Apparatus and method for inspection, and recording medium recording command for inspection

Abstract

The present disclosure provides an apparatus. An apparatus according to the present disclosure includes: at least one first light source that irradiates illumination light to an object located on a reference surface; one or more cameras that capture one or more illumination images generated by reflection of the illumination light from the object body; and one or more processors which can determine one or more contour lines representing the edge of the object on the one or more illumination images, determine the height value of the edge of the object relative to the reference surface based on the one or more contour lines, and determine the height value of the edge of the object relative to the reference surface based on the determined height value. A first angle between the upper surface of the object and the reference surface is determined based on the height value.

Description

Technical Field

[0001] The present disclosure relates to a technique for inspection. Background Art

[0002] During the semiconductor manufacturing process, various inspections are performed to confirm whether various semiconductor processes and procedures are being properly performed. For example, an inspection can be performed to determine whether components such as dies disposed on a semiconductor substrate are positioned where they should be.

[0003] In particular, for a die mounted on a semiconductor substrate, an inspection can be performed to confirm whether there is a tilt between the mounted die and the substrate. Typically, the die can be mounted on top of the solder or solder ball when solder or solder balls are coated on the substrate. At this time, the bottom of the die should be mounted in parallel with the reference plane of the semiconductor substrate, but the die may also be mounted in a form tilted at a certain angle to the semiconductor substrate due to prescribed factors (for example, the coating state of the solder or solder ball). Since this is a factor that may cause a defect in the semiconductor device, during the inspection of the semiconductor, it should be possible to confirm whether the die is tilted, and if so, to confirm the tilt angle.

[0004] To inspect this tilt, a 3D inspection device can be used that illuminates the semiconductor die with three-dimensional illumination. However, this method makes it difficult to obtain the three-dimensional shape of the die being inspected. This is because the top surface of a semiconductor die is typically reflective, and the 3D illumination can cause total reflection on the top surface of the die, resulting in poor imaging of the die.

[0005] Alternatively, equipment using white light scanning interferometry (WLI) can be used to inspect the tilt. However, this method has the problem of not being able to form an image of a semiconductor crystal grain if it is tilted by a specific angle (e.g., 1.2 degrees) or more relative to a reference plane.

[0006] To detect tilt, two-dimensional area inspection can be used. When photographing a tilted quadrilateral semiconductor crystal, the degree of tilt appears as a parallelogram. Two-dimensional area inspection uses the shape of this parallelogram to derive the degree of tilt. However, if the semiconductor crystal's top surface is formed by a reflector, this method still fails to accurately capture the parallelogram shape, resulting in an inability to accurately predict the degree of tilt. Summary of the Invention

[0007] Technical problems to be solved

[0008] The present disclosure aims to solve the above-mentioned defects of the prior art and provide a technology for inspection.

[0009] Means of solving the problem

[0010] As one aspect of the present disclosure, a device for inspection can be provided. The device according to one aspect of the present disclosure includes: at least one first light source, which irradiates an object located on a reference plane with illumination light; one or more cameras, which capture one or more illumination images generated by the illumination light reflected from the object; and one or more processors, which can determine one or more contour lines representing the edges of the object on the one or more illumination images, determine the height values ​​of the edges on the object relative to the reference plane based on the one or more contour lines, and determine a first angle between the object and the reference plane based on the height values.

[0011] In one embodiment, the invention further includes: at least one second light source, which irradiates pattern light toward the object, and the one or more cameras capture one or more pattern images generated by the pattern light reflected from the object. The processor can determine the one or more contour lines representing the corners of the object based on the one or more illumination images and the one or more pattern images, determine a virtual plane corresponding to the top of the object based on the one or more contour lines, and determine a second angle between the virtual plane and the reference plane as the first angle.

[0012] In one embodiment, the one or more cameras may include: an upper camera, which captures an illumination image or a pattern image from above the object; and one or more side cameras, which capture an illumination image or a pattern image from one or more sides of the object.

[0013] In one embodiment, a portion of the one or more illumination images and the one or more pattern images may be upper images captured by the upper camera from above the object, and another portion of the one or more illumination images and the one or more pattern images may be side images captured by the one or more side cameras from one or more sides of the object.

[0014] In one embodiment, the processor may extract a first contour line on the upper image and a second contour line on the one or more side images representing the same corner of the object, and may determine the virtual plane based on the first contour line and the second contour line.

[0015] In one embodiment, the processor may determine first relationship information indicating a corresponding relationship between the first contour line and the second contour line, and may determine the virtual plane based on the first relationship information.

[0016] In one embodiment, the first point on the first contour line may correspond to the second point on the second contour line, and the first relationship information may include a pair of coordinate information, the coordinate information being the coordinate information of the first point on the upper image and the coordinate information of the second point on the one or more side images.

[0017] In one embodiment, the processor can extract a third contour line on a side image and a fourth contour line on another side image representing the same corner of the object body from the one or more side images, determine second relationship information indicating the correspondence between the third contour line and the fourth contour line, and can determine the virtual plane based on the first relationship information and the second relationship information.

[0018] In one embodiment, the processor may select an image having a contrast value of the surface of the object greater than a preset value from the one or more illumination images and the one or more pattern images to determine the virtual plane.

[0019] In one embodiment, the object body may have a form in which multiple layers are stacked, and the one or more contour lines may indicate corners of each of the multiple layers. The processor may determine multiple virtual planes corresponding to the multiple layers based on the one or more contour lines, and may determine the intervals between each of the multiple layers based on the multiple virtual planes corresponding to the multiple layers.

[0020] As one aspect of the present disclosure, a method for inspection can be provided. The method according to one aspect of the present disclosure can include: a step of irradiating an object located on a reference surface with illumination light from at least one first light source; a step of capturing one or more illumination images generated by the illumination light reflected from the object by one or more cameras; a step of determining one or more contour lines representing corners of the object on the one or more illumination images by one or more processors; a step of determining, based on the one or more contour lines, height values ​​of the corners on the object relative to the reference surface; and a step of determining, based on the height values, a first angle between the object surface and the reference surface.

[0021] In one embodiment, the method may further include: a step of at least one second light source irradiating patterned light onto the object; a step of the one or more cameras capturing one or more pattern images generated by the patterned light reflected from the object; a step of the processor determining the one or more contour lines representing the corners of the object based on the one or more illumination images and the one or more pattern images; a step of the processor determining a virtual plane corresponding to the top of the object based on the one or more contour lines; and a step of the processor determining a second angle between the virtual plane and the reference plane as the first angle.

[0022] In one embodiment, the one or more cameras may include: an upper camera, which captures an illumination image or a pattern image from above the object; and one or more side cameras, which capture an illumination image or a pattern image from one or more sides of the object.

[0023] In one embodiment, a portion of the one or more illumination images and the one or more pattern images may be upper images captured by the upper camera from above the object, and another portion of the one or more illumination images and the one or more pattern images may be side images captured by the one or more side cameras from one or more sides of the object.

[0024] In one embodiment, the step of determining the virtual plane may include: the step of extracting a first contour line on the upper image and a second contour line on the one or more side images representing the same corner of the object body; and the step of determining the virtual plane based on the first contour line and the second contour line.

[0025] In one embodiment, the step of determining the virtual plane may include: determining first relationship information indicating a corresponding relationship between the first contour line and the second contour line; and determining the virtual plane based on the first relationship information.

[0026] In one embodiment, the first point on the first contour line may correspond to the second point on the second contour line, and the first relationship information may include a pair of coordinate information, the pair of coordinate information being the coordinate information of the first point on the upper image and the coordinate information of the second point on the one or more side images.

[0027] In one embodiment, the step of determining the virtual plane may also include: the step of extracting a third contour line on a side image and a fourth contour line on another side image representing the same corner of the object body in the one or more side images; the step of determining second relationship information indicating the correspondence between the third contour line and the fourth contour line; and the step of determining the virtual plane based on the first relationship information and the second relationship information.

[0028] In one embodiment, the step of determining the virtual plane may include: selecting an image having a contrast value of the surface of the object greater than a preset value from the one or more illumination images and the one or more pattern images to determine the virtual plane.

[0029] According to one aspect of the present disclosure, a non-transitory computer-readable recording medium having recorded thereon commands for inspection can be provided. The commands recorded on the recording medium according to one aspect of the present disclosure, as commands for execution on a computer, when executed by one or more processors, can include executable commands that cause the one or more processors to perform the following steps: determining one or more contour lines representing corners of an object located on a reference plane on one or more illumination images generated by reflection of illumination light irradiated onto the object; determining heights of corners on the top surface of the object relative to the reference plane based on the one or more contour lines; and determining a first angle between the top surface of the object and the reference plane based on the heights.

[0030] Effects of the Invention

[0031] According to various embodiments of the present disclosure, when the upper surface of an object is a reflector, the tilt angle of the object can also be measured.

[0032] According to various embodiments of the present disclosure, even when the object is tilted at a specific angle (eg, 1.2 degrees) or more relative to the reference plane of the substrate, the degree of tilt can be derived.

[0033] According to various embodiments of the present disclosure, even when it is difficult to obtain a clear image of an object due to other obstructions on the substrate, the tilt angle of the object can be measured.

[0034] According to various embodiments of the present disclosure, an image can be obtained using light irradiated from various types of light sources at various angles and directions, thereby generating an accurate virtual three-dimensional model of an object. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a diagram showing a process of operating the device according to the present disclosure.

[0036] Figure 2 is a diagram illustrating a block diagram of a device according to an embodiment of the present disclosure.

[0037] Figure 3 FIG. 1 is a diagram illustrating a process of generating a virtual plane relative to an upper surface of an object according to an embodiment of the present disclosure.

[0038] Figure 4 FIG. 1 is a diagram illustrating image combination for obtaining diverse images and determining relationship information according to an embodiment of the present disclosure.

[0039] Figure 5 FIG. 1 is a diagram illustrating an arrangement of one or more pattern light sources according to an embodiment of the present disclosure.

[0040] Figure 6 This is a diagram showing an arrangement of one or more illumination light sources according to an embodiment of the present disclosure.

[0041] Figure 7 FIG. 4 is a diagram illustrating an image filtering process according to an embodiment of the present disclosure.

[0042] Figure 8 : is a diagram illustrating a process of deriving another contour line from one contour line of an image according to an embodiment of the present disclosure.

[0043] Figure 9 FIG2 is a diagram illustrating an embodiment of a method for inspection that can be performed by the apparatus according to the present disclosure. DETAILED DESCRIPTION

[0044] The various embodiments described herein are presented for the purpose of clearly illustrating the technical concept of the present disclosure, and are not intended to limit the present disclosure to specific implementation forms. The technical concept of the present disclosure includes various modifications, equivalents, alternatives, and embodiments that selectively combine all or part of the various embodiments described herein. In addition, the scope of the claims of the technical concept of the present disclosure is not limited to the various embodiments described below or the specific description of the embodiments.

[0045] The terms used herein, including technical or scientific terms, shall have the meanings generally understood by those skilled in the art to which the present disclosure belongs unless otherwise defined.

[0046] As used herein, expressions such as "include," "may include," "have," "may have," "have," and "may have" imply the presence of a target feature (e.g., function, action, or component) and do not preclude the presence of additional features. In other words, these expressions should be understood as open-ended terms that may include other embodiments.

[0047] As long as not mentioned differently, the singular expressions used in the present disclosure may include the plural meaning, and the same applies to the singular expressions recited in the claims.

[0048] Unless otherwise mentioned, the expressions "first", "second" or "first", "second" etc. used in this article are used to distinguish one object from other objects when describing multiple objects of the same type, and do not limit the order or importance between the objects.

[0049] As used herein, expressions such as "A, B, and C," "A, B or C," "A, B and / or C," or "at least one of A, B, and C," "at least one of A, B, or C," or "at least one of A, B, and / or C" may refer to each listed item or all possible combinations of the listed items. For example, "at least one of A or B" may refer to (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.

[0050] The expression "based on..." used in this article is used to describe one or more factors that influence the determination, judgment, behavior, or action described in the sentence or article containing the corresponding expression. The expression does not exclude additional factors that influence the corresponding determination, judgment, behavior, or action.

[0051] As used in this document, expressions such as “connected” or “linked” between a certain component (e.g., the first component) and other components (e.g., the second component) not only mean that the certain component is directly connected or linked to the other components, but also mean that it is connected or linked through a new other component (e.g., the third component).

[0052] The expression “configured to” as used herein may also have the meanings of “set to”, “having the ability to”, “changed to”, “prepared to”, “capable of”, etc., depending on the context.

[0053] In this disclosure, a target object may refer to an object located on a specified surface and serving as the inspection target of an apparatus. In one embodiment, the target object may be a die, semiconductor element, or chip located on a semiconductor substrate. The target object may be mounted on the semiconductor substrate. As described above, the target object may be mounted such that the lower surface of the target object is tilted at a certain angle relative to the reference surface of the substrate.

[0054] In the present disclosure, a substrate may be a predetermined surface on which an object is located. In one embodiment, the substrate may be a semiconductor substrate. The reference surface of the substrate may refer to the surface of the substrate, or may be a reference surface for measuring an inclination angle with respect to the bottom surface of the object.

[0055] The following describes various embodiments of the present disclosure in detail with reference to the accompanying drawings. In the drawings and descriptions of the drawings, identical or substantially equivalent components may be denoted by the same reference numerals. Furthermore, in the following descriptions of various embodiments, repeated descriptions of identical or corresponding components may be omitted, but this does not mean that such components are not included in the embodiments.

[0056] Figure 1 1 is a diagram illustrating the operation of an apparatus according to the present disclosure. The apparatus 1 according to the present disclosure can measure the tilt angle θ of an object 2 relative to a reference surface (e.g., a substrate, etc.). The object 2 (e.g., a crystal grain, etc.) can be placed on the reference surface. Depending on the situation, the object 2 can be placed 100 with a predetermined tilt angle θ relative to the reference surface. The apparatus 1 according to the present disclosure can inspect whether there is a problem with the placement of the object 2 by measuring (determining) the tilt angle.

[0057] Specifically, the illumination light source of apparatus 1 can irradiate two-dimensional illumination light toward object 2 located on a reference surface. The illumination light can be reflected from the surface of object 2, i.e., the top or one or more side surfaces. One or more cameras 50 of apparatus 1 can capture one or more illumination images generated by the illumination light reflected from object 2.

[0058] The device 1 can determine multiple contour lines representing the edges of the object 2 on one or more obtained illumination images. Each illumination image can be a two-dimensional photograph representing the object 2 or even the edges of the object 2. The edges of the object in the present disclosure can refer to the edges of the shape of the object existing in the real three-dimensional space (for example, the edges of a cube). On the one hand, the contour lines in the present disclosure can refer to the edges of the object represented in the image of the object (that is, the lines represented on the image). In the present disclosure, the contour lines can also be called dividing lines.

[0059] Based on the determined multiple contour lines, the apparatus 1 can determine the heights of the corners of the top surface of the object 2 relative to the reference plane. Furthermore, based on these heights, the apparatus 1 can determine the inclination angle θ between the top surface of the object 2 and the reference plane. The heights of the corners of the top surface of the object can represent the distance from each point of the corner to the reference plane. Therefore, by determining the heights of the corners of the top surface of the object, the angle of the corner relative to the reference plane can be determined, and thus the angle of inclination of the top surface of the object relative to the reference plane can be determined.

[0060] In one embodiment, in addition to illumination light, apparatus 1 may also utilize patterned light to determine the tilt angle of an object. Specifically, the patterned light source of apparatus 1 may illuminate object 2 with patterned light. The patterned light may be reflected from the surface of object 2, i.e., the top or one or more side surfaces. One or more cameras 50 of apparatus 1 may capture one or more pattern images generated by the patterned light reflected from object 2.

[0061] In this case, the apparatus 1 may determine a plurality of contour lines representing the corners of the object 2 based on one or more illumination images and / or one or more pattern images. The apparatus 1 may determine a virtual plane 3 (110) corresponding to the top surface of the object 2 based on the determined plurality of contour lines. The virtual plane 3 may be a plane that is virtually realized using computer graphics as the top surface of the object 2 in a real three-dimensional space. The apparatus 1 may determine the angle between the virtual plane 3 and the reference plane as the angle θ (110) between the object 2 and the reference plane.

[0062] In one embodiment, the object body 2 may be a stacked memory having a plurality of stacked layers. In this case, the extracted plurality of contour lines may indicate the corners of the plurality of layers of the object body 2. The device 1 may determine a plurality of virtual planes based on the plurality of contour lines obtained from the captured image of the object body 2 having a stacked structure. The process of determining the virtual planes is as described above. Each of the plurality of determined virtual planes may correspond to a plurality of layers of the object body 2 of the stacked structure, respectively. That is, a virtual plane may represent one of the plurality of layers. The device 1 may determine the gaps between the plurality of layers of the object body 2 of the stacked structure based on each of the plurality of determined virtual planes. In one embodiment, the process may be executed by the processor of the device 1.

[0063] According to various embodiments of the present disclosure, when it is difficult to obtain an image for inspection of the upper surface of the object 2 to be inspected because the upper surface is a reflective body, the device 1 can also use a virtual plane for the object 2 to measure the tilt angle of the object 2. In addition, according to various embodiments of the present disclosure, when the object is tilted at a specific angle (for example, 1.2 degrees) or more relative to the reference plane, the device 1 can also generate a virtual plane to derive the degree of tilt. In addition, according to various embodiments of the present disclosure, even when it is difficult to extract a clear image or contour line of the object 2 to be inspected due to various semiconductor elements and structures located on the substrate, necessary relevant information can be obtained to perform inspection of the object 2. In addition, according to various embodiments of the present disclosure, it is possible to obtain an image using light irradiated from various types of light sources at various angles and directions, thereby performing accurate measurement of the object.

[0064] Figure 2 is a diagram showing a block diagram of an apparatus according to an embodiment of the present disclosure. In one embodiment, the apparatus 1 may include: one or more processors 10, one or more memories 20, one or more pattern light sources 30, one or more illumination light sources 40 and / or one or more cameras 50. In one embodiment, at least one of these constituent elements of the apparatus 1 may be omitted, or other constituent elements may be added to the apparatus 1. In one embodiment, some constituent elements may be implemented additionally or alternatively, or in the form of singular or plural individuals. In the present disclosure, one or more constituent elements (e.g., one or more processors 10) may be expressed as constituent elements (e.g., processor 10), and may mean a collection of one or more constituent elements unless otherwise explicitly mentioned in the text. In one embodiment, at least some of the components inside / outside the device 1 can be interconnected via a bus, GPIO (general purpose input / output), SPI (serial peripheral interface) or MIPI (mobile industry processor interface), and can send and receive data and / or signals.

[0065] The illumination light source 40 can emit two-dimensional illumination light toward the object 2. In one embodiment, the illumination light can be RGB light, and each illumination light source 40 can include a red light source emitting red light, a green light source emitting green light, and a cyan light source emitting cyan light. In one embodiment, each illumination light source 40 can emit red light, green light, and cyan light to saturate the surface of the object.

[0066] The pattern light source 30 can irradiate pattern light onto the object 2. In one embodiment, the pattern light can be structured light, grid light (light whose luminosity changes in a sinusoidal wave pattern), linear light, etc. In one embodiment, the pattern light source 30 can irradiate pattern light that is phase-shifted by a grid transfer mechanism such as PZT (piezo actuator), or can irradiate pattern light by DLP (Digital Light Processing) or LCoS (Liquid Crystal on Silicon). In one embodiment, one or more pattern light sources 30 can irradiate pattern light according to a method selected from the PZT method, the DLP method, and the LCoS method. In one embodiment, one or more pattern light sources 30 (for example, 8) can be used to irradiate pattern light in different directions above the object 2.

[0067] As described above, the camera 50 can capture an illumination image and / or a pattern image. In one embodiment, the camera 50 can be a CCD (Charge Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor) camera. In one embodiment, the one or more cameras 50 may include a top camera and / or one or more side cameras. The top camera can capture an illumination image or pattern image reflected from the object 2 from above. The one or more side cameras can capture illumination images or pattern images reflected from the object 2 from the side or even the upper side of the object 2. In one embodiment, the one or more side cameras may include four cameras corresponding to the front, rear, left, and right sides of the object 2. In one embodiment, the illumination image or pattern image captured from above the object 2 by the top camera may be referred to as a top image. Furthermore, the illumination image or pattern image captured from one or more sides of the object 2 by the one or more side cameras may be referred to as a side image. That is, a portion of the one or more illumination images and the one or more pattern images may be an upper image, and another portion may be a side image.

[0068] The processor 10 can drive software (e.g., commands, programs, etc.) to control at least one component of the device 1 connected to the processor 10. In addition, the processor 10 can perform various operations related to the present disclosure, such as calculations, processing, data generation, and processing. In addition, the processor 10 can load data from the memory 20 or store it in the memory 20. In one embodiment, the processor 10 can determine multiple contour lines of the object 2 on one or more illumination images, determine the height values ​​of the corners on the object based on the multiple contour lines, and determine the angle θ between the object surface and the reference plane based on the height values. In addition, in one embodiment, the processor 10 can determine multiple contour lines representing the corners of the object 2 based on one or more illumination images and / or one or more pattern images, determine a virtual plane 3 corresponding to the object surface based on the multiple contour lines, and determine the angle between the virtual plane 3 and the reference plane as the inclination angle θ of the object 2.

[0069] The memory 20 can store a variety of data. The data stored in the memory 20 is data obtained, processed, or used by at least one component of the device 1, and may include software (e.g., commands, programs, etc.). The memory 20 may include volatile and / or non-volatile memory. In the present disclosure, commands and even programs, as software stored in the memory 20, may include an operating system for controlling the resources of the device 1, applications, and / or middleware that provides various functions to the applications so that the applications can utilize the resources of the device 1. In one embodiment, the memory 20 may store commands that, when executed by the processor 10, cause the processor 10 to perform operations.

[0070] In one embodiment, the device 1 may further include a transceiver (not shown) that can perform wireless or wired communication between the device 1 and a server or between the device 1 and other devices. For example, the transceiver can perform wireless communications such as eMBB (enhanced Mobile Broadband), URLLC (Ultra Reliable Low-Latency Communications), MMTC (Massive Machine Type Communications), LTE (Long-Term Evolution), LTE-A (LTE Advance), NR (New Radio), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (Code Division Multiple Access), WCDMA (Wideband CDMA), WiBro (Wireless Broadband), WiFi (Wireless Fidelity), Bluetooth, NFC (Near Field Communication), GPS (Global Positioning System) or GNSS (Global Navigation Satellite System). For example, the transceiver can implement wired communication via USB (Universal Serial Bus), HDMI (High Definition Multimedia Interface), RS-232 (Recommended Standard-232), or POTS (Plain Old Telephone Service). In one embodiment, the processor 10 can control the transceiver to obtain information from the server. The information obtained from the server can be stored in the memory 20. In one embodiment, the information obtained from the server can include extracting contour lines from an image, deriving relationship information between the extracted contour lines, or generating parameters or even information required for the virtual three-dimensional model 3.In one embodiment, the information obtained from the server may include intrinsic or extrinsic parameter values ​​of the camera 50 for calibrating the camera 50 .

[0071] In one embodiment, the device 1 may further include a user interface (not shown). The user interface may receive input from a user and output (display) information to the user. In one embodiment, the user interface may include an input device and / or an output device. The input device may be a device that receives data from the outside for transmission to at least one component of the device 1. For example, the input device may include a mouse, a keyboard, a touchpad, etc. The output device may be a device that provides various data of the device 1 to the user in a visual form. For example, the output device may include a display, a projector, or a hologram. In one embodiment, the user interface may receive parameters input by the user for inspecting the object body 2, or display the derived tilt angle θ of the object body 2 to the user.

[0072] In one embodiment, device 1 can be a variety of devices. For example, device 1 can be a portable communication device, a computer device, a portable multimedia device, or a device that is a combination of one or more of the aforementioned devices. However, device 1 of the present disclosure is not limited to the aforementioned devices.

[0073] The various embodiments of the apparatus 1 according to the present disclosure can be combined with one another. The various embodiments can be combined in any desired manner, and the resulting embodiments of the apparatus 1 still fall within the scope of the present disclosure. Furthermore, the aforementioned internal and external components of the apparatus 1 according to the present disclosure can be added, modified, replaced, or deleted depending on the embodiment. Furthermore, the aforementioned internal and external components of the apparatus 1 can be implemented as hardware components.

[0074] Figure 3 3 is a diagram illustrating a process of generating a virtual plane relative to an object according to an embodiment of the present disclosure. The process of generating the virtual plane 3 may include a calibration process, a contour line extraction process, a process of deriving relationship information between contour lines, and / or a virtual plane generation process.

[0075] During the calibration process (not shown), the device 1 can perform calibration of the camera 50. The device 1 can obtain the basic information necessary to generate a virtual plane relative to the object body through the calibration process. Specifically, during this process, the device 1 can obtain the position information of each camera 50, that is, the origin position information. Among them, the origin position of the camera 50 refers to the position of the point in three-dimensional space when the camera is regarded as a point. According to an embodiment, the position of the camera 50 can be expressed as a relative position relative to the object body 2, or as an absolute position in a three-dimensional space coordinate system. In addition, during the calibration process, the device 1 can obtain the viewing angle information and / or intrinsic parameters of each camera 50. The information obtained during the calibration process can be stored in the memory 20. In one embodiment, the calibration-related information can be obtained from the server via a transceiver. In one embodiment, the calibration process can be omitted as an optional process.

[0076] The following describes the process of extracting contour lines, deriving relationship information between contour lines, and generating a virtual plane. These processes can be performed using any two images selected from the illumination image and the pattern image, but for ease of explanation, the description is based on the top image (top) and a side image (left).

[0077] In the contour line extraction process s3010, the processor 10 may extract contour lines from the top image and the side image. Among the extracted contour lines, there may be contour lines representing the same corner of the object 2 in different images. For example, the processor 10 may extract a contour line representing a corner of the object 2 in the top image (hereinafter referred to as a first contour line), and extract a contour line representing the same corner in the side image (hereinafter referred to as a second contour line).

[0078] In the process s3020 of deriving the relationship information between the contour lines, the processor 10 can determine (derive) the relationship information between the first contour line and the second contour line (hereinafter referred to as the first relationship information). Among them, the corner of the object body 2 jointly represented by the two contour lines can be the corner shared by two adjacent surfaces of the object body 2. Specifically, a point on the first contour line (hereinafter referred to as the first point) and a point on the second contour line (hereinafter referred to as the second point) jointly represent a point on the real corner of the object body (hereinafter referred to as the object point), and therefore can correspond to each other. When the object point is observed from the upper camera, the object point can be represented as the first point on the upper image. Assuming that there is a straight line in three-dimensional space connecting the object point from the origin of the upper camera, if the straight line is observed from the side camera, the straight line will be represented as a two-dimensional straight line on the side image. The two-dimensional straight line is called the epipolar line relative to the first point (epipolar line). On the side image, the intersection between the epipolar line relative to the first point and the second contour line can be derived, and the intersection can be the second point (i.e., the point corresponding to the first point). The transformation relationship from the first point to the epipolar line relative to the first point can be defined by a fundamental matrix. The fundamental matrix is ​​based on the concept of epipolar geometry and can be derived from the aforementioned camera calibration information.

[0079] In one embodiment, the first relationship information may include information about two corresponding points (a first point and a second point). In one embodiment, the first relationship information may be a pair of two-dimensional pixel coordinates of the two points (the first point and the second point), or a pair of three-dimensional spatial coordinates, or a transformation relationship or transformation matrix between the two coordinates.

[0080] In the virtual plane generation process s3030, the processor 10 may use the determined first relationship information to determine (generate) a virtual plane 3 corresponding to the top surface of the object. Specifically, the processor 10 may use the first relationship information to obtain a vector from the origin of the top camera to the coordinates of the first point, and a vector from the origin of the side camera to the coordinates of the second point, thereby deriving the point where the two vectors intersect. These points may be used to perform 3D modeling of the entire top surface of the object.

[0081] Although the generation process of the virtual plane 3 is described with reference to the top image (top) and a side image (left), as described above, the virtual plane 3 can also be generated using two different side images from a plurality of side images. In this case, the processor 10 can extract the contour line representing a corner of the object 2 in one of the side images (hereinafter referred to as the third contour line), and extract the contour line representing the same corner in the other image in the side image (hereinafter referred to as the fourth contour line). The processor 10 can also obtain relationship information indicating the correspondence between the third contour line and the fourth contour line (hereinafter referred to as the second relationship information). The processor 10 can determine the virtual plane 3 based on the first relationship information and / or the second relationship information.

[0082] In one embodiment, the processor 10 may repeatedly perform the process s3020 of obtaining relationship information from an image multiple times. The process of obtaining relationship information may be repeated multiple times based on combinations of top images, side images, etc. In one embodiment, the processor 10 may use different images for obtaining relationship information based on the type of illumination image or pattern image (top, front, back, left, right, etc.) and perform the process of obtaining relationship information multiple times for the same surface. In one embodiment, the processor 10 may perform the process of obtaining relationship information multiple times for the same surface based on the illumination angle and direction of the illumination light source or pattern light source.

[0083] Figure 4 This figure illustrates image combination for obtaining multiple images and determining relationship information according to one embodiment of the present disclosure. Apparatus 1 can obtain multiple images from an object. The images that can be obtained vary depending on the type of light used, the direction of light illumination, and the angle of light illumination.

[0084] In one embodiment, the apparatus 1 can obtain multiple illumination images or pattern images for one direction of the object depending on whether the light used for capture is illumination light or pattern light. That is, for the same surface of the object, multiple images can also be obtained depending on the type of light used.

[0085] In one embodiment, the device 1 can obtain multiple images for one direction of the object according to the direction and even the angle of light irradiated by the pattern light source 30 or the illumination light source 40 to the object. For example, one or more pattern light sources 30 can be arranged in various positions (for example, 8 directions (8-way)) to irradiate pattern light to the object. The configuration of the pattern light source 30 will be described later. Depending on the direction of the irradiated pattern light, multiple pattern images can be obtained for the same surface of the object. In addition, for example, one or more illumination light sources 40 can irradiate illumination light to the object at various angles. The configuration of the illumination light source 40 will be described later. Depending on the angle of the irradiated illumination light, multiple illumination images can be obtained for the same surface of the object.

[0086] In one embodiment, as described above, the device 1 can select two images in different directions (an upper image, side images in various directions, etc.) from the illumination images or pattern images obtained in various directions of the object, thereby deriving the relationship information between the contour lines from the two images. First, the processor 10 can select an upper image and a side image, thereby deriving the first relationship information. The upper image can represent the upper surface of the object, and a side image can represent the upper surface and one or more side surfaces of the object. Since the upper surface and side surfaces of the object share corners, the first relationship information can be derived from the upper image and the side image. For example, when the object is a rectangular parallelepiped, four pieces of first relationship information can be derived respectively.

[0087] Next, the processor 10 can select two side images in different directions from one or more side images to derive second relationship information. The two side images can represent two adjacent sides of the object. Because the two adjacent sides share the corners of the object, the second relationship information can be derived from the two side images. For example, if the object is a rectangular parallelepiped, four pieces of second relationship information can be derived.

[0088] Third, the processor 10 can select two side images representing two sides of an object body that do not share common corners from more than one side image, thereby deriving relationship information (hereinafter referred to as the third relationship information). In this case, the third relationship information can also be derived using the polar line. Specifically, when the object body is a rectangular parallelepiped, two parallel side faces can share two parallel corners. A point on a contour line on a side image can be represented as a straight line (polar line) on another parallel side image. The point where the polar line intersects the contour line on the other side image can correspond to a point on the first side image. The third relationship information indicating this corresponding relationship can be derived. For example, when the object body is a rectangular parallelepiped, two pieces of third relationship information can be derived separately.

[0089] The processor 10 can use at least one of the first relationship information, the second relationship information, and / or the third relationship information to derive a virtual plane 3 relative to the object. A variety of relationship information can be obtained based on the selectable combination of two images, illumination directions, angles, and lighting types from which the relationship information is derived. The processor 10 can use this variety of relationship information to more accurately generate a virtual plane.

[0090] Figure 5 1 is a diagram illustrating the configuration of one or more pattern light sources 30 according to an embodiment of the present disclosure. As described above, the apparatus 1 may include one or more pattern light sources 30. The pattern light sources 30 may be arranged at various locations around the object 2 to irradiate the object 2 with patterned light.

[0091] In the illustrated embodiment 5010, the pattern light source 30 can be a light source that is located on the normal 4 above the object 2 and irradiates pattern light onto the object 2. Specifically, a normal 4 perpendicular to the top of the object 2 can be drawn. A pattern light source 30 can be located at a point on the normal 4. The pattern light source 30 can irradiate pattern light onto the top of the object 2. In one embodiment, the normal 4 can be drawn as a half-straight line extending from the center of the top of the object 2 in the opposite direction of the reference plane. According to the embodiment, the distance between the pattern light source 30 and the object 2 can be changed. In one embodiment, the pattern light source 30 can change the phase of the pattern and irradiate the pattern light onto the object 2 multiple times in sequence.

[0092] In the illustrated embodiment 5020, the pattern light sources 30 may be multiple pattern light sources 30 arranged on concentric circles 5 above the object. Specifically, a concentric circle 5 parallel to the reference plane and centered at a point on the normal 4 on the object can be assumed. Multiple pattern light sources 30 may be arranged around the circumference of the concentric circle 5. These multiple pattern light sources 30 may irradiate patterned light onto the object 2. Depending on the embodiment, the radius of the concentric circle 5 may be varied. Depending on the embodiment, the spacing between the multiple pattern light sources 30 arranged on the concentric circle 5 may be varied. In one embodiment, the multiple pattern light sources 30 may be arranged at equal spacing around the circumference of the concentric circle 5, or at arbitrarily set, mutually varying spacings. In one embodiment, a total of eight pattern light sources 30 may be arranged at 45-degree intervals. In this case, the arrangement of the multiple pattern light sources 30 may be referred to as an 8-way arrangement. In one embodiment, the multiple pattern light sources 30 may sequentially irradiate patterned light onto the object 2. In one embodiment, the patterns of the patterned lights emitted by the plurality of patterned light sources 30 may have different phases.

[0093] Figure 6 1 is a diagram illustrating an arrangement of one or more illumination light sources according to an embodiment of the present disclosure. As described above, the apparatus 1 may include one or more illumination light sources 40. The illumination light sources 40 may be arranged at various positions around an object to illuminate the object.

[0094] In the illustrated embodiment, the illumination light sources 40 may be multiple illumination light sources 40 arranged on a concentric circle 6 perpendicular to a reference plane. Specifically, it can be assumed that the concentric circle 6 is perpendicular to the reference plane and has the object 2 as its center. Multiple illumination light sources 40 may be arranged on the circumference of the concentric circle 6. These multiple illumination light sources 40 may irradiate illumination light toward the object.

[0095] Depending on the embodiment, the radius of the concentric circle 6 can be changed. Depending on the embodiment, the spacing between the multiple illumination light sources 40 arranged in the concentric circle 6 can be changed. In one embodiment, the multiple illumination light sources 40 can be arranged at equal spacing around the circumference of the concentric circle 6, or at arbitrarily set, mutually different spacings around the circumference. In one embodiment, the multiple illumination light sources 40 can be arranged at any angle. The angle here refers to the angle between the position of the illumination light source 40 on the circumference and the normal to the reference plane. In one embodiment, at least one of the multiple illumination light sources 40 can be arranged at a high angle of 70 degrees or greater (a low angle relative to the reference plane). In one embodiment, the multiple illumination light sources 40 can be arranged at angles of 17 degrees, 40 degrees, 58 degrees, and 70 degrees, respectively. In one embodiment, the multiple illumination light sources 40 can be arranged only above the reference plane. In one embodiment, the multiple illumination light sources 40 can sequentially illuminate the object 2. In one embodiment, the illumination light emitted by the multiple illumination light sources 40 can have different wavelengths. In one embodiment, the multiple illumination light sources 40 may include a red light source, a green light source, and a cyan light source capable of emitting red, green, and cyan light, respectively. In one embodiment, the multiple illumination light sources 40 may first irradiate a target with light of a single wavelength (e.g., red light) sequentially at a configured angle. Then, the multiple illumination light sources 40 may irradiate a target with light of a different wavelength (e.g., green light) sequentially at a configured angle, and then irradiate the target with light of the remaining wavelength (e.g., cyan light) sequentially at a configured angle. In one embodiment, the multiple illumination light sources 40 may sequentially irradiate red, green, and cyan light at one angle, and then sequentially irradiate red, green, and cyan light at the next angle. In one embodiment, the irradiation order of red, green, and cyan light may be changed according to the designer's intent. In one embodiment, when irradiating two-dimensional illumination light, each illumination light source 40 may simultaneously irradiate red, green, and cyan light, thereby irradiating a mixed white light onto the target. In one embodiment, each illumination light source 40 may irradiate a target with normal white light instead of mixed white light. In one embodiment, the plurality of illumination light sources 40 arranged at various angles can simultaneously irradiate mixed white light to saturate the surface of the object. In this case, the reflected illumination light can be captured to obtain an illumination image of the surface of the object.

[0096] Figure 7is a diagram showing the filtering process of an image according to an embodiment of the present disclosure. In one embodiment, the processor 10 may filter the images obtained according to various factors (type of light, illumination direction, angle, etc.) according to preset standards, and then generate the virtual plane 3 using only the filtered images. When the device 1 inspects the object body (for example, grains) on the substrate, it may be affected by other components and bumps of the substrate. Due to these influences, the obtained images themselves may be distorted, or the contour lines obtained from these images may be distorted. When the virtual plane 3 is generated based on the distorted image or even the contour line, it may be difficult to accurately measure or even inspect the object body 2. To prevent this, only the filtered images can be filtered out according to the preset standards and used for the generation of the virtual plane 3.

[0097] In one embodiment, the processor 10 may perform filtering based on whether the contrast value of the surface of the object 2 appearing in the image is greater than a preset value. The processor 10 may discard images with contrast values ​​below the preset value and select only images with contrast values ​​greater than the preset value to determine (derive) the aforementioned relationship information.

[0098] In one embodiment, the processor 10 may perform filtering based on the continuity of contour lines appearing in the image. The processor 10 may determine the aforementioned relationship information by selecting only images in which the contour line representing a corner of the object is represented by a continuous line segment. Furthermore, in one embodiment, the processor 10 may statistically calculate the average or median value of the position coordinates of contour lines extracted from multiple images and exclude from these values ​​images in which contour lines with position coordinates deviating by more than a preset ratio are extracted. In this case, the processor 10 may determine the aforementioned relationship information only from images that have not been excluded.

[0099] Figure 8 is a diagram illustrating the process of deriving one contour line from another contour line in an image according to one embodiment of the present disclosure. In one embodiment, processor 10 can utilize pre-stored specification information of object 2 to determine another contour line from one contour line. As described above, when device 1 inspects a substrate object, it may be affected by other components, unevenness, and other factors on the substrate. Consequently, it may be impossible to extract the contour lines necessary to determine relationship information. In this case, to supplement the contour line extraction process, pre-stored specification information of object 2 can be utilized to derive another contour line from one contour line.

[0100] Specifically, the memory 20 may include specification information of the object 2. The specification information may include length, width, and height information of the object 2. After the processor 10 derives a contour line from the image, the processor 10 may use the specification information to predict and derive the position of a contour line located parallel to the contour line. For example, it may be assumed that the lower contour line 8010 of the object 2 is extracted, and the upper contour line located parallel to the lower contour line cannot be extracted due to a prescribed distortion. In this case, the processor 10 may obtain height information of the object 2 based on the specification information, apply the height information to the extracted lower contour line, and thereby predict the position of the upper contour line located parallel to the lower contour line. In this manner, the upper contour line 8020 can be extracted.

[0101] If the contour lines necessary for determining the relationship information cannot be derived due to distortion of the image or even the contour lines, the processor 10 may first extract contour lines parallel to the necessary contour lines and then extract the necessary contour lines using the specification information as described above. The processor 10 may determine the relationship information 8030 using the contour lines necessary for determining the relationship information.

[0102] In one embodiment, when the object 2 is not a cuboid, the specification information may also include information about the object's shape. For example, if the object is cylindrical, the specification information may store information about the radius and height of the cylinder. The processor 10 may use the radius and height information to derive other contours necessary to determine the relationship information from one contour.

[0103] In one embodiment, the memory 20 can function as a database storing specification information of an object to be inspected. The memory 20 can store specification information regarding objects located on a substrate or commonly used (e.g., components, chips, etc.). In one embodiment, the processor 10 identifies the object from the acquired image and can then read the specification information corresponding to the identified object from the memory 20. Based on the obtained specification information, the processor 10 can derive other contours necessary to determine relationship information from one contour of the identified object. In one embodiment, the processor 10 can control the transceiver to obtain the specification information of the object from a server. The obtained specification information can be stored in the memory 20.

[0104] Figure 9is a diagram showing an embodiment of a method for inspection that can be performed by the device 1 according to the present disclosure. The method for inspection according to the present disclosure can be a method implemented by a computer. In the flowchart shown in the figure, the steps of the method or algorithm according to the present disclosure are described in sequence, but in addition to being performed in sequence, the steps can also be performed in an order that can be arbitrarily combined by the present disclosure. The description according to this flowchart does not exclude changes or modifications to the method or algorithm, and does not mean that any step is necessary or preferred. In one embodiment, at least some of the steps can be performed in parallel, repeatedly or heuristically. In one embodiment, at least some of the steps can be omitted, or other steps can be added.

[0105] The apparatus 1 according to the present disclosure can execute the inspection method according to various embodiments of the present disclosure. The method according to one embodiment of the present disclosure may include: step S910 of irradiating an object located on a reference surface with illumination light; step S920 of capturing one or more illumination images generated by reflection of the illumination light from the object; step S930 of determining multiple contour lines representing corners of the object on the one or more illumination images; step S940 of determining height values ​​of the corners on the object relative to the reference surface based on the multiple contour lines; and / or step S950 of determining a first angle between the object surface and the reference surface based on the height values.

[0106] In step S910, at least one first light source (e.g., illumination light source 40) of apparatus 1 may irradiate illumination light toward object 2 located on a reference surface. In step S920, one or more cameras 50 of apparatus 1 may capture one or more illumination images generated by the illumination light reflected from object 2.

[0107] In step S930, one or more processors 10 of apparatus 1 may determine, in one or more illumination images, a plurality of contour lines representing corners of object 2. In step S940, processor 10 may determine, based on the plurality of contour lines, heights of the corners on the top surface of the object relative to a reference plane. In step S950, processor 10 may determine, based on the heights, a first angle (e.g., angle θ) between the top surface of the object and the reference plane.

[0108] In one embodiment, the method according to the present disclosure may further include: a step of at least one second light source (e.g., pattern light source 30) of device 1 irradiating pattern light onto object 2; a step of one or more cameras 50 capturing one or more pattern images generated by reflection of the pattern light from object 2; a step of the processor 10 determining a plurality of contour lines representing the corners of object 2 based on the one or more illumination images and the one or more pattern images; a step of the processor 10 determining a virtual plane 3 corresponding to the top of object 2 based on the plurality of contour lines; and / or a step of the processor 10 determining a second angle between the virtual plane 3 and the reference plane as a first angle.

[0109] In one embodiment, the step of determining the virtual plane may include: the step of extracting a first contour line on the upper image and a second contour line on the one or more side images representing the same corner of the object body 2; and / or the step of determining the virtual plane 3 based on the first contour line and the second contour line.

[0110] In one embodiment, the step of determining the virtual plane may further include: determining first relationship information indicating a corresponding relationship between the first contour line and the second contour line; and / or determining the virtual plane 3 according to the first relationship information.

[0111] In one embodiment, the step of determining the virtual plane may also include: the step of extracting a third contour line on a side image and a fourth contour line on another side image representing the same corner of the object body in more than one side image; the step of determining second relationship information indicating the correspondence between the third contour line and the fourth contour line; and / or the step of determining the virtual plane 3 based on the first relationship information and the second relationship information.

[0112] In one embodiment, the step of determining the virtual plane may include: selecting an image having a contrast value of the surface of the object greater than a preset value from the one or more illumination images and the one or more pattern images to determine the virtual plane 3 .

[0113] Various embodiments of the present disclosure can be implemented as software recorded on a machine-readable recording medium. The software can be software used to implement the various embodiments of the present disclosure. The software can be inferred by a programmer skilled in the art from the various embodiments of the present disclosure. For example, the software can be machine-readable commands (e.g., code or code segments) or programs. A machine is a device that can operate according to commands invoked from a recording medium, such as a computer. In one embodiment, the machine can be an electronic device 1 according to an embodiment of the present disclosure. In one embodiment, the processor of the machine can execute the invoked commands, thereby causing the components of the machine to perform functions corresponding to the commands. In one embodiment, the processor can be one or more processors 10 according to an embodiment of the present disclosure. The recording medium can refer to any type of recording medium that is machine-readable and can store data. For example, the recording medium can include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. In one embodiment, the recording medium can be one or more memories 20. In one embodiment, the recording medium can be implemented in a distributed form on a computer system connected by a network. The software can be stored and executed in a distributed manner on a computer system, etc. The recording medium may be a non-transitory recording medium. A non-transitory recording medium means a tangible medium that does not store data semi-permanently or temporarily and does not include a signal that is transmitted transiently.

[0114] While the technical concepts of the present disclosure have been described above based on various embodiments, the technical concepts of the present disclosure encompass various substitutions, modifications, and alterations that are within the scope of understanding of a person skilled in the art. Furthermore, it should be understood that such substitutions, modifications, and alterations are encompassed within the scope of the appended claims.

Claims

1. An inspection device, in, include: at least one light source, the light source irradiating illumination light toward an object located on the reference surface; a plurality of cameras, the plurality of cameras capturing a plurality of illumination images generated by the illumination light reflecting from the object; and More than one processor, the one or more processors, determining a plurality of contour lines representing corners of the object from two or more illumination images captured in different directions among the plurality of illumination images, Based on the multiple contour lines, a virtual plane corresponding to the upper surface of the object body is determined, An angle between the virtual plane and the reference plane is determined.

2. The inspection device according to claim 1, wherein: The multiple cameras include: an upper camera that captures an illuminated image from above the object; and One or more side cameras are configured to capture illumination images from one or more sides of the object.

3. The inspection device according to claim 2, wherein: A portion of the plurality of illumination images is an upper image captured by the upper camera from above the object body, Another part of the plurality of illumination images is side images captured by the one or more side cameras from one or more sides of the object.

4. The inspection device according to claim 3, wherein: the one or more processors, extracting a first contour line on the upper image and a second contour line on the one or more side images that represent the same corner of the object, The virtual plane is determined based on the first contour line and the second contour line.

5. The inspection device according to claim 4, wherein: the one or more processors, determining first relationship information indicating a corresponding relationship between the first contour line and the second contour line, The virtual plane is determined based on the first relationship information.

6. The inspection device according to claim 5, wherein: A first point on the first contour line corresponds to a second point on the second contour line, The first relationship information includes a pair of coordinate information, wherein the pair of coordinate information is coordinate information of the first point on the upper image and coordinate information of the second point on the one or more side images.

7. The inspection device according to claim 6, wherein: the one or more processors, extracting a third contour line on one side image and a fourth contour line on another side image that represent the same corner of the object from the one or more side images, determining second relationship information indicating a corresponding relationship between the third contour line and the fourth contour line, The virtual plane is determined based on the first relationship information and the second relationship information.

8. The inspection device according to claim 1, wherein: An image contrast value of the surface of the object in the two or more illumination images captured from the different directions is greater than a preset value.

9. The inspection device according to claim 1, wherein: The at least one light source includes a plurality of light sources arranged on the circumference of concentric circles perpendicular to the reference plane.

10. An inspection device, in, include: at least one light source, the light source irradiating pattern light toward an object located on a reference surface; a plurality of cameras, the plurality of cameras capturing a plurality of pattern images generated by the pattern light being reflected from the object; and More than one processor, the one or more processors, determining a plurality of contour lines representing corners of the object from two or more pattern images captured in different directions among the plurality of pattern images, Based on the multiple contour lines, a virtual plane corresponding to the upper surface of the object body is determined, An angle between the virtual plane and the reference plane is determined.

11. The inspection device according to claim 10, wherein: The at least one light source includes a light source located on a normal line above the object volume.

12. The inspection device according to claim 10, wherein: The at least one light source includes a plurality of light sources centered at a point on the normal line of the upper surface of the object and located on the circumference of concentric circles parallel to the reference plane.

13. The inspection device according to claim 10, wherein: The multiple cameras include: an upper camera, the upper camera capturing a pattern image from above the object; and One or more side cameras are used to capture pattern images from one or more sides of the object.

14. The inspection device according to claim 13, wherein: A portion of the plurality of pattern images is an upper image captured by the upper camera from above the object. Another part of the plurality of pattern images is side images captured by the one or more side cameras from one or more sides of the object.

15. The inspection device according to claim 14, wherein: the one or more processors, extracting a first contour line on the upper image and a second contour line on the one or more side images that represent the same corner of the object, The virtual plane is determined based on the first contour line and the second contour line.

16. The inspection device according to claim 15, wherein: the one or more processors, determining first relationship information indicating a corresponding relationship between the first contour line and the second contour line, The virtual plane is determined based on the first relationship information.

17. The inspection device according to claim 16, wherein: A first point on the first contour line corresponds to a second point on the second contour line, The first relationship information includes a pair of coordinate information, wherein the pair of coordinate information is coordinate information of the first point on the upper image and coordinate information of the second point on the one or more side images.

18. The inspection device according to claim 17, wherein: the one or more processors, extracting a third contour line on one side image and a fourth contour line on another side image representing the same corner of the object from among the one or more side images, determining second relationship information indicating a corresponding relationship between the third contour line and the fourth contour line; and The virtual plane is determined based on the first relationship information and the second relationship information.

19. The inspection device according to claim 10, wherein: An image contrast value of the surface of the object in the two or more pattern images captured from the different directions is greater than a preset value.

Images