Method and apparatus for inspecting defects in tempered glass

Abstract

The present disclosure relates to a method and apparatus for inspecting defects in tempered glass. Provided may be the apparatus for inspecting defects in tempered glass, according to an embodiment of the present disclosure, which: acquires a first image by capturing the tempered glass; acquires a second image by preprocessing the first image; generates a third image in which classification has been carried out according to types of defects by inputting the second image into a deep learning model; and outputs a result of analyzing a state of the tempered glass on the basis of the third image.

Description

Method and device for inspecting defects in tempered glass

[0001] The present disclosure relates to a method and apparatus for inspecting defects in tempered glass.

[0002] A photovoltaic power generation system is a technology that converts solar energy into electrical energy using solar modules, and tempered glass, one of the core components of a solar module, plays an important role in protecting solar cells and maintaining the module's performance against the external environment.

[0003] Meanwhile, a technology for inspecting defects in tempered glass as described above is automated optical inspection (AOI) conducted by tempered glass manufacturers. Automated optical inspection is a defect inspection method based on visible light images of tempered glass, and while it is possible to detect foreign matter defects, there was a problem in that it was difficult to detect defects that allow light to pass through due to the material properties of tempered glass.

[0004] In addition, when tempered glass that has undergone the above-mentioned automated optical inspection is received by solar panel manufacturers, there was a problem in that the defect detection rate of the tempered glass was not high because the manufacturers detected defects through manual sampling inspection.

[0005] Therefore, there is a need for a more efficient method and device for inspecting defects in tempered glass.

[0006] The aforementioned background technology is technical information that the inventor possessed for the derivation of the present invention or acquired during the process of deriving the present invention, and it cannot be considered as prior art disclosed to the general public prior to the filing of the present invention.

[0007] The technical problem that the present disclosure aims to solve is to provide a method and apparatus for inspecting defects in tempered glass. The problem that the present disclosure aims to solve is not limited to the problem mentioned above, and other problems and advantages of the present disclosure not mentioned can be understood from the following description and will be more clearly understood from the embodiments of the present disclosure. Furthermore, it will be seen that the problems and advantages that the present disclosure aims to solve can be realized by the means and combinations thereof set forth in the claims.

[0008] As a technical means for achieving the technical problem described above, the first aspect of the present disclosure may provide a method for inspecting defects in tempered glass, comprising: a step of obtaining a first image of the tempered glass; a step of obtaining a second image by preprocessing the first image; a step of generating a third image by classifying the type of defect by inputting the second image into a deep learning model; and a step of outputting a result in which the state of the tempered glass is analyzed based on the third image.

[0009] A second aspect of the present disclosure provides an apparatus for inspecting defects in tempered glass, wherein the apparatus acquires a first image of the tempered glass, preprocesses the first image to acquire a second image, generates a third image classified according to the type of defect by inputting the second image into a deep learning model, and outputs a result in which the condition of the tempered glass is analyzed based on the third image.

[0010] A third aspect of the present disclosure may provide a computer-readable recording medium having a program for executing the method of the first aspect of the present disclosure on a computer.

[0011] According to the means for solving the problem of the present disclosure described above, it is possible to provide the effect of improving the defect detection rate of tempered glass and reducing the risk of breakage of tempered glass products.

[0012] Embodiments of the present disclosure will be described with reference to the accompanying drawings described below, wherein similar reference numerals indicate similar elements, but are not limited thereto.

[0013] FIG. 1 is a drawing for explaining a device for inspecting defects in tempered glass according to one embodiment.

[0014] FIG. 2 is a block diagram of a device for inspecting defects in tempered glass according to one embodiment.

[0015] FIG. 3 is an exemplary drawing for explaining the process of acquiring a first image according to one embodiment.

[0016] FIG. 4 is an exemplary drawing for explaining the process of acquiring a second image according to one embodiment.

[0017] FIG. 5 is an exemplary drawing for explaining the process of generating a third image according to one embodiment.

[0018] FIG. 6 is an exemplary drawing for explaining the process of outputting analyzed results to a display according to one embodiment.

[0019] FIG. 7 is a flowchart illustrating a method for inspecting defects in tempered glass according to one embodiment.

[0020] FIG. 8 is a flowchart illustrating the process of acquiring a second image according to one embodiment.

[0021] FIG. 9 is a flowchart illustrating the process of generating a third image according to one embodiment.

[0022] FIG. 10 is a flowchart illustrating the overall process of a method for inspecting defects in tempered glass according to one embodiment.

[0023] An apparatus according to one embodiment of the present disclosure is an apparatus for inspecting defects in tempered glass, wherein the apparatus acquires a first image of the tempered glass, preprocesses the first image to acquire a second image, generates a third image classified according to the type of defect by inputting the second image into a deep learning model, and outputs a result in which the condition of the tempered glass is analyzed based on the third image.

[0024] The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments presented below, but can be implemented in various different forms and should be understood to include all modifications, equivalents, and substitutions that fall within the spirit and scope of the present disclosure. The embodiments presented below are provided to make the present disclosure complete and to fully inform those skilled in the art of the scope of the invention. In describing the present disclosure, detailed descriptions of related prior art are omitted where it is determined that such detailed descriptions may obscure the essence of the present invention.

[0025] The terms used herein are used merely to describe specific embodiments and are not intended to limit the disclosure. Unless otherwise defined, all terms used herein have the same meaning as generally understood by those skilled in the art to which this disclosure pertains.

[0026] In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. Furthermore, terms such as "comprising" or "having" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0027] Additionally, terms including ordinal numbers, such as "first" or "second" as used herein, may be used to describe various components, but the components should not be limited by the terms. The terms are used solely for the purpose of distinguishing one component from another.

[0028] Phrases such as "in one embodiment," "according to one embodiment," "related to one embodiment," or "according to an implementation of one embodiment" in this specification do not necessarily refer to the same embodiment. Furthermore, throughout this specification, "examples" are arbitrary distinctions to facilitate the description of the present disclosure, and each embodiment does not need to be mutually exclusive. For example, configurations mentioned for the description of one embodiment may be applied and / or implemented in other embodiments, and may be modified and applied and / or implemented to the extent that they do not depart from the scope of the present disclosure.

[0029] Some embodiments of the present disclosure may be represented by functional block configurations and various processing steps. Some or all of the functional blocks may be implemented by various numbers of hardware and / or software configurations that perform specific functions. For example, the functional blocks of the present disclosure may be implemented by one or more microprocessors or by circuit configurations for a specific function.

[0030] For example, the functional blocks of the present disclosure may be implemented in various programming or scripting languages. The functional blocks may be implemented as algorithms executed on one or more processors. Additionally, the present disclosure may employ prior art for electronic configuration, signal processing, and / or data processing, etc. Terms such as "mechanism," "element," "means," and "configuration" may be used broadly and are not limited to mechanical and physical configurations. Furthermore, terms such as "-part," "-model," etc. refer to a unit that processes at least one function or operation, which may be implemented in hardware or software, or as a combination of hardware and software.

[0031] Furthermore, the connecting lines or connecting members between the components depicted in the drawings are merely illustrative of functional connections and / or physical or circuit connections. In the actual device, connections between components may be represented by various alternative or added functional connections, physical connections, or circuit connections.

[0032] In addition, some components in the drawings may be depicted with their size or proportions slightly exaggerated. Also, components depicted in one drawing may not be depicted in another drawing.

[0033] In addition, the term "AOI (automated optical inspection) camera" may refer to a camera used for automated optical inspection.

[0034] The present disclosure will be described in detail below with reference to the attached drawings.

[0035] FIG. 1 is a drawing for explaining a device for inspecting defects in tempered glass according to one embodiment.

[0036] Referring to FIG. 1, a device (100) for inspecting defects in tempered glass, tempered glass (110), and defects (120) within the tempered glass are shown.

[0037] In the present disclosure, the tempered glass (110) may be tempered glass used in a solar module, but is not limited thereto.

[0038] According to one embodiment, a device (100) for inspecting defects in tempered glass can detect defects in tempered glass by inputting tempered glass (110) and outputting defects (120) in the tempered glass.

[0039] Specifically, a device (100) for inspecting defects in tempered glass can obtain a first image by photographing the tempered glass using a bottom light, at least one polarizing film, and an AOI camera.

[0040] Here, the first image represents a polarized image acquired through an AOI camera as light emitted from the lower illumination passes through a polarizing film and reinforced glass.

[0041] According to one embodiment, a device (100) for inspecting defects in tempered glass can acquire a first image through an AOI camera as light emitted from a lower light passes through a lower polarizing film, tempered glass (110), and an upper polarizing film in sequence.

[0042] Specifically, light emitted from an underlight can be made to vibrate in only a specific direction as it passes through a polarizing film. In defect-free areas of the tempered glass, the emitted light can pass through in a consistent direction. However, in defect-containing areas of the tempered glass, the direction of the emitted light may be distorted or partially scattered, appearing in a different pattern.

[0043] In other words, the polarizing film can control the polarization state of light and perform the role of detecting changes in polarization of defects, and the AOI camera can perform the role of acquiring polarized images at high resolution.

[0044] Meanwhile, the types of defects within the tempered glass (110) may include at least one of surface defects, internal defects, or optical defects. Surface defects may include scratches, contaminants, and blemishes, internal defects may include bubbles, foreign substances, and cracks, and optical defects may include non-uniformity of light transmission, reflection imbalance, and refraction errors.

[0045] If defects in the tempered glass (110) are detected only through a conventional AOI camera, optical defects cannot be detected. The present invention uses a polarizing film together with an AOI camera to detect optical defects, thereby enabling more accurate detection of defects (120) within the tempered glass.

[0046] Specifically, the device (100) for inspecting defects in tempered glass can acquire a first image of the tempered glass (110). The first image may include a polarized image, which is an image acquired through an AOI camera as light from a bottom light passes through a polarizing film.

[0047] Meanwhile, in the case of tempered glass (110) used in the solar module production process, a stress of -50 MPa to -100 MPa may exist in the surface layer. At this time, if there is a defect inside the tempered glass (110), the stress within that location may change abnormally. The polarized image included in the first image contains information regarding the internal stress change of the tempered glass (110) as described above, which can improve the defect detection rate.

[0048] Additionally, the tempered glass (110) may be weakened against external thermal shock and physical shock due to the influence of changes in stress around defects (120) within the tempered glass during the production of the solar module.

[0049] Accordingly, the device (100) for inspecting defects in tempered glass can prevent breakage of the tempered glass (110) by determining defects (120) within the tempered glass according to the type and size of the defects based on changes in stress around the defects as described above.

[0050] FIG. 2 is a block diagram of a device for inspecting defects in tempered glass according to one embodiment.

[0051] Referring to FIG. 2, a device (200) for inspecting defects in tempered glass may include a processor (210), a communication module (220), and a memory (230). Additionally, the processor (210) may include a deep learning model (211).

[0052] For convenience of explanation, only the components related to the present invention are shown in FIG. 2. Accordingly, in addition to the components shown in FIG. 2, other general-purpose components may be further included in the device (200) for inspecting defects in tempered glass.

[0053] According to one embodiment, the processor (210) can obtain a second image by preprocessing a first image containing a polarized image using optical equipment and a polarizing film.

[0054] Here, the second image represents an image obtained by preprocessing the first image and dividing the first image into multiple regions.

[0055] Additionally, the processor (210) can generate a third image classified according to the type of defect by inputting the second image into the deep learning model (211).

[0056] Here, the third image represents the second image classified according to the type of defect in the tempered glass through a deep learning model (211).

[0057] Additionally, the processor (210) can output the result of analyzing the state of the tempered glass based on the third image to the display.

[0058] Meanwhile, the communication module (220) may be configured to receive a first image through a communication network. The communication module (220) may provide the received first image to the processor (210). Additionally, the communication module (220) may be configured to transmit an image processed by the processor (210) through a communication network.

[0059] Meanwhile, the processor (210) can obtain a second image by preprocessing a first image received through a communication module (220), and can generate a third image by performing defect classification on the second image through a deep learning model. In one embodiment, the processor (210) may include at least one of a CPU (central processing unit), a GPU (graphic processing unit), a DSP (digital signal processor), an FPGA (Field Programmable Gate Array), and an ASIC (Application Specific Integrated Circuit) to perform any computational operation.

[0060] Meanwhile, memory (230) should be broadly interpreted to include any electronic component capable of storing electronic information. Memory (230) may refer to various types of processor-readable media such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erase-programmable read-only memory (EPROM), electrically eraseable PROM (EEPROM), flash memory, magnetic or optical data storage devices, registers, etc.

[0061] If the processor (210) can read information from and / or write information to the memory (230), the memory (230) is said to be in an electronic communication state with the processor (210). The memory integrated in the processor (210) is in an electronic communication state with the processor (210).

[0062] FIG. 3 is an exemplary drawing for explaining the process of acquiring a first image according to one embodiment.

[0063] Referring to FIG. 3, a lower light (310), a lower polarizing film (320), a reinforced glass (330), a defect (340), an upper polarizing film (350), and an AOI camera (360) are shown.

[0064] According to one embodiment, the processor (210) can check the internal stress of the tempered glass (330) as light emitted from the lower illumination (310) passes through the lower polarizing film (320), the tempered glass (330), and the upper polarizing film (350) in sequence. Additionally, the processor (210) can acquire a first image through an AOI camera (360). The first image may include a polarized image showing a change in the internal stress of the tempered glass (330).

[0065] Specifically, the processor (210) can accurately detect stress concentration areas and microcracks that are not visible with a normal light source by using polarizing films (320 and 350). Additionally, surface and internal defects can be identified without damaging the tempered glass (330). Furthermore, because it utilizes the polarization properties of light, the processor (210) can detect defects such as stress non-uniformity within the tempered glass (330) with high precision.

[0066] FIG. 4 is an exemplary drawing for explaining the process of acquiring a second image according to one embodiment.

[0067] Referring to FIG. 4, a second image (400) and one of the divided regions, region A1 (401), are shown.

[0068] According to one embodiment, the processor (210) can first remove unnecessary regions from the first image to obtain the second image (400).

[0069] Here, the unnecessary area refers to an area that is not used to detect defects (340) in the tempered glass (330).

[0070] For example, removing unnecessary areas may involve removing unnecessary noise from the first image. Specifically, the acquired first image may include a background area surrounding the tempered glass (330) in addition to the tempered glass. Accordingly, the processor (210) may remove the background area surrounding the tempered glass (330) using either a rule-based or deep learning-based technique.

[0071] For example, edge detection or corner detection techniques may be used in rule-based techniques, but are not limited thereto. Additionally, image segmentation techniques may be used in deep learning-based techniques, but are not limited thereto.

[0072] Next, according to one embodiment, the processor (210) can normalize the first image from which unnecessary regions have been removed. Specifically, the processor (210) can adjust the pixel value range of the first image from which unnecessary regions have been removed to a specific range so that a defect detection algorithm or a deep learning model can process consistent data.

[0073] Next, according to one embodiment, the processor (210) can adjust the brightness and contrast of the first uniformized image. Specifically, the processor (210) can enhance the contrast of the first uniformized image to make the defect (340) more distinct.

[0074] For example, the processor (210) can adjust the dark parts of the uniformized first image to be bright and the bright parts to be dark to provide uniform contrast overall.

[0075] Next, according to one embodiment, the processor (210) can obtain a second image (400) by dividing the adjusted first image into a plurality of regions. Specifically, when generating the third image, the processor (210) can perform an efficient defect classification task by first performing classification only on region A1 (401), which is one of the divided regions.

[0076] FIG. 5 is an exemplary drawing for explaining the process of generating a third image according to one embodiment.

[0077] Referring to FIG. 5, a second image (510), a deep learning model (520), and a third image (530) are shown.

[0078] According to one embodiment, by inputting a second image (510) into a deep learning model (520), a third image (530) that has undergone defect classification can be output.

[0079] According to one embodiment, the deep learning model (520) can first identify at least one defect (340) included in the second image (510). Specifically, the deep learning model (520) can perform data labeling on the type of defect (340) included in the second image (510). For example, the deep learning model (520) can identify at least one defect (340) included in the second image (510) by performing data labeling according to the type of defect (340), such as scratches, bubbles, cracks, foreign substances, etc.

[0080] Next, the deep learning model (520) can acquire a third image (530) based on the type of the identified defect (340). For example, the deep learning model (520) can perform classification by the type of the identified defect and apply a bounding box to the defect (340) area to visualize the location of the defect (340) within the tempered glass (330).

[0081] According to one embodiment, the deep learning model (520) may use a pre-trained Convolutional Neural Network (CNN), but is not limited thereto, and may also be based on a neural network of a different structure.

[0082] For example, the deep learning model (520) may include a deep neural network (DNN) containing multiple layers and may classify the type and location of one or more defects included in a received image using a pre-trained convolutional neural network. Here, the pre-trained convolutional neural network may be composed of one or more layers that perform convolution operations on input values ​​and may infer output values ​​by performing convolution operations on input values.

[0083] Information regarding the defects (340) classified in this way can be processed by the processor (210), and information regarding the defects (340) classified in this way or information processed therefrom can be transmitted to the display of the third image (530) through the communication module (220).

[0084] FIG. 6 is an exemplary drawing for explaining the process of outputting analyzed results to a display according to one embodiment.

[0085] Referring to FIG. 6, a display (610) showing the analyzed results and a defect image (620) are shown.

[0086] According to one embodiment, the third image (530) may include information regarding defects in the tempered glass (330). Specifically, information regarding at least one of the location, size, or type of the defect may be included in the third image (530). The processor (210) may finally determine whether the tempered glass (330) is defective based on the size of the defect included in the third image (530) and output it to the display (610) as a defect image (620).

[0087] According to one embodiment, the processor (210) may output an analyzed result including a defect image (620) to a display (610). The output display (610) may include at least one of the location of the defect, the type of defect, the size of the defect, or statistical data regarding the defect included in the tempered glass (330).

[0088] For example, the display (610) may output information visualizing the location, type, and size of defects within the tempered glass (330) based on the third image (530), and may display the quantity and location data of defects detected in the tempered glass (330) in the form of a table.

[0089] Additionally, the display (610) may output cumulative statistical data of defects based on the production quantity of tempered glass (330). The user can utilize the cumulative statistical data of defects for process alarms and in-house inspections.

[0090] FIG. 7 is a flowchart illustrating a method for inspecting defects in tempered glass according to one embodiment.

[0091] Referring to FIG. 7, in step 710, the processor (210) can obtain a first image of the tempered glass (330).

[0092] According to one embodiment, the processor (210) can check the internal stress of the tempered glass (330) as light emitted from the lower illumination (310) passes through the lower polarizing film (320), the tempered glass (330), and the upper polarizing film (350) in sequence. Additionally, the processor (210) can acquire a first image through an AOI camera (360). The first image may include a polarized image showing a change in the internal stress of the tempered glass (330).

[0093] In step 720, the processor (210) can preprocess the first image to obtain the second image (510).

[0094] In step 730, the processor (210) can generate a third image (530) that has been classified according to the type of defect by inputting the second image (510) into the deep learning model (211).

[0095] Specifically, the processor (210) can identify at least one defect included in the second image (510) through a deep learning model (211) and generate a third image (530) based on the type of the identified defect.

[0096] In step 740, the processor (210) can output the result of analyzing the state of the reinforced glass (330) based on the third image (530).

[0097] Specifically, the processor (210) can output at least one of the location of the defect included in the tempered glass, the type of defect, the size of the defect, or statistical data about the defect to the display.

[0098] FIG. 8 is a flowchart illustrating the process of acquiring a second image according to one embodiment.

[0099] Referring to FIG. 8, in step 810, the processor (210) can remove unnecessary regions from the first image.

[0100] Here, the unnecessary area refers to an area that is not used to detect defects (340) in the tempered glass (330).

[0101] In step 820, the processor (210) can homogenize the first image from which unnecessary areas have been removed.

[0102] Specifically, the processor (210) can adjust the pixel value range of the first image from which unnecessary areas have been removed to a specific range so that the defect detection algorithm or deep learning model (211) can process consistent data.

[0103] In step 830, the processor (210) can adjust the brightness and contrast ratio of the first uniformized image.

[0104] Specifically, the processor (210) can improve the contrast of the uniformized first image to make the defect (340) more distinct.

[0105] In step 840, the processor (210) can obtain a second image (510) by dividing the adjusted first image into multiple regions.

[0106] Specifically, the processor (210) can perform an efficient defect classification task by first performing classification only on area A1 (401), which is one of the divided areas, when generating the third image.

[0107] FIG. 9 is a flowchart illustrating the process of generating a third image according to one embodiment.

[0108] Referring to FIG. 9, in step 910, the processor (210) can identify at least one defect included in the second image (510) through a deep learning model (211).

[0109] Specifically, the deep learning model (211) can perform data labeling on the types of defects (340) included in the second image (510). For example, the deep learning model (211) can identify at least one defect (340) included in the second image (510) by performing data labeling on the types of defects (340), such as scratches, bubbles, cracks, foreign substances, etc.

[0110] In step 920, the processor (210) can generate a third image (530) based on the type of identified defect.

[0111] For example, the deep learning model (211) can perform classification by type of identified defect and apply a bounding box to the defect (340) area to visualize the location of the defect (340) within the tempered glass (330).

[0112] FIG. 10 is a flowchart illustrating the overall process of a method for inspecting defects in tempered glass according to one embodiment.

[0113] Referring to Fig. 10, first in step 1010, tempered glass can be fed into the AOI equipment.

[0114] In step 1020, a polarized image can be generated through optical equipment and a polarizing film.

[0115] In step 1030, the processor (210) can acquire a first image through an AOI camera.

[0116] In step 1040, the processor (210) can analyze the acquired first image and store it in memory (230).

[0117] In step 1050, the processor (210) can preprocess the first image to obtain the second image (510).

[0118] Specifically, in step 1051, the processor (210) can remove unnecessary regions from the first image.

[0119] In step 1052, the processor (210) can normalize the first image from which unnecessary regions have been removed.

[0120] In step 1053, the processor (210) can adjust the brightness and contrast of the first uniformized image.

[0121] In step 1054, the processor (210) can obtain a second image (510) by dividing the adjusted first image into multiple regions. For a more specific process of preprocessing the first image, refer to FIG. 5.

[0122] In step 1060, the processor (210) can generate a third image (530) by performing classification based on a previously trained deep learning model.

[0123] Specifically, in step 1061, the processor (210) can perform classification based on a deep learning model (211).

[0124] In step 1062, the processor (210) can determine the size of the defect based on the defect-specific spec.

[0125] In step 1063, the processor (210) can store information data about the defect in memory (230). Refer to FIG. 6 for the process of generating a more specific third image (530).

[0126] In step 1070, the processor (210) can make a final defect determination including at least one of the location and size of the defect in the tempered glass. Refer to FIGS. 6 and 7 for a specific final defect determination process.

[0127] In step 1080, the processor (210) may output the determined result to a display. The output display (610) may include at least one of the location of the defect contained in the tempered glass, the type of the defect, the size of the defect, or statistical data regarding the defect. For a more specific display output process, refer to FIG. 7.

[0128] The device described above may be implemented as computer-readable code on a computer-readable recording medium. A computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices. Additionally, computer-readable recording media may be distributed across networked computer systems, allowing computer-readable code to be stored and executed in a distributed manner. Furthermore, functional programs, codes, and code segments for implementing the above embodiments can be easily inferred by programmers skilled in the art to which the present invention pertains.

[0129] The techniques described herein may be implemented by various means. For example, these techniques may be implemented in hardware, firmware, software, or a combination thereof. Those skilled in the art will further understand that the various exemplary logical blocks, models, circuits, and algorithmic steps described in connection with the disclosure herein may be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability of hardware and software, various exemplary components, blocks, models, circuits, and steps have been generally described above in terms of their functionality. Whether such functionality is implemented as hardware or as software depends on the design constraints imposed on the specific application and the overall system. Those skilled in the art may implement the described functionality in various ways for each specific application, but such implementation decisions should not be construed as departing from the scope of this disclosure.

[0130] In a hardware implementation, the processing units used to perform the techniques may be implemented in one or more ASICs, DSPs, digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions described herein, computers, or a combination thereof.

[0131] Accordingly, the various exemplary logic blocks, models, and circuits described in connection with the disclosure herein may be implemented or performed by any combination of general-purpose processors, DSPs, ASICs, FPGAs or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or those designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but alternatively, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, for example, a DSP and a microprocessor, multiple microprocessors, one or more microprocessors coupled with a DSP core, or any other combination of such configurations.

[0132] In firmware and / or software implementations, techniques may be implemented as instructions stored on a computer-readable medium such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable PROM (EEEPROM), flash memory, compact disc (CD), magnetic or optical data storage devices, etc. The instructions may be executable by one or more processors, and the processor(s) may be enabled to perform specific aspects of the functions described herein.

[0133] When implemented in software, the functions may be stored on a computer-readable medium as one or more instructions or code, or transmitted through a computer-readable medium. Computer-readable media include both computer storage media and communication media, including any medium that facilitates the transmission of a computer program from one place to another. Storage media may be any available media accessible by a computer. As a non-limiting example, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium accessible by a computer that can be used to transfer or store desired program code in the form of instructions or data structures. Additionally, any connection is appropriately referred to as a computer-readable medium.

[0134] For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair cable, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, coaxial cable, fiber optic cable, twisted pair cable, digital subscriber line, or wireless technologies such as infrared, radio, and microwave are included within the definition of a medium. As used herein, disk and disc include CD, laser disc, optical disc, DVD (digital versatile disc), floppy disk, and Blu-ray disc, wherein disks usually play data magnetically, whereas discs play data optically using a laser. The above combinations should also be included within the scope of computer-readable media.

[0135] The software model may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other known form of storage medium. An exemplary storage medium may be coupled to a processor so that the processor can read information from the storage medium or write information to the storage medium. Alternatively, the storage medium may be integrated into the processor. The processor and the storage medium may exist within an ASIC. The ASIC may exist within a user terminal. Alternatively, the processor and the storage medium may exist as separate components within the user terminal.

[0136] The foregoing description of the present disclosure is provided to enable those skilled in the art to practice or use the present disclosure. Various modifications of the present disclosure will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to various variations without departing from the spirit or scope of the present disclosure. Accordingly, the present disclosure is not intended to be limited to the examples described herein, but is intended to be given the broadest possible scope consistent with the principles and novel features disclosed herein.

[0137] Although the subject matter has been described in language specific to structural features and / or methodological operations, it will be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or operations described above. Rather, the specific features and operations described above are described as exemplary forms for implementing the claims.

[0138] Although the method described in this specification has been explained through specific embodiments, it is possible to implement it as computer-readable code on a computer-readable recording medium. A computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. Additionally, computer-readable recording media may be distributed across networked computer systems, so that computer-readable code can be stored and executed in a distributed manner. Furthermore, functional programs, codes, and code segments for implementing the embodiments can be easily inferred by programmers skilled in the art to which the present invention pertains.

[0139] Although the present disclosure has been described in relation to some embodiments, various modifications and changes may be made without departing from the scope of the present disclosure as understood by a person skilled in the art to which the present invention pertains. Furthermore, such modifications and changes should be considered to fall within the scope of the claims appended to this specification.

Claims

1. In a method for inspecting defects in tempered glass, A step of obtaining a first image of the above-mentioned tempered glass; A step of obtaining a second image by preprocessing the first image; A step of generating a third image classified according to the type of defect by inputting the second image into a deep learning model; and A method comprising the step of outputting a result in which the state of the tempered glass is analyzed based on the third image above.

2. In Paragraph 1, The step of acquiring the first image above is, A method comprising the step of acquiring the first image using a bottom light, at least one polarizing film, and an automated optical inspection (AOI) camera.

3. In Paragraph 2, The step of acquiring the first image above is, A step of verifying the internal stress of the tempered glass as light emitted from the lower illumination passes through the lower polarizing film, the tempered glass, and the upper polarizing film in sequence; and The method includes the step of acquiring the first image through the AOI camera; and The above first image represents a change in the internal stress, a method.

4. In Paragraph 1, The above first image includes a polarized image, a method.

5. In Paragraph 1, The step of acquiring the second image above is, A step of removing unnecessary regions from the first image; A step of normalizing the first image from which the above-mentioned unnecessary region has been removed; A step of adjusting the brightness and contrast of the first uniformized image; and A method comprising the step of dividing the above-mentioned adjusted first image into a plurality of regions.

6. In Paragraph 1, The step of generating the third image above is, A step of identifying at least one defect included in the second image through the deep learning model; and A step of generating the third image based on the type of defect identified above; A method including 7. In Paragraph 1, The above outputting step is, A method comprising the step of determining whether the tempered glass is defective based on the size of the defect included in the third image.

8. In Paragraph 1, The above method is, A step of outputting the above-mentioned analyzed results to a display; A method that further includes.

9. In Paragraph 8, The step of outputting to the above display is, A step of outputting at least one of the location of the defect included in the tempered glass, the type of the defect, the size of the defect, or statistical data regarding the defect to the display; A method that further includes.

10. In a device for inspecting defects in tempered glass, The above device is, An apparatus that acquires a first image of the tempered glass, preprocesses the first image to acquire a second image, generates a third image classified according to the type of defect by inputting the second image into a deep learning model, and outputs a result in which the state of the tempered glass is analyzed based on the third image.

11. In Paragraph 10, Acquiring the above first image is, A device for acquiring the first image using a bottom light, at least one polarizing film, and an AOI camera.

12. In Paragraph 10, Acquiring the above first image is, Light emitted from the lower illumination passes through the lower polarizing film, the tempered glass, and the upper polarizing film in sequence to check the internal stress of the tempered glass, and the first image is acquired through the AOI camera, and The above first image is a device showing a change in the internal stress.

13. In Paragraph 10, The above first image is a device including a polarized image.

14. In Paragraph 10, Acquiring the above second image is, An apparatus for removing unnecessary regions from the first image, homogenizing the first image from which the unnecessary regions have been removed, adjusting the brightness and contrast ratio of the homogenized first image, and dividing the adjusted first image into a plurality of regions.

15. In Paragraph 10, Generating the above third image is, An apparatus that identifies at least one defect included in the second image through the deep learning model and generates the third image based on the type of the identified defect.

16. In Paragraph 10, The output above is, A device for determining whether the tempered glass is defective based on the size of the defect included in the third image above.

17. In Paragraph 10, The above device is, A device that outputs the above-mentioned analyzed results to a display.

18. In Paragraph 17, What is output to the above display is, A device for outputting at least one of the location of the defect included in the tempered glass, the type of the defect, the size of the defect, or statistical data regarding the defect to the display.

19. A computer-readable recording medium storing a program for executing the method according to claim 1 on a computer.

Images