Method and Apparatus for Increasing Data Using a Moire pattern image of a Printed Circuit Board

Abstract

A method and apparatus for increasing data using a moiré pattern image of a printed circuit board are disclosed. According to one embodiment of the present disclosure, a method for increasing data may include the steps of acquiring a moiré pattern image of a printed circuit board, identifying a reflection area within the moiré pattern image and a shadow area within the moiré pattern image, and increasing the data based on at least one of the reflection area, the shadow area, a predetermined group, and a grid.

Description

Technology Field

[0001] The present invention relates to a method and apparatus for increasing data using a moiré pattern image of a printed circuit board. More specifically, the invention relates to a method and apparatus for increasing data by considering shadow areas and reflection areas within a moiré pattern image of a printed circuit board. Background Technology

[0002] The following description merely provides background information related to the present embodiment and does not constitute prior art.

[0003] A Printed Circuit Board (PCB) is a board that forms an electronic circuit by fixing electronic components, such as resistors, capacitors, and integrated circuits, to the surface of the board and connecting them with copper wiring. Equipment for inspecting PCBs requires precise and fast algorithms and demands a high level of accuracy. Artificial neural networks are a form of artificial intelligence technology and are machine learning models created in software to mimic the structure of human neurons. There is an increasing need to apply artificial neural networks to equipment for inspecting PCBs. However, when applying artificial neural networks to such equipment, training may be difficult due to a lack of training data and the characteristics of the PCBs. Accordingly, to apply artificial neural networks to PCB inspection equipment, it is necessary to increase the training data by taking into account the characteristics of the PCBs. The problem to be solved

[0004] The present disclosure can increase training data by using a moiré pattern image of a printed circuit board (PCB).

[0005] In addition, according to one embodiment, the training data can be increased by considering the reflection area and the shadow area within the moiré pattern image of the printed circuit board.

[0006] In addition, according to one embodiment, the height of the printed circuit board can be accurately measured.

[0007] In addition, according to one embodiment, the performance of the equipment for inspecting printed circuit boards can be improved.

[0008] The problems that the present invention aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0009] A method for increasing data according to the present disclosure may include the steps of acquiring a moiré pattern image of a printed circuit board, identifying a reflection area and a shadow area within the moiré pattern image, and increasing the data based on at least one of the reflection area, the shadow area, a predetermined group, and a grid.

[0010] According to the present disclosure, an apparatus for increasing data comprises a memory and at least one processor, wherein the at least one processor acquires a moiré pattern image of a printed circuit board, identifies a reflection area and a shadow area within the moiré pattern image, and can increase the data based on at least one of the reflection area, the shadow area, a predetermined group, and a grid.

[0011] A computer program according to the present disclosure may be stored in a recording medium readable by a computing device to execute the steps of acquiring a moiré pattern image of a printed circuit board, identifying a reflection area and a shadow area within the moiré pattern image, and increasing data based on at least one of the reflection area, the shadow area, a predetermined group, and a grid.

[0012] A computer-readable recording medium according to the present disclosure is a computer-readable recording medium in which instructions are stored, wherein the instructions, when executed by the computer, may cause the computer to perform the steps of acquiring a moiré pattern image of a printed circuit board, identifying a reflection area within the moiré pattern image and a shadow area within the moiré pattern image, and increasing data based on at least one of the reflection area, the shadow area, a predetermined group, and a grid. Effects of the invention

[0013] According to the present disclosure, there is an effect of increasing training data by using a moiré pattern image of a printed circuit board (PCB).

[0014] In addition, according to one embodiment, there is an effect of increasing training data by considering reflection areas and shadow areas within the moiré pattern image of a printed circuit board.

[0015] In addition, according to one embodiment, there is an effect of being able to accurately measure the height of a printed circuit board.

[0016] In addition, according to one embodiment, there is an effect of improving the performance of equipment for inspecting printed circuit boards.

[0017] The effects obtainable from the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art to which the present disclosure pertains from the description below. Brief explanation of the drawing

[0018] FIG. 1 is a diagram illustrating a process of augmenting training data using a moiré pattern image of a printed circuit board (PCB) according to one embodiment of the present disclosure. FIGS. 2a and 2b are drawings for explaining a method for obtaining a moiré pattern image of a printed circuit board and the obtained moiré pattern image according to one embodiment of the present disclosure. FIG. 3 is a drawing for explaining a method for generating a 3D model of a printed circuit board according to one embodiment of the present disclosure. FIG. 4 is a drawing for explaining a method for generating a 3D model of a printed circuit board according to another embodiment of the present disclosure. FIGS. 5A and 5B are drawings for illustrating a reflection area and a shadow area within a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure, and a 3D model of the printed circuit board created using such a moiré pattern image. FIGS. 6a and 6b are drawings for explaining the cause of reflection areas and shadow areas occurring within a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure. FIG. 7 is a drawing for explaining a method of increasing training data using images according to one embodiment of the present disclosure. FIG. 8 is a drawing for explaining a method of increasing training data using a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure. FIGS. 9a and 9b are drawings for explaining a method of increasing training data using reflection areas and shadow areas within a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure. FIGS. 10a and FIGS. 10b are drawings for explaining a method of increasing training data by dividing a moiré pattern image of a printed circuit board into a grid according to one embodiment of the present disclosure. FIGS. 11a and FIGS. 11b are drawings for illustrating a method of increasing training data by dividing a moiré pattern image of a printed circuit board into a grid according to another embodiment of the present disclosure. FIGS. 12a and 12b are drawings for illustrating an original image of a component in a printed circuit board and a reflective area in a moiré pattern image of a component in a printed circuit board according to one embodiment of the present disclosure. FIGS. 13a and FIGS. 13b are drawings for explaining a method of increasing training data by dividing a moiré pattern image of a component in a printed circuit board into a grid, according to one embodiment of the present disclosure. FIG. 14 is a drawing for explaining a method of increasing training data according to one embodiment of the present disclosure. Specific details for implementing the invention

[0019] Some embodiments of the present disclosure are described in detail below with reference to exemplary drawings. It should be noted that in assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings. Furthermore, in describing the present disclosure, if it is determined that a detailed description of related known components or functions could obscure the essence of the present disclosure, such detailed description is omitted.

[0020] In describing the components of the embodiments according to the present disclosure, symbols such as first, second, i), ii), a), b), etc., may be used. These symbols are intended only to distinguish the components from other components, and the essence, order, or sequence of the components is not limited by the symbols. When a part in the specification is described as 'comprising' or 'having' a component, this means that, unless explicitly stated otherwise, it does not exclude other components but may include additional components.

[0021] The detailed description set forth below, together with the accompanying drawings, is intended to describe exemplary embodiments of the present disclosure and is not intended to represent the only embodiment in which the present disclosure can be practiced.

[0022] FIG. 1 is a diagram illustrating a process of augmenting training data using a moiré pattern image of a printed circuit board (PCB) according to one embodiment of the present disclosure.

[0023] Referring to FIG. 1, an original image and a moiré pattern image of a printed circuit board can be obtained (S110). The original image and the moiré pattern image of the printed circuit board can be obtained using an optical system. Using the original image of the printed circuit board, rectangular and circular components can be classified. Data can be augmented using the moiré pattern image of the printed circuit board (S120). Features of components within the printed circuit board can be extracted. Data can be augmented using these features of components. The augmented data can be used as training data (S130). The augmented data can be used as training data for a training model that generates a 3D model of the printed circuit board.

[0024] FIGS. 2a and 2b are drawings for explaining a method for obtaining a moiré pattern image of a printed circuit board and the obtained moiré pattern image according to one embodiment of the present disclosure.

[0025] Referring to FIGS. 2a and 2b, a printed circuit board can be photographed using an optical camera. When photographing the printed circuit board with the optical camera, a pattern projector can project a pattern onto the printed circuit board. The height of a component within the printed circuit board may correspond to H. H can be calculated using H = △px sinθ, where θ corresponds to the angle between the printed circuit board and the projected pattern. Through this method, a moiré pattern image of the printed circuit board in FIG. 2b can be obtained.

[0026] FIG. 3 is a drawing for explaining a method for generating a 3D model of a printed circuit board according to one embodiment of the present disclosure.

[0027] Referring to FIG. 3, an original image of a printed circuit board and a moiré pattern image of the printed circuit board can be input into a learning model that generates a 3D model of the printed circuit board. Such a learning model may correspond to a deep learning-based model. The device according to the present disclosure may additionally include a learning unit (not shown) for pre-training the learning model. The learning unit may pre-train the learning model using supervised learning, unsupervised learning, semi-supervised learning, and / or reinforcement learning. Here, the specific method by which the learning unit trains the learning model based on learning data is common in the field, so a detailed description is omitted.

[0028] When the original image of the printed circuit board and the moiré pattern image of the printed circuit board are input into a training model that generates a 3D model of the printed circuit board, a 3D model of the printed circuit board can be output. The training model that generates the 3D model of the printed circuit board may correspond to a neural network model. Through the 3D model of the printed circuit board, the heights of the components within the printed circuit board can be verified.

[0029] FIG. 4 is a drawing for explaining a method for generating a 3D model of a printed circuit board according to another embodiment of the present disclosure.

[0030] Referring to FIG. 4, the learning model for generating the 3D model of the printed circuit board described in FIG. 3 may correspond to a network of encoder structures and decoder structures. The RGB (Red Green Blue) of the original image of the printed circuit board and the depth value of the original image of the printed circuit board may be inputs. The RGB of the moiré pattern image of the printed circuit board and the depth value of the moiré pattern image of the printed circuit board may be inputs. Accordingly, the 3D model of the printed circuit board can be restored.

[0031] FIGS. 5A and 5B are drawings for illustrating a reflection area and a shadow area within a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure, and a 3D model of the printed circuit board created using such a moiré pattern image.

[0032] Referring to FIG. 5a, depending on the characteristics of the components within the printed circuit board, there may be reflective areas within the moiré pattern image of the printed circuit board. Depending on the characteristics of the components within the printed circuit board, there may be shadow areas within the moiré pattern image of the printed circuit board.

[0033] Referring to Fig. 5b, when the original image of the printed circuit board and the moiré pattern image of the printed circuit board are input into a training model that generates a 3D model of the printed circuit board, a 3D model of the printed circuit board can be output. Here, it can be observed that there is an error in the height of some parts measured in the 3D model of the printed circuit board. Such errors may be caused by reflective and shadowed areas within the moiré pattern image of the printed circuit board. If pixel values ​​similar to those of surrounding parts are assigned to the parts corresponding to the reflective and shadowed areas within the moiré pattern image of the printed circuit board, such errors may not occur.

[0034] FIGS. 6a and 6b are drawings for explaining the cause of reflection areas and shadow areas occurring within a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure.

[0035] Referring to FIG. 6a, a printed circuit board can be photographed using an optical camera. When photographing the printed circuit board with the optical camera, a pattern projector can project a pattern onto the printed circuit board. Tall components within the printed circuit board may obstruct the pattern projected by the pattern projector, so the pattern may not be projected onto the short components. A shadow area may be created on the short components behind the tall components within the printed circuit board. Consequently, a shadow area may be generated within the moiré pattern image of the printed circuit board. The shadow area may correspond to an area that appears dark.

[0036] Referring to FIG. 6b, a printed circuit board can be photographed using an optical camera. When photographing the printed circuit board with the optical camera, a pattern projector can project a pattern onto the printed circuit board. Due to the characteristics of the components within the printed circuit board, a reflection phenomenon may occur on those components. Due to this reflection phenomenon, a reflective area may be created within the moiré pattern image of the printed circuit board. The reflective area may correspond to a brightly shining white area. The height of the reflective area may not be measurable.

[0037] FIG. 7 is a drawing for explaining a method of increasing training data using images according to one embodiment of the present disclosure.

[0038] Referring to Fig. 7, when an image is training data, there may be methods to increase the training data. For example, the training data image may be an image containing a dog. There may be methods to increase the training data by cropping and resizing the image. There may be methods to increase the training data by cropping, resizing, and flipping the image. There may be methods to increase the training data by rotating the image. There may be methods to increase the training data by distorting the colors of the image. There may be methods to increase the training data by cutting out the image. There may be methods to increase the training data by applying Gaussian noise to the image. There may be methods to increase the training data by applying Gaussian blur to the image. There may be methods to increase the training data by applying a Sobel filter to the image. Here, the most effective methods are cropping and resizing images to increase training data, and rotating images to increase training data.

[0039] FIG. 8 is a diagram illustrating a method for increasing training data using a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure. It may not be suitable to apply the method for increasing training data described in FIG. 7 directly to the moiré pattern image of a printed circuit board. This is due to the characteristics of the moiré pattern image of the printed circuit board.

[0040] Referring to Fig. 8, when a moiré pattern image is training data, there may be a method to increase the training data. For example, the moiré pattern image serving as training data may correspond to a moiré pattern image of a printed circuit board. There may be a method to increase the training data by cropping and resizing the moiré pattern image of the printed circuit board. There may be a method to increase the training data by cropping, resizing, and flipping the moiré pattern image of the printed circuit board. There may be a method to increase the training data by rotating the moiré pattern image of the printed circuit board. There may be a method to increase the training data by cutting out the moiré pattern image of the printed circuit board.

[0041] FIGS. 9a and 9b are drawings for explaining a method of increasing training data using reflection areas and shadow areas within a moiré pattern image of a printed circuit board according to one embodiment of the present disclosure.

[0042] Referring to Fig. 9a, dark areas within the moiré pattern image of the printed circuit board may correspond to shadow areas. Bright areas within the moiré pattern image of the printed circuit board may correspond to reflection areas. In the 3D model of the printed circuit board, height measurements may be impossible due to height errors in the parts corresponding to the shadow and reflection areas.

[0043] Referring to Fig. 9b, shadow and reflection areas within the moiré pattern image of a printed circuit board can be cropped to a certain size. For example, there may be a method to increase training data by flipping the shadow and reflection areas horizontally or vertically. For example, there may be a method to increase training data by flipping the parts that are not shadow and reflection areas horizontally or vertically.

[0044] FIGS. 10a and FIGS. 10b are drawings for explaining a method of increasing training data by dividing a moiré pattern image of a printed circuit board into a grid according to one embodiment of the present disclosure.

[0045] Referring to FIG. 10a, the moiré pattern image of a printed circuit board can be divided into multiple grids of a certain size. There may be a method to increase training data by switching the grids with each other within the moiré pattern image of the printed circuit board.

[0046] Referring to FIG. 10b, numbers can be assigned to multiple grids of a certain size in FIG. 10a. The grids can be switched with each other using the numbers assigned to each grid.

[0047] FIGS. 11a and FIGS. 11b are drawings for illustrating a method of increasing training data by dividing a moiré pattern image of a printed circuit board into a grid according to another embodiment of the present disclosure.

[0048] Referring to FIG. 11a, the moiré pattern image of a printed circuit board can be divided into multiple grids of a certain size. If the training data is increased simply by switching these grids with each other, the moiré pattern may be damaged. Accordingly, grids corresponding to the areas where the moiré pattern is applied and grids corresponding to the areas where the moiré pattern is not applied can be distinguished. There may be a method to increase the training data by switching the grids corresponding to the areas where the moiré pattern is applied with each other. There may also be a method to increase the training data by switching the grids corresponding to the areas where the moiré pattern is not applied with each other.

[0049] Referring to FIG. 11b, numbers can be assigned to multiple grids of a fixed size in FIG. 11a. For example, a grid assigned a red number may correspond to a grid corresponding to an area with a moiré pattern. A grid assigned a blue number may correspond to a grid corresponding to an area without a moiré pattern. Training data can be increased by switching between grids assigned red numbers. Training data can be increased by switching between grids assigned blue numbers.

[0050] FIGS. 12a and 12b are drawings for illustrating an original image of a component in a printed circuit board and a reflective area in a moiré pattern image of a component in a printed circuit board according to one embodiment of the present disclosure.

[0051] Referring to FIGS. 12a and 12b, an original image of a component within a printed circuit board can be obtained using an optical camera. The component within the printed circuit board may correspond to a circular component. A pattern projector can incident a pattern onto this circular component. Accordingly, a moiré pattern image of the component within the printed circuit board shown in FIG. 12b can be obtained. Due to the characteristics of the component within the printed circuit board, reflective areas may be generated in the moiré pattern image of the component within the printed circuit board. These reflective areas may correspond to bright areas. An image histogram can be generated for the moiré pattern image of the component within the printed circuit board. Bright and dark areas can be distinguished through the image histogram. Binarization can be performed through the image histogram.

[0052] FIGS. 13a and FIGS. 13b are drawings for explaining a method of increasing training data by dividing a moiré pattern image of a component in a printed circuit board into a grid, according to one embodiment of the present disclosure.

[0053] Referring to Fig. 13a, the moiré pattern image of a component on a printed circuit board can be divided into multiple grids. There may be a method to increase training data by switching the grids among themselves within the moiré pattern image of a component on a printed circuit board. Grids corresponding to dark areas and grids corresponding to bright areas can be distinguished. There may be a method to increase training data by switching the grids corresponding to dark areas among themselves. There may be a method to increase training data by switching the grids corresponding to bright areas among themselves.

[0054] Referring to FIG. 13b, numbers can be assigned to multiple grids in FIG. 13a. For example, a grid assigned a red number may correspond to a grid corresponding to a dark area. A grid assigned a blue number may correspond to a grid corresponding to a bright area. Training data can be increased by switching between grids assigned red numbers. Training data can be increased by switching between grids assigned blue numbers.

[0055] FIG. 14 is a drawing for explaining a method of increasing training data according to one embodiment of the present disclosure.

[0056] Referring to FIG. 14, a data-increasing device can acquire a moiré pattern image of a printed circuit board (S1410). The moiré pattern image can be acquired by an optical camera and a pattern projector. The data-increasing device can identify a reflection area and a shadow area within the moiré pattern image (S1420). The data-increasing device can increase data based on at least one of a reflection area, a shadow area, a predetermined group, and a grid (S1430). The step of increasing data may include the step of cropping the reflection area and the shadow area to a predetermined size and the step of increasing data by flipping a first area within the moiré pattern image left and right. The first area may correspond to an area within the moiré pattern image that is not a reflection area or a shadow area.

[0057] The step of increasing data may include a step of cropping the reflection area and the shadow area to a predetermined size, and a step of increasing data by inverting the first area within the moiré pattern image vertically. The first area may correspond to an area within the moiré pattern image that is not a reflection area or a shadow area. The step of increasing data may include a step of dividing the area within the moiré pattern image into a plurality of grids, a step of assigning numbers to the plurality of grids, and a step of increasing data by switching the plurality of grids based on the numbers.

[0058] The step of increasing data may include dividing an area within a moiré pattern image into a plurality of grids, assigning numbers to the plurality of grids, dividing the plurality of grids into one or more groups, and increasing data by switching the plurality of grids based on one or more groups and numbers. The data may be used as training data for a training model that generates a 3D model of a printed circuit board. The 3D model may represent the height of the printed circuit board.

[0059] Each component of the device or method according to the present invention may be implemented in hardware or software, or in a combination of hardware and software. Additionally, the function of each component may be implemented in software, and a microprocessor may be implemented to execute the function of the software corresponding to each component.

[0060] Various embodiments of the systems and techniques described herein may be realized as digital electronic circuits, integrated circuits, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include being implemented as one or more computer programs executable on a programmable system. A programmable system comprises a storage system, at least one input device, and at least one programmable processor (which may be a special-purpose processor or a general-purpose processor) coupled to receive data and instructions from and transmit data and instructions to at least one output device. Computer programs (which are also known as programs, software, software applications, or code) include instructions for the programmable processor and are stored on a "computer-readable recording medium."

[0061] Computer-readable recording media include all types of recording devices in which data that can be read by a computer system is stored. Such computer-readable recording media may be non-volatile or non-transitory media such as ROM, CD-ROM, magnetic tape, floppy disk, memory card, hard disk, magneto-optical disk, and storage device, and may also include transitory media such as data transmission media. Additionally, computer-readable recording media may be distributed across networked computer systems, and computer-readable code may be stored and executed in a distributed manner.

[0062] Although the flowcharts in this specification describe each process as being executed sequentially, this is merely an illustrative explanation of the technical concept of one embodiment of the present disclosure. In other words, a person skilled in the art to which one embodiment of the present disclosure belongs may modify and adapt the flowchart in various ways, such as by changing the order described in the flowchart or executing one or more of the processes in parallel, without departing from the essential characteristics of one embodiment of the present disclosure; therefore, the flowchart is not limited to a chronological order.

[0063] The above description is merely an illustrative explanation of the technical concept of the present embodiment, and a person skilled in the art to which the present embodiment belongs would be able to make various modifications and variations within the scope of the essential characteristics of the present embodiment. Accordingly, the present embodiments are intended to explain, not limit, the technical concept of the present embodiment, and the scope of the technical concept of the present embodiment is not limited by these embodiments. The scope of protection of the present embodiment shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of the present embodiment.

Claims

Claim 1 A method performed by a device for increasing data, comprising: acquiring a moiré pattern image of a printed circuit board; identifying a reflection area and a shadow area within the moiré pattern image; and increasing the data based on at least one of a predetermined group and a grid composed of the reflection area, the shadow area, and a plurality of grids, wherein the data is used as training data for a training model that generates a 3D model of the printed circuit board, and the 3D model represents the height of the printed circuit board. Claim 2 A method according to claim 1, wherein the moiré pattern image is obtained by an optical camera and a pattern projector. Claim 3 delete Claim 4 A method according to claim 1, wherein the step of increasing the data comprises: a step of cutting the reflection area and the shadow area into a predetermined size; and a step of increasing the data by flipping the first area within the moiré pattern image left and right, wherein the first area corresponds to an area within the moiré pattern image that is not the reflection area and the shadow area. Claim 5 A method according to claim 1, wherein the step of increasing the data comprises: cutting the reflection area and the shadow area into a predetermined size; and inverting the first area within the moiré pattern image vertically to increase the data, wherein the first area corresponds to an area within the moiré pattern image that is not the reflection area and the shadow area. Claim 6 A method according to claim 1, wherein the step of increasing the data comprises: dividing an area within the moiré pattern image into a plurality of grids; assigning numbers to the plurality of grids; and increasing the data by switching the plurality of grids based on the numbers. Claim 7 A method according to claim 1, wherein the step of increasing the data comprises: dividing an area within the moiré pattern image into a plurality of grids; assigning numbers to the plurality of grids; dividing the plurality of grids into one or more groups; and increasing the data by switching the plurality of grids based on the one or more groups and the numbers. Claim 8 A device for increasing data comprises: a memory; and at least one processor, wherein the at least one processor acquires a moiré pattern image of a printed circuit board, identifies a reflection area within the moiré pattern image and a shadow area within the moiré pattern image, and increases the data based on at least one of a predetermined group and a grid composed of the reflection area, the shadow area, and a plurality of grids, and the data is used as training data for a training model that generates a 3D model of the printed circuit board, and the 3D model represents the height of the printed circuit board. Claim 9 In claim 8, the device wherein the moiré pattern image is acquired by an optical camera and a pattern projector. Claim 10 delete Claim 11 In claim 8, the device wherein at least one processor cuts the reflection area and the shadow area to a predetermined size, flips a first area within the moiré pattern image left and right to increase the data, and the first area corresponds to an area within the moiré pattern image that is not the reflection area and the shadow area. Claim 12 In claim 8, the device wherein at least one processor cuts the reflection area and the shadow area to a predetermined size, inverts the first area within the moiré pattern image vertically to increase the data, and the first area corresponds to an area within the moiré pattern image that is not the reflection area and the shadow area. Claim 13 In claim 8, the device wherein at least one processor divides an area within the moiré pattern image into a plurality of grids, assigns numbers to the plurality of grids, and increases the data by switching the plurality of grids based on the numbers. Claim 14 In claim 8, the device wherein at least one processor divides an area within the moiré pattern image into a plurality of grids, assigns numbers to the plurality of grids, divides the plurality of grids into one or more groups, and increases the data by switching the plurality of grids based on the one or more groups and the numbers. Claim 15 A computer program that is combined with a computing device, which is hardware, to perform the steps of: acquiring a moiré pattern image of a printed circuit board; identifying a reflection area and a shadow area within the moiré pattern image; and storing the data in a non-transient recording medium readable by the computing device to perform the step of increasing data based on at least one of a predetermined group and a grid composed of the reflection area, the shadow area, and a plurality of grids, wherein the data is used as training data for a training model that generates a 3D model of the printed circuit board, and the 3D model represents the height of the printed circuit board. Claim 16 A computer-readable, non-transient recording medium in which a command is stored, wherein the command, when executed by the computer, causes the computer to perform the steps of: acquiring a moiré pattern image of a printed circuit board; identifying a reflection area and a shadow area within the moiré pattern image; and increasing data based on at least one of a predetermined group and a grid composed of the reflection area, the shadow area, and a plurality of grids, wherein the data is used as training data for a training model that generates a 3D model of the printed circuit board, and the 3D model represents the height of the printed circuit board.

Images