Apparatus for predicting contact angle through extraction of droplet boundary characteristic of low-level image on basis of learning model and method therefor

Abstract

A method for predicting a contact angle comprises the steps of: generating, by a data processing unit, training data including a droplet image vector, which is a low-level droplet image having reduced resolution, and a target vector indicating a contact angle calculated corresponding to the droplet image vector; and generating, by a model generation unit, a learning model for predicting a contact angle of a droplet through weighted calculation from the droplet image vector by using the training data.

Description

Device and method for predicting contact angle by extracting droplet boundary characteristics from low-level images based on learning models

[0001] The present invention relates to a technique for predicting a contact angle, and more particularly, to a device and method for predicting a contact angle by extracting droplet boundary characteristics from a low-level image based on a learning model.

[0002] There is a liquid droplet on a solid surface in the air, and the angle formed between the solid surface and the tangent line at the triple point, that is, the point of contact of the three phases (solid, liquid (liquid droplet), and gas (air), is called the contact angle of the liquid droplet on the solid.

[0003] The purpose of the present invention is to provide a device and method for predicting a contact angle by extracting droplet boundary characteristics from a low-level image based on a learning model.

[0004] A method for predicting a contact angle according to a preferred embodiment of the present invention for achieving the above-described purpose includes a step of generating learning data including a droplet image vector, which is a low-level droplet image with reduced resolution, and a target vector representing a contact angle calculated in response to the droplet image vector, by a data processing unit, and a step of generating a learning model for predicting the contact angle of a droplet through a weight operation from the droplet image vector using the learning data, by a model generation unit.

[0005] The step of generating the above learning data includes a step in which the data processing unit generates a droplet image, a step in which the data processing unit calculates a contact angle of the droplet image, and a step in which the data processing unit reduces the resolution of the droplet image to generate a droplet image vector, which is a low-level droplet image.

[0006] The step of generating the above droplet image vector includes a step in which the data processing unit horizontally cuts an ellipse based on the triple point of the droplet to generate a droplet image having the overall shape of the droplet with a ratio of the major axis to the minor axis within a predetermined range, and a step in which the data processing unit extracts a portion of the droplet image having the overall shape to generate a droplet image including at least the triple point and the highest point.

[0007] The ratio of the major axis to the minor axis is characterized by being 1.00 to 0.80.

[0008] The step of calculating the contact angle of the above droplet image is performed by the data processing unit using the mathematical formula

[0009]

[0010] The contact angle corresponding to the droplet image is calculated according to the following, wherein θ is the contact angle of the droplet image, a is the major axis of the ellipse that is the basis of the droplet image, b is the minor axis of the ellipse that is the basis of the droplet image, x1 and y1 are the center coordinates of the ellipse that is the basis of the droplet image, and x and y are the coordinates of the triple point of the droplet image.

[0011] The step of generating the droplet image vector, which is the low-level droplet image, is characterized in that the data processing unit resizes the droplet image having a first size and a first resolution corresponding to the first size to generate a low-level droplet image having a second resolution that is lower than the first resolution while having the first size.

[0012] The step of generating the learning model includes a step in which the model generation unit inputs the droplet image vector into the learning model, a step in which the learning model performs a plurality of operations to which weights for which learning has not been completed between multiple layers are applied to the droplet image vector to derive a prediction vector representing a predicted contact angle of a droplet image corresponding to the droplet image vector, a step in which the model generation unit derives a loss representing a difference between the target vector and the prediction vector through a loss function, and a step in which the model generation unit performs optimization to update the weights of the learning model so that the loss is minimized.

[0013] In order to achieve the above-described object, a method for predicting a contact angle according to a preferred embodiment of the present invention includes a step in which, when a droplet image of which a contact angle is unknown is input by a preprocessing unit, the input droplet image is preprocessed to generate a droplet image vector, a step in which an inference unit inputs the droplet image vector into a learning model learned using learning data including a learning droplet image vector which is a low-level droplet image with a reduced resolution and a target vector representing a contact angle calculated in response to the learning droplet image vector, a step in which the learning model performs a plurality of operations to which weights learned between a plurality of layers are applied to the droplet image vector to derive a prediction vector that predicts the contact angle of a droplet of the droplet image, and a step in which the inference unit outputs the contact angle of the prediction vector.

[0014] The step of generating the above droplet image vector includes a step in which the preprocessing unit sequentially performs gray scale, Gaussian blur, and binarization on the input droplet image to remove external noise, and a step in which the preprocessing unit performs an image-to-image operation using a floodfill method on the droplet image from which the external noise has been removed to remove internal noise.

[0015] The above learning data is characterized by including a target vector, which is a contact angle, calculated from a droplet image generated based on an ellipse, and a droplet image vector, which is a low-level droplet image generated by lowering the resolution of the droplet image generated based on the ellipse.

[0016] In order to achieve the above-described purpose, a device for predicting a contact angle according to a preferred embodiment of the present invention includes a data processing unit that generates learning data including a droplet image vector, which is a low-level droplet image with reduced resolution, and a target vector representing a contact angle calculated in response to the droplet image vector, and a model generation unit that generates a learning model that predicts the contact angle of a droplet through weight calculation from the droplet image vector using the learning data.

[0017] The above data processing unit is characterized by generating a droplet image, calculating a contact angle of the generated droplet image, and then reducing the resolution of the generated droplet image to generate a droplet image vector, which is a low-level droplet image.

[0018] The above data processing unit is characterized in that it generates a droplet image having the overall shape of the droplet with a ratio of the major axis and minor axis within a predetermined range by cutting an ellipse horizontally based on the triple point of the droplet, and extracts a part of the droplet image having the overall shape to generate a droplet image including at least the triple point and the highest point.

[0019] The ratio of the major axis to the minor axis is characterized by being 1.00 to 0.80.

[0020] The above data processing unit uses mathematical formulas

[0021]

[0022] The contact angle corresponding to the droplet image is calculated according to the following, wherein θ is the contact angle of the droplet image, a is the major axis of the ellipse that is the basis of the droplet image, b is the minor axis of the ellipse that is the basis of the droplet image, x1 and y1 are the center coordinates of the ellipse that is the basis of the droplet image, and x and y are the coordinates of the triple point of the droplet image.

[0023] The data processing unit is characterized in that it generates a low-level droplet image having a second resolution that is lower than the first resolution while having the first size by resizing the droplet image having a first size and a first resolution corresponding to the first size.

[0024] The above model generation unit inputs the droplet image vector into a learning model, and the learning model performs a plurality of operations to which weights that have not been learned between multiple layers are applied to the droplet image vector to derive a prediction vector representing a predicted contact angle of a droplet image corresponding to the droplet image vector, and derives a loss representing the difference between the target vector and the prediction vector through a loss function, and performs optimization to update the weights of the learning model so that the loss is minimized.

[0025] In order to achieve the above-described object, a device for predicting a contact angle according to a preferred embodiment of the present invention includes a preprocessing unit for preprocessing an input droplet image to generate a droplet image vector when a droplet image of which a contact angle is unknown is input, a learning model trained using learning data including a learning droplet image vector which is a low-level droplet image with a reduced resolution and a target vector representing a contact angle calculated in response to the learning droplet image vector, and an inference unit for outputting the contact angle of the predicted vector when the learning model performs a plurality of operations to which weights learned between a plurality of layers are applied to the droplet image vector to derive a prediction vector that predicts the contact angle of a droplet of the droplet image.

[0026] The above preprocessing unit sequentially performs gray scale, Gaussian blur, and binarization on the input droplet image to remove external noise, and performs an image-to-image operation using a floodfill method on the droplet image from which the external noise has been removed to remove internal noise.

[0027] The above learning data is characterized by including a target vector, which is a contact angle, calculated from a droplet image generated based on an ellipse, and a droplet image vector, which is a low-level droplet image generated by lowering the resolution of the droplet image generated based on the ellipse.

[0028] According to the present invention, even in the case of low-level images, the contact angle can be accurately predicted by extracting droplet boundary characteristics through a learning model.

[0029] FIG. 1 is a drawing for explaining the configuration of a device for predicting a contact angle through extraction of droplet boundary characteristics of a low-level image based on a learning model according to an embodiment of the present invention.

[0030] FIG. 2 is a diagram illustrating a learning model for predicting a contact angle through extraction of droplet boundary characteristics of a low-level image based on a learning model according to an embodiment of the present invention.

[0031] Figure 3 is a flowchart illustrating a method for generating learning data for a learning model according to an embodiment of the present invention.

[0032] FIG. 4 is an exemplary diagram illustrating a method for generating learning data for a learning model according to an embodiment of the present invention.

[0033] Figure 5 is a flowchart illustrating a method for creating a learning model using learning data according to an embodiment of the present invention.

[0034] FIG. 6 is a flowchart illustrating a method for predicting a contact angle through extraction of droplet boundary characteristics of a low-level image based on a learning model according to an embodiment of the present invention.

[0035] FIG. 7 is an exemplary diagram illustrating a method for capturing a droplet image according to an embodiment of the present invention.

[0036] FIG. 8 is an exemplary diagram illustrating a method for predicting a contact angle through extraction of droplet boundary characteristics of a low-level image based on a learning model according to an embodiment of the present invention.

[0037] FIG. 9 is a drawing showing a computing device according to an embodiment of the present invention.

[0038] Before going into a detailed description of the present invention, it should be understood that the embodiments described in this specification and the configurations illustrated in the drawings are merely the most preferred embodiments of the present invention and do not represent all of the technical ideas of the present invention, and therefore, there may be various equivalents and modified examples that can replace them at the time of filing this application.

[0039] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. It should be noted that, where possible, identical components are represented by identical reference numerals throughout the drawings. Furthermore, detailed descriptions of well-known functions and structures that may obscure the gist of the present invention will be omitted. For the same reason, some components in the attached drawings are exaggerated, omitted, or schematically depicted, and the sizes of each component do not fully reflect their actual sizes.

[0040] In addition, the terms and words used in the present specification and claims described below should not be interpreted as limited to their usual or dictionary meanings, but should be interpreted as meanings and concepts that conform to the technical idea of ​​the present invention based on the principle that the inventor can appropriately define the concept of the term to explain his or her own invention in the best way.

[0041] First, a device for predicting a contact angle through extraction of droplet boundary characteristics from a low-level image based on a learning model according to an embodiment of the present invention will be described. Fig. 1 is a diagram for explaining the configuration of a device for predicting a contact angle through extraction of droplet boundary characteristics from a low-level image based on a learning model according to an embodiment of the present invention. Fig. 2 is a diagram for explaining a learning model for predicting a contact angle through extraction of droplet boundary characteristics from a low-level image based on a learning model according to an embodiment of the present invention.

[0042] Referring to FIG. 1, a prediction device (10) according to an embodiment of the present invention includes a data processing unit (100), a model generation unit (200), a preprocessing unit (300), and an inference unit (400).

[0043] The data processing unit (100) is for generating learning data according to an embodiment of the present invention. The learning data includes a droplet image vector, which is a low-level droplet image, and a target vector, which is a contact angle, obtained from a high-level droplet image corresponding to the droplet image vector. Here, the high-level droplet image means an image having a resolution greater than the size of the image, and the low-level droplet image means an image having a resolution less than the size of the image. The data processing unit (100) generates a high-level droplet image, calculates a target vector, which is a contact angle, from the high-level droplet image, and then reduces the resolution of the droplet image to generate a droplet image vector, which is a low-level droplet image. Here, the droplet image vector can be a binary representation of the pixel values ​​of the droplet image.

[0044] The model generation unit (200) is for generating a learning model (LM) through learning (Deep Learning) using learning data. The model generation unit (200) generates a learning model (LM) having learned weights by learning (Deep Learning) a prototype of a learning model (LM) having unlearned weights using learning data. According to an embodiment of the present invention, the learning model (LM) that has completed learning can receive a droplet image vector, which is an image of a droplet, and predict the contact angle of the droplet image to produce a prediction vector, which is the predicted contact angle. The learning model (LM) may be a convolutional neural network (CNN). The learning model (LM) may include an input layer (IL), at least one pair of convolutional layers (CL) and pooling layers (PL) that are alternately repeated, at least one fully-connected layer (FL), and an output layer (OL). An example of such a learning model (ML), which is a convolutional neural network, is shown in Figure 2.

[0045] As illustrated, the learning model (LM) includes multiple layers, namely, sequentially, an input layer (IL), a convolutional layer (CL), a pooling layer (PL), a fully connected layer (FL), and an output layer (OL). Each of the multiple layers receives a weighted value based on the computational results of the previous layer, performs an operation on the weighted input value using a predetermined activation function, and transmits the computational result to the next layer. Then, the computational result of the final layer is output as the output value of the learning model (LM), i.e., a prediction vector. This prediction vector represents the predicted contact angle of the droplet.

[0046] The input layer (IL) serves as a buffer for receiving a droplet image vector, which is a droplet image, and is formed according to the specifications of the droplet image so that each of a plurality of pixels of the input droplet image vector maintains position information (e.g., pixel coordinates).

[0047] The convolution layer (CL) is composed of at least one first feature map (FM 1: Feature Map). When a droplet image vector is input to the input layer (IL), the convolution layer (CL) performs a convolution operation using the first filter (or kernel) on the input data of the input layer (IL), applies weights, and performs an operation using an activation function to derive multiple first feature maps (FM1).

[0048] The pooling layer (PL) is composed of at least one second feature map (FM 2: Feature Map 2). The pooling layer (PL) performs a pooling (or sub-sampling) operation using a second filter (or kernel) on at least one first feature map (FM1) of the convolutional layer (CL) to derive at least one second feature map (FM2).

[0049] A fully connected layer (FL) is composed of a plurality of operation nodes (f1 to fn). The plurality of operation nodes (f1 to fn) of the fully connected layer (FL) produce a plurality of operation values ​​through an operation by an activation function on at least one second feature map (FM2) of the pooling layer (PL). Optionally, a flattening layer (FLL: FLatten Layer) may be interposed between the pooling layer (PL) and the fully connected layer (FL), which changes the shape of the values ​​of the previous layer to match the shape of the values ​​of the next layer and inputs them to the next layer. That is, the flattening layer (FLL) changes the values ​​of the second feature map (FM2) in the matrix form of the pooling layer (PL) to match the shape of the values ​​of the operation nodes of the fully connected layer (FL) and inputs them. For example, if the second feature map of the pooling layer (PL) is in the form of (row × column × channel), the flattening layer (FLL) can change it into the form of (column × channel), which is the form of the operation node of the fully connected layer (FL), and input it to the operation node.

[0050] The output layer (OL) includes one output node (O). Each of the plurality of operation nodes (f1 to fn) of the fully connected layer (FL) is connected to the output node (O) of the output layer (OL) by a channel having a weight (W). In other words, the plurality of operation values ​​of the plurality of operation nodes (f1 to fn) are input to the output node (O) with the weights applied. Accordingly, the output node (O) of the output layer (OL) produces a prediction vector through an operation by an activation function for the plurality of operation values ​​to which the weights of the fully connected layer (FL) are applied. This prediction vector represents the contact angle of the droplet of the droplet image vector, which is the input droplet image.

[0051] Activation functions used in operations in the aforementioned convolutional layer (CL), pooling layer (PL), fully connected layer (FL), and output layer (OL) include sigmoid, hyperbolic tangent (tanh), exponential linear unit (ELU), rectified linear unit (ReLU), leaky ReLU, Maxout, Minout, and softmax. The convolutional layer (CL), fully connected layer (FL), and output layer (OL) select one of these activation functions to perform operations.

[0052] Referring back to FIG. 1, the preprocessing unit (300) is for generating a droplet image vector through preprocessing on the captured droplet image. The preprocessing unit (300) sequentially performs gray scale, Gaussian blur, and binarization on the input droplet image to remove external noise. Then, the preprocessing unit (300) removes internal noise by performing an image-to-image operation using a floodfill method on the droplet image from which external noise has been removed. Subsequently, the droplet image from which internal noise has been removed is vertically cropped, and an image obtained by extracting the cropped left portion can be generated as a droplet image vector.

[0053] The inference unit (400) analyzes the droplet image vector using the learning model (LM) for which learning has been completed to derive a prediction vector that predicts the contact angle of the droplet in the droplet image.

[0054] Next, a method for generating training data for a learning model according to an embodiment of the present invention will be described. Figure 3 is a flowchart illustrating a method for generating training data for a learning model according to an embodiment of the present invention. Figure 4 is an exemplary diagram illustrating a method for generating training data for a learning model according to an embodiment of the present invention.

[0055] Referring to FIG. 3, the data processing unit (100) generates a high-level droplet image at step S110. Here, a high-level droplet image means an image with a resolution greater than or equal to the size of the image. For example, if the size of the image is 300×300 pixels, it means an image with a resolution greater than or equal to 300×300 pixels.

[0056] At this time, the data processing unit (100) horizontally cuts the ellipse based on the triple point of the droplet to generate a high-level droplet image having the entire shape of the droplet with a ratio of the major axis to the minor axis within a predetermined range. In this droplet image, the inside of the droplet boundary is 0 (black), and the outside is 1 (white). Then, the data processing unit (100) extracts a portion of the droplet image having the entire shape to generate a droplet image including at least the triple point and the highest point.

[0057] For example, as illustrated in Fig. 4, an ellipse (e) having a ratio of a major axis (a) to a minor axis (b) within a predetermined range is prepared. At this time, the ratio of the major axis to the minor axis is preferably 1.00 to 0.80. Then, after setting the triple point Q(x, y) of the droplet, a horizontal cut is made based on the set triple point Q(x, y) to generate a droplet image (fdi) having the entire shape of the droplet having a ratio of a major axis to a minor axis within a predetermined range.

[0058] Next, the data processing unit (100) calculates the target vector, which is the contact angle, from the high-level droplet image in step S120. At this time, the data processing unit (100) can calculate the contact angle according to the following mathematical expression 1.

[0059]

[0060] Here, θ represents the contact angle of the droplet image, i.e., the target vector. Referring to Fig. 4, a represents the major axis of the ellipse that is the basis of the droplet image, and b represents the minor axis of the ellipse that is the basis of the droplet image. In addition, x1 and the above y1 represent the center coordinates P(x1, y1) of the ellipse that is the basis of the droplet image, and x and the above y represent the coordinates Q(x, y) of the triple point of the droplet image.

[0061] Next, the data processing unit (100) reduces the resolution of the droplet image in step S130 to generate a droplet image vector, which is a low-level droplet image. A low-level droplet image means an image with a resolution lower than the size of the image. For example, if the size of the image is 300×300 pixels, it means an image with a resolution lower than 300×300 pixels.

[0062] At this time, the data processing unit (100) can generate a low-level droplet image having a first size and a second resolution that is lower than the first resolution by resizing a high-level droplet image having a first size and a first resolution corresponding to the first size.

[0063] For example, assume that the first size of the image is 300×300 pixels and the first resolution is 300×300 pixels. Here, while maintaining the first size of the image as 300×300 pixels, a low-level droplet image can be generated by resizing the image to drop the first resolution to the second resolution, 50×50 pixels.

[0064] Next, a method for creating a learning model using learning data according to an embodiment of the present invention will be described. Figure 5 is a flowchart illustrating a method for creating a learning model using learning data according to an embodiment of the present invention.

[0065] Referring to FIG. 5, the model generation unit (200) can receive learning data generated as described with reference to FIGS. 3 and 4 described above in step S210. When the learning data is input, the model generation unit (200) inputs the droplet image vector into the learning model (LM) in step S220.

[0066] Then, the learning model (LM) performs multiple operations in step S230, in which multiple inter-layer learning weights are applied to the droplet image vector, to derive a prediction vector representing the predicted contact angle of the droplet of the droplet image corresponding to the droplet image vector.

[0067] Accordingly, the model generation unit (200) can calculate a loss representing the difference between the predicted vector and the target vector through a loss function at step S240. That is, the model generation unit (100) can calculate a loss representing the difference between the predicted contact angle of the predicted vector and the calculated contact angle of the target vector through a loss function.

[0068] Next, the model generation unit (200) performs optimization to update the weights of the learning model (LM) so that the loss derived through the loss function in step S250 is minimized.

[0069] The above-described steps S220 to S250 are repeatedly performed using a plurality of different learning data, and the weights of the learning model (LM) are repeatedly updated according to this repetition. And this repetition is repeated until a predetermined condition is satisfied. According to one embodiment, the predetermined condition may be a case where the loss converges and is below a predetermined value. Therefore, the model generation unit (100) determines in step S260 whether the loss calculated previously (S240) converges and is below a preset target value, and if the loss is below the preset target value, the learning for the detection model (DM) is completed in step S270.

[0070] Next, a method for predicting a contact angle through droplet boundary feature extraction of a low-level image based on a learning model according to an embodiment of the present invention will be described. Fig. 6 is a flowchart for explaining a method for predicting a contact angle through droplet boundary feature extraction of a low-level image based on a learning model according to an embodiment of the present invention. Fig. 7 is an exemplary diagram for explaining a method for capturing a droplet image according to an embodiment of the present invention. Fig. 8 is an exemplary diagram for explaining a method for predicting a contact angle through droplet boundary feature extraction of a low-level image based on a learning model according to an embodiment of the present invention.

[0071] Referring to FIG. 6, the preprocessing unit (300) can receive a droplet image with an unknown contact angle at step S310. This droplet image is captured by a camera (73) while illuminating the droplet (D) with light from a light source (72) when the droplet (D) is dropped on a substrate (S) via a needle (71), as illustrated in FIG. 8. In this way, the droplet image generated through the capture has an unknown contact angle.

[0072] When a droplet image is input, the preprocessing unit (300) preprocesses the input droplet image in step S320 to generate a droplet image vector. At this time, as illustrated in FIG. 8, the preprocessing unit (300) sequentially performs gray scale, Gaussian blur, and binarization on the input droplet image to remove external noise. Then, the preprocessing unit (300) removes internal noise by performing an image-to-image operation using a floodfill method on the droplet image from which external noise has been removed. Next, the droplet image from which internal noise has been removed is vertically cropped, and an image obtained by extracting the cropped left portion is generated as a droplet image vector.

[0073] When the droplet image vector is generated, the inference unit (400) inputs the droplet image vector into the learning model (LM) learned in the manner described through FIG. 5 in step S330.

[0074] Then, the learning model (LM) performs multiple operations in which weights learned across multiple layers are applied to the droplet image vector in step S340 to derive a prediction vector that predicts the contact angle of the droplet corresponding to the droplet image vector. Then, the inference unit (400) outputs the contact angle of the prediction vector in step S350.

[0075] FIG. 9 is a diagram illustrating a computing device according to an embodiment of the present invention. The computing device (TN100) of FIG. 9 may be a device described herein, for example, a prediction device (10).

[0076] In the embodiment of FIG. 9, the computing device (TN100) may include at least one processor (TN110), a transceiver (TN120), and a memory (TN130). In addition, the computing device (TN100) may further include a storage device (TN140), an input interface device (TN150), an output interface device (TN160), and the like. The components included in the computing device (TN100) may be connected by a bus (TN170) to communicate with each other.

[0077] The processor (TN110) can execute program commands stored in at least one of the memory (TN130) and the storage device (TN140). The processor (TN110) may refer to a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor in which methods according to embodiments of the present invention are performed. The processor (TN110) may be configured to implement procedures, functions, methods, etc. described in relation to embodiments of the present invention. The processor (TN110) may control each component of the computing device (TN100).

[0078] The memory (TN130) and the storage device (TN140) can each store various information related to the operation of the processor (TN110). The memory (TN130) and the storage device (TN140) can each be configured with at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory (TN130) can be configured with at least one of a read-only memory (ROM) and a random access memory (RAM).

[0079] The transceiver (TN120) can transmit or receive wired or wireless signals. The transceiver (TN120) can be connected to a network to perform communication.

[0080] In particular, the data processing unit (100), model generation unit (200), preprocessing unit (300), and inference unit (400) according to an embodiment of the present invention may be implemented in the form of a program readable by a computing device, stored in a memory (TN130), and then executed by a processor (TN110), or implemented as a lower module of the processor (TN110).

[0081] Meanwhile, the method according to the embodiment of the present invention described above may be implemented in the form of a program readable by various computer means and recorded on a computer-readable recording medium. Here, the recording medium may include program commands, data files, data structures, etc., singly or in combination. The program commands recorded on the recording medium may be those specially designed and configured for the present invention, or may be those known and usable by those skilled in the art of computer software. For example, the recording medium includes magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands, such as ROMs, RAMs, and flash memories. Examples of program commands may include not only machine language wires generated by a compiler, but also high-level language wires that can be executed by a computer using an interpreter, etc. These hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

[0082] While the present invention has been described using several preferred embodiments, these embodiments are illustrative and not limiting. As such, those skilled in the art will appreciate that various changes and modifications can be made in accordance with the doctrine of equivalents without departing from the spirit of the invention and the scope of the claims.

Claims

1. In a method for predicting a contact angle, A step of generating learning data including a droplet image vector, which is a low-level droplet image with reduced resolution, and a target vector representing a contact angle calculated in response to the droplet image vector; and A step of generating a learning model that predicts the contact angle of a droplet by calculating a weight from the droplet image vector using the learning data by the model generating unit; characterized by including A method for predicting contact angle.

2. In paragraph 1, The steps for generating the above learning data are A step in which the above data processing unit generates a droplet image; The step of the data processing unit calculating the contact angle of the droplet image; and A step in which the data processing unit reduces the resolution of the droplet image to generate a droplet image vector, which is a low-level droplet image; characterized by including A method for predicting contact angle.

3. In paragraph 2, The step of generating the above droplet image vector is The step of the above data processing unit generating a droplet image having the entire shape of the droplet with a ratio of the major axis and minor axis within a predetermined range by cutting the ellipse horizontally based on the triple point of the droplet; and A step in which the data processing unit extracts a part of the droplet image having the entire shape and generates a droplet image including at least a triple point and a highest point; characterized by including A method for predicting contact angle.

4. In paragraph 3, The ratio of the above major and minor axes is characterized by being 1.00 to 0.80 A method for predicting contact angle.

5. In paragraph 2, The step of calculating the contact angle of the above droplet image is The above data processing unit Mathematical formula The contact angle corresponding to the above droplet image is calculated according to The above θ is the contact angle of the droplet image, The above a is the major axis of the ellipse that is the basis of the droplet image, The above b is the short axis of the ellipse that is the basis of the droplet image, The above x1 and y1 are the center coordinates of the ellipse that is the basis of the droplet image, The above x and y are characterized in that they are coordinates of the triple point of the droplet image. A method for predicting contact angle.

6. In paragraph 2, The step of generating the droplet image vector, which is the low-level droplet image above, is The data processing unit is characterized in that it generates a low-level droplet image having a second resolution that is lower than the first resolution while having the first size by resizing the droplet image having a first size and a first resolution corresponding to the first size. A method for predicting contact angle.

7. In paragraph 1, The steps for creating the above learning model are A step in which the above model generation unit inputs the above droplet image vector into the learning model; A step of the learning model performing a plurality of operations to which weights for which learning has not been completed between multiple layers are applied to the droplet image vector to derive a prediction vector representing a predicted contact angle of the droplet image corresponding to the droplet image vector; A step in which the above model generation unit derives a loss representing the difference between the target vector and the prediction vector through a loss function; A step of performing optimization by which the model generation unit updates the weights of the learning model so that the loss is minimized; characterized by including A method for predicting contact angle.

8. In a method for predicting the contact angle, A step of preprocessing a droplet image of which a contact angle is unknown when a droplet image is input, and generating a droplet image vector by preprocessing the input droplet image; A step of inputting the droplet image vector into a learning model learned using learning data including a learning droplet image vector, which is a low-level droplet image with a reduced resolution, and a target vector representing a contact angle calculated in response to the learning droplet image vector; A step of the learning model performing multiple operations to which weights learned between multiple layers are applied to the droplet image vector to derive a prediction vector predicting the contact angle of the droplet of the droplet image; and Step where the inference unit outputs the contact angle of the above prediction vector: characterized by including A method for predicting contact angle.

9. In paragraph 8, The step of generating the above droplet image vector is The above preprocessing unit sequentially performs gray scale, Gaussian blur, and binarization on the input droplet image to remove external noise; and A step in which the above preprocessing unit performs an image-to-image operation using a floodfill method on the droplet image from which the external noise has been removed to remove internal noise; characterized by including A method for predicting contact angle.

10. In paragraph 8, The above learning data is It is characterized by including a target vector which is a contact angle calculated from a droplet image generated based on an ellipse and a droplet image vector which is a low-level droplet image generated by lowering the resolution of the droplet image generated based on the ellipse. A method for predicting contact angle.

11. In a device for predicting a contact angle, A data processing unit that generates learning data including a droplet image vector, which is a low-level droplet image with reduced resolution, and a target vector representing a contact angle calculated in response to the droplet image vector; and A model generation unit that generates a learning model that predicts the contact angle of a droplet through weight calculation from the droplet image vector using the above learning data; characterized by including A device for predicting contact angle.

12. In paragraph 11, The above data processing unit Generate droplet images, After calculating the contact angle of the generated droplet image, It is characterized by generating a droplet image vector, which is a low-level droplet image, by lowering the resolution of the generated droplet image. A device for predicting contact angle.

13. In paragraph 12, The above data processing unit A droplet image having the overall shape of a droplet with a ratio of the major axis to the minor axis within a predetermined range is created by cutting the ellipse horizontally based on the triple point of the droplet, A method characterized in that a part of a droplet image having the above overall shape is extracted to generate a droplet image including at least a triple point and a highest point. A device for predicting contact angle.

14. In paragraph 13, The ratio of the above major and minor axes is characterized by being 1.00 to 0.80 A device for predicting contact angle.

15. In paragraph 12, The above data processing unit Mathematical formula The contact angle corresponding to the above droplet image is calculated according to The above θ is the contact angle of the droplet image, The above a is the major axis of the ellipse that is the basis of the droplet image, The above b is the short axis of the ellipse that is the basis of the droplet image, The above x1 and y1 are the center coordinates of the ellipse that is the basis of the droplet image, The above x and y are characterized in that they are coordinates of the triple point of the droplet image. A device for predicting contact angle.

16. In paragraph 12, The above data processing unit A low-level droplet image having a second resolution lower than the first resolution but having the first size is generated by resizing the droplet image having a first size and a first resolution corresponding to the first size. A device for predicting contact angle.

17. In paragraph 11, The above model generation part Input the above droplet image vector into the learning model, When the learning model performs multiple operations to which weights that have not been learned between multiple layers are applied to the above droplet image vector, a prediction vector representing the predicted contact angle of the droplet image corresponding to the above droplet image vector is derived. A loss function is used to derive a loss that represents the difference between the target vector and the predicted vector. It is characterized by performing optimization to update the weights of the learning model so that the loss is minimized. A device for predicting contact angle.

18. In a device for predicting a contact angle, When a droplet image of unknown contact angle is input, a preprocessing unit that preprocesses the input droplet image to generate a droplet image vector; and The droplet image vector is input into a learning model trained using learning data including a learning droplet image vector, which is a low-level droplet image with reduced resolution, and a target vector representing a contact angle calculated in response to the learning droplet image vector. When the learning model performs multiple operations to which weights learned between multiple layers are applied to the above droplet image vector, a prediction vector is derived that predicts the contact angle of the droplet in the above droplet image. Inference unit that outputs the contact angle of the above prediction vector: characterized by including A device for predicting contact angle.

19. In Article 18, The above preprocessing unit Gray scale, Gaussian blur, and binarization are sequentially performed on the input droplet image to remove external noise. It is characterized in that the internal noise is removed by performing an image-to-image operation using the floodfill method on the droplet image from which the external noise has been removed. A device for predicting contact angle.

20. In paragraph 18, The above learning data is It is characterized by including a target vector which is a contact angle calculated from a droplet image generated based on an ellipse and a droplet image vector which is a low-level droplet image generated by lowering the resolution of the droplet image generated based on the ellipse. A device for predicting contact angle.

Images