Dip coating coating shape measurement method.
The method of claim 11, wherein in the third step,
Detecting four vertices corresponding to the substrate in the black and white substrate image, and obtaining location information of the substrate using the four vertices of the substrate
Dip coating coating shape measurement method.
The method of claim 9, wherein in the fourth step,
Determining whether there is a capillary bridge connected to the fluid between the coating image and the coating pool, and proceeding with the image preprocessing when the capillary bridge is broken
Dip coating coating shape measurement method.
The method of claim 13, wherein in the fourth step,
Components not connected to the substrate include the coating pool.
Dip coating coating shape measurement method.
The method of claim 9, wherein in the sixth step,
The coating layer shape index includes the cross-sectional area of the coating layer and the volume of the coating layer.
Dip coating coating shape measurement method.
The method of claim 15, wherein in the sixth step,
The cross-sectional area and volume of the coating layer are obtained by counting all the points constituting the coating layer, and obtaining the cross-sectional area and volume in pixel units based on this.
Dip coating coating shape measurement method.

Images