KR102446658B1 - Part defect inspection apparatus and method using image recognition based on artificial intelligence - Google Patents

Abstract

Disclosed is a part defect inspection apparatus using image recognition based on artificial intelligence, comprising: a camera; a plurality of light source modules which irradiate a light source to a target inspection part at different irradiation angles to form an imaging light; a transfer module for moving positions of the plurality of light source modules; a memory in which image characteristic information of the target inspection part and position characteristic information of the plurality of light source modules are stored in association for each imaging light; and a processor which, when an image containing the target inspection part is received from the camera, extracts the image characteristic information of the target inspection part in the image based on artificial intelligence, searches for the position characteristic information of the plurality of light source modules associated with the image characteristic information extracted from the memory, and controls the transfer module to change the positions of the plurality of light source modules based on the searched position characteristic information. Therefore, the present invention can accurately check defect of a target inspection part.

Description

Apparatus and method for component defect inspection using image recognition based on artificial intelligence {PART DEFECT INSPECTION APPARATUS AND METHOD USING IMAGE RECOGNITION BASED ON ARTIFICIAL INTELLIGENCE}

The present disclosure relates to a component defect inspection apparatus. More specifically, the present disclosure relates to an apparatus and method for inspecting component defects using artificial intelligence-based image recognition.

In general, parts to be mounted on various electronic products, such as mobile phones, digital cameras, and telephones, various patterns or symbols are formed through lasers, etc., and the material of the parts may also be made of various materials such as plastics, metals, and ceramics.

Even if a required pattern or engraving is formed on the surface of the part, a pattern or engraving that is not appropriate to the standard required for the part may be formed. In addition, defects such as scratches, foreign substances, bubbles, irregularities, tears, warping or folding may occur on the surface during the production and transportation process.

As such, when a defect in the engraving or pattern of the part occurs, the reliability in the operation of the part or the performance of the part may be greatly reduced.

Therefore, there is a need for a process of inspecting whether a defect in a pattern or an engraving formed on a part occurs.

However, the visual inspection by the operator has a problem in that workability is poor, defects can be overlooked, and the inspection time for checking defects is increased.

Korean Patent Publication No. 10-2298218 (2021.09.06. Announcement)

An object of the embodiments disclosed in the present disclosure is to provide a thing capable of preventing a defect of a component to be inspected from being overlooked.

In addition, an embodiment disclosed in the present disclosure has an object of providing that a defect of a component to be inspected can be accurately identified.

In addition, an embodiment disclosed in the present disclosure has an object of providing a method capable of shortening an inspection time for confirming a defect of a component to be inspected.

The problems to be solved by the present disclosure are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

A component defect inspection apparatus using artificial intelligence-based image recognition according to an aspect of the present disclosure for achieving the above-described technical problem includes: a camera; a plurality of light source modules for irradiating light sources to parts to be inspected at different irradiation angles to form photographic lighting; a transport module for moving the plurality of light source modules; a memory in which image characteristic information of a corresponding inspection target part for each photographing light and location characteristic information of a plurality of light source modules are linked and stored; and when an image including the part to be inspected is received from the camera, image characteristic information of the part to be inspected is extracted from the image based on artificial intelligence, and linked with the extracted image characteristic information in the memory a processor that searches for positional characteristic information of the plurality of light source modules and controls the transfer module to change positions of the plurality of light source modules based on the found positional characteristic information; may include.

In addition, the memory, the light color information and the light intensity information of the plurality of light source modules are further linked and stored; The processor further searches for light color information and light intensity information of a plurality of light source modules associated with the extracted image characteristic information in the memory, and based on the found light color information and light intensity information, the plurality of It may be characterized in that the LED element is turned on so that the light source is irradiated with the light color and light intensity of at least one of a Red LED element, a Green LED element, a Blue LED element, and a White LED element in the light source module.

In addition, it further comprises a plate for mounting the part to be inspected; It is disposed along the periphery of the part to be inspected, and forms a movement path of the transfer module, further comprising a support provided on the plate to be curved toward the central axis of the plate; The transfer module may be characterized in that it changes the height and irradiation angle of the plurality of light source modules along the longitudinal direction of the support.

In addition, it further comprises a plate for mounting the part to be inspected; It is disposed along the periphery of the part to be inspected, forms a movement path of the transfer module, further comprising a support provided on the plate parallel to the central axis of the plate; The transfer module may include: a first transfer module for changing a first height and a first irradiation angle of a plurality of first light source modules among the plurality of light source modules; and a second transfer module for changing a second height and a second irradiation angle of a plurality of second light source modules among the plurality of light source modules; including, wherein the plurality of first light source modules are arranged at equal intervals from each other, the plurality of second light source modules are arranged at equal intervals from each other, and the distance between the plurality of first light source modules is equal to the distance between the plurality of second light sources. It may be characterized in that it is wider than the interval between modules.

In addition, the first transfer module, while being supported on the support, a first frame that moves in the longitudinal direction of the support; a first connecting member connecting the first frame and the plurality of first light source modules; and a first angle adjusting member connected to the plurality of first light source modules through side surfaces of the first connecting member and configured to adjust irradiation angles of the plurality of first light source modules. may include.

In addition, a first diffusion plate mounted on the plurality of first light source modules to diffuse the light sources of the plurality of first light source modules may be further included.

In addition, the second transfer module, while being supported on the support, a second frame moving in the longitudinal direction of the support; a second connecting member connecting the second frame and the plurality of second light source modules; and a second angle adjusting member connected to the plurality of second light source modules through side surfaces of the second connecting member and adjusting irradiation angles of the plurality of second light source modules. may include.

In addition, a second diffusion plate mounted on the plurality of second light source modules to diffuse the light sources of the plurality of second light source modules may be further included.

In addition, it further comprises a third light source module disposed under the camera; It is mounted on the lower part of the third light source module and further includes a third diffusion plate for diffusing the light source of the third light source module, wherein the third light source module comprises white LED elements arranged at equal intervals from each other can do.

In addition, the component defect inspection method using artificial intelligence-based image recognition according to another aspect of the present disclosure is a component defect inspection method using artificial intelligence-based image recognition performed by a device, the processor of the device through, determining whether an image including the part to be inspected is received from the camera of the device; when an image including the part to be inspected is received through the processor of the apparatus, extracting image characteristic information of the part to be inspected from within the image based on artificial intelligence; searching for positional characteristic information of a plurality of light source modules associated with the extracted image characteristic information in a memory of the apparatus through a processor of the apparatus; controlling, through the processor of the apparatus, the transfer module of the apparatus to change the positions of the plurality of light source modules based on the searched position characteristic information; forming photographic lighting by irradiating light sources to the parts to be inspected at different irradiation angles through the plurality of light source modules; and photographing the part to be inspected through the camera of the device; may include.

According to the above-mentioned problem solving means of this indication, the effect which can prevent overlooking the defect of an inspection object component is provided.

In addition, according to the above-described problem solving means of the present disclosure, it provides the effect of accurately confirming the defect of the component to be inspected.

In addition, according to the above-described problem solving means of the present disclosure, it is possible to reduce the inspection time for confirming the defect of the inspection target component is provided.

Effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a perspective view illustrating the configuration of a component defect inspection apparatus according to the present disclosure as an example.
FIG. 2 is a view illustrating a process in which the transfer module of FIG. 1 moves as an example.
3 and 4 are views illustrating an arrangement structure between a plurality of light source modules of FIG. 1 as an example.
5 is a view illustrating a plurality of light source modules and a diffusion plate mounted on the camera of FIG. 1 as an example.
FIG. 6 is a view illustrating a process of irradiating a light source to a part to be inspected at different irradiation angles through a plurality of light source modules of FIG. 1 and photographing the part to be inspected through a camera as an example.
7 is a perspective view illustrating the configuration of the component defect inspection apparatus according to the present disclosure as another example.
8 is a view showing an example of a process in which the transfer module of FIG. 7 moves.
9 is a view showing the configuration of a driving device included in the component defect inspection apparatus according to the present disclosure.
FIG. 10 is a diagram illustrating an example of a process of storing location characteristic data, light color data, and light amount data in the memory of FIG. 9 .
11 is a flowchart illustrating a component defect inspection method according to the present disclosure.
12 is a diagram illustrating an arrangement structure of a red LED element, a green LED element, a blue LED element, and a white LED element electrically connected to the sixth and seventh driving units of FIG. 9 as an example.
13 is a view illustrating a process of turning on the Red LED device through the sixth and seventh driving units of FIG. 9 as an example.
FIG. 14 is a diagram illustrating a process of turning on a Red LED device and a Green LED device through a sixth driving unit and a seventh driving unit of FIG. 9 as an example.
15 is a view illustrating a process of turning on the Red LED device through the eighth driving unit of FIG. 9 as an example.
16 is a diagram illustrating a process of turning on a Red LED device and a Green LED device through the eighth driving unit of FIG. 9 as an example.
17 to 22 are diagrams illustrating an example of a process of configuring a camera and a plurality of light source modules with an optical illumination system and an imaging system optimized for each display sample characteristic according to the present disclosure.

Like reference numerals refer to like elements throughout this disclosure. The present disclosure does not describe all elements of the embodiments, and general contents in the technical field to which the present disclosure pertains or overlapping contents between the embodiments are omitted. The term 'unit, module, member, block' used in the specification may be implemented in software or hardware, and a plurality of 'part, module, member, block' may be implemented as one component, It is also possible for one 'part, module, member, block' to include a plurality of components.

Throughout the specification, when a part is "connected" with another part, it includes not only a direct connection but also an indirect connection, and the indirect connection includes connection through a wireless communication network. do.

Also, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated.

Throughout the specification, when a member is said to be located "on" another member, this includes not only a case in which a member is in contact with another member but also a case in which another member exists between the two members.

Terms such as 1st, 2nd, etc. are used to distinguish one component from another component, and the component is not limited by the above-mentioned terms.

The singular expression includes the plural expression unless the context clearly dictates otherwise.

In each step, the identification code is used for convenience of description, and the identification code does not describe the order of each step, and each step may be performed differently from the specified order unless the specific order is clearly stated in the context. have.

Hereinafter, the working principle and embodiments of the present disclosure will be described with reference to the accompanying drawings.

In the present specification, the control unit of the component defect inspection apparatus according to the present disclosure includes various devices capable of providing a result to a user by performing an arithmetic process. For example, the control unit of the component defect inspection apparatus according to the present disclosure includes all of a computer, a server apparatus, and a portable terminal, or may be in any one form.

Here, the computer may include, for example, a laptop equipped with a web browser, a desktop, a laptop, a tablet PC, a slate PC, and the like.

The server device is a server that processes information by communicating with an external device, and may include an application server, a computing server, a database server, a file server, a mail server, a proxy server, a web server, and the like.

A portable terminal is, for example, a wireless communication device that guarantees portability and mobility, and includes a Personal Communication System (PCS), a Global System for Mobile communications (GSM), a Personal Digital Cellular (PDC), a Personal Handyphone System (PHS), and a PDA (Personal Communication System). Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminal, smart phone All kinds of handheld-based wireless communication devices, such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or wearable devices such as head-mounted-devices (HMDs) may include

The part defect inspection apparatus using artificial intelligence-based image recognition according to the present disclosure extracts image characteristic information of the inspection object part from the image based on artificial intelligence when an image including the inspection object part is received, and memory Searching for positional characteristic information of a plurality of light source modules associated with the extracted image characteristic information in the , by irradiating a light source to the parts to be inspected at different irradiation angles to form photographing lighting, and may be provided to photograph the parts to be inspected.

Such a part defect inspection device using artificial intelligence-based image recognition can prevent ignoring defects of the parts to be inspected, accurately identify the defects of the parts to be inspected, and The inspection time can be shortened.

Hereinafter, an apparatus for inspecting parts defects using artificial intelligence-based image recognition will be described in detail.

1 is a perspective view illustrating the configuration of a component defect inspection apparatus according to the present disclosure as an example. FIG. 2 is a view illustrating a process in which the transfer module of FIG. 1 moves as an example.

3 and 4 are views illustrating an arrangement structure between a plurality of light source modules of FIG. 1 as an example. 5 is a view illustrating a plurality of light source modules and a diffusion plate mounted on the camera of FIG. 1 as an example.

1 to 5, the component defect inspection apparatus 1000 includes a plate 101, supports 102a, 102b, transfer modules 103, 104, 111, a plurality of light source modules 105, 106, 109, 112 ), diffusion plates 107 , 108 , 110 , 113 , and a camera 120 .

The plate 101 may be provided to seat the component A to be inspected. In this case, the inspection target component (A) may be any component for inspecting defects of components, such as components mounted on a board, equipment components, and the like.

The support 102a is disposed along the periphery of the part to be inspected (A), forms a movement path of the transfer modules 103 and 104, and is parallel to the central axis of the plate 101 on the upper portion of the plate 101 can be provided. In this case, the support 102 may include a guide rail, and the transfer modules 103 and 104 may ascend or descend along the guide rail.

The transfer modules 103 and 104 may be provided to move the plurality of light source modules 105 and 106 . Here, the plurality of light source modules 105 and 106 may irradiate light sources to the part A to be inspected at different irradiation angles to form photographing illumination. In this case, the plurality of light source modules 105 and 106 may include four-division LED elements 105a and 106a each including a Red LED element, a Green LED element, a Blue LED element, and a White LED element.

The transfer modules 103 and 104 may include a first transfer module 103 and a second transfer module 104 .

The first transfer module 103 may be provided to change the first height and the first irradiation angle of the plurality of first light source modules 105 . The second transfer module 104 may be provided to change the second height and the second irradiation angle of the plurality of second light source modules 106 .

At this time, as shown in FIGS. 3 and 4 , the plurality of first light source modules 105 and the plurality of second light source modules 106 may be disposed at equal intervals from each other, and the plurality of first light source modules 105 may be disposed at equal intervals. The interval L1 between the plurality of second light source modules 106 may be wider than the interval L2 between the plurality of second light source modules 106 .

The first transfer module 103 may include a first frame 103a, a first connection member 103b, and a first angle adjustment member 103c. The first frame 103a may move in the longitudinal direction of the support 102a while being supported by the support 102a. For example, the first frame 103a may be raised or lowered along a guide rail included in the support 102a. The first connecting member 103b may be provided to connect the first frame 103a and the plurality of first light source modules 105 . The first angle adjusting member 103c is connected to the plurality of first light source modules 105 through the side surface of the first connecting member 103b, and can adjust the irradiation angles of the plurality of first light source modules 105 . . Here, the first angle adjusting member 103c may be axially connected to the plurality of first light source modules 105 , and may adjust the irradiation angles of the plurality of first light source modules 105 according to the tilting angle. In this case, the quadrant LED elements 105a included in the plurality of first light source modules 105 may be provided to face the component A to be inspected.

The component defect inspection apparatus 1000 according to the present disclosure may further include a first diffusion plate 107 . The first diffusion plate 107 may be mounted on the plurality of first light source modules 105 or provided on the front surface of the plurality of first light source modules 105 to diffuse the light sources of the plurality of first light source modules 105 . In this case, the first diffusion plate 107 may be made of a polycarbonate material that spreads light. The first diffusion plate 107 may increase light uniformity and maximum brightness, and may disperse the light emitted from the quadruple LED element 105a as a whole.

The second transfer module 104 may include a second frame 104a, a second connection member 104b, and a second angle adjustment member 104c. The second frame 104a may move in the longitudinal direction of the support 102a while being supported by the support 102a. For example, the second frame 104a may be raised or lowered along a guide rail included in the support 102a. The second connecting member 104b may be provided to connect the second frame 104a and the plurality of second light source modules 106 . The second angle adjusting member 104c is connected to the plurality of second light source modules 106 through the side surface of the second connecting member 104b, and can adjust the irradiation angles of the plurality of second light source modules 106 . . Here, the second angle adjusting member 104c may be axially connected to the plurality of second light source modules 106 and may adjust the irradiation angles of the plurality of second light source modules 106 according to the tilting angle. In this case, the quadrant LED element 106a included in the plurality of second light source modules 106 may be provided to face the component A to be inspected.

The component defect inspection apparatus 1000 according to the present disclosure may further include a second diffusion plate 108 . The second diffusion plate 108 may be mounted on the plurality of second light source modules 106 or provided on the front surface of the plurality of second light source modules 106 to diffuse the light sources of the plurality of second light source modules 106 . In this case, the second diffusion plate 108 may be made of a polycarbonate material that spreads light. The second diffusion plate 108 may increase light uniformity and maximum brightness, and may disperse the light emitted from the quadruple LED element 106a as a whole.

5 , the component defect inspection apparatus 1000 according to the present disclosure may further include a third light source module 109 and a third diffusion plate 110 .

The third light source module 109 may be disposed under the camera 120 . In this case, the third light source module 109 may include white LED elements 109a disposed at equal intervals from each other. The third diffusion plate 110 may be mounted on the lower portion of the third light source module 109 or provided on the front surface of the third light source module 109 to diffuse the light source of the third light source module 109 . In this case, the third diffusion plate 110 may be made of a polycarbonate material that spreads light. The third diffusion plate 110 may increase light uniformity and maximum brightness, and may disperse the light emitted from the white LED element 109a as a whole.

FIG. 6 is a view illustrating a process of irradiating a light source to a part to be inspected at different irradiation angles through a plurality of light source modules of FIG. 1 and photographing the part to be inspected through a camera as an example.

Referring to FIG. 6 , the component defect inspection apparatus 1000 according to the present disclosure includes a plurality of first light source modules 105 including a quadruple LED 105a, and a plurality of second light source modules including a quadruple LED 106a. The light source module 106, the third light source module 109 including the White LED element 109a, the transfer modules 103, 104, and the diffusion plate 107, 108, 110 through the optimal light source irradiation angle, Light color, optimal light intensity, optimal light uniformity, and optimal brightness may be irradiated to the part A to be inspected, and the part A to be inspected may be photographed through the camera 120 .

7 is a perspective view illustrating the configuration of the component defect inspection apparatus according to the present disclosure as another example. 8 is a view showing an example of a process in which the transfer module of FIG. 7 moves.

On the other hand, as shown in FIGS. 7 and 8 , the support 102b of the component defect inspection apparatus 1000 according to the present disclosure is disposed along the circumference of the inspection target component A, and the movement of the transfer module 111 Forming a path, it may be provided on the upper portion of the plate 101 to be curved toward the central axis of the plate (101). In this case, the transfer module 111 may change the height and irradiation angle of the plurality of fourth light source modules 112 along the longitudinal direction of the support 102b.

The transfer module 111 may include a third frame 111a and a third connection member 111b. The third frame 111a may move in the longitudinal direction of the support 102b while being supported by the support 102b. For example, the third frame 111a may be raised or lowered along a guide rail included in the support 102b. The third connecting member 111b may be provided to connect the third frame 111a and the plurality of fourth light source modules 112 to each other. Here, as the plurality of fourth light source modules 112 move in the longitudinal direction of the support 102b by the transfer module 111 , the height and the irradiation angle may be adjusted. In this case, the quadrant LED element 112a included in the plurality of fourth light source modules 112 may be provided to face the component A to be inspected.

The component defect inspection apparatus 1000 according to the present disclosure may further include a fourth diffusion plate 113 . The fourth diffusion plate 113 may be mounted on the plurality of fourth light source modules 112 or provided on front surfaces of the plurality of fourth light source modules 112 to diffuse the light sources of the plurality of fourth light source modules 112 . In this case, the fourth diffusion plate 113 may be made of a polycarbonate material that spreads light. The fourth diffusion plate 113 may increase light uniformity and maximum brightness, and may disperse the light emitted from the quadruple LED element 112a as a whole.

9 is a view showing the configuration of a driving device included in the component defect inspection apparatus according to the present disclosure.

Referring to FIG. 9 , the controller 210 uses an algorithm for controlling the operation of components in the device or a memory 211 for storing data for a program reproducing the algorithm, and data stored in the memory 211 . Thus, it may be implemented as at least one processor 212 that performs the above-described operation. Here, the memory 211 and the processor 212 may be implemented as separate chips. Also, the memory 211 and the processor 212 may be implemented as a single chip.

The memory 211 may store data supporting various functions of the device, a program for the operation of the control unit, store input/output data, and a plurality of application programs (application programs or applications) driven in the device. (application)), data for operation of the device, and commands may be stored. At least some of these application programs may be downloaded from an external server through wireless communication.

The memory 211 may include a flash memory type, a hard disk type, a solid state disk type (SSD), a silicon disk drive type (SDD), and a multimedia card micro type. micro type), card type memory (such as SD or XD memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable memory (EEPROM) It may include a storage medium of at least one type of a programmable read-only memory), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. In addition, although the memory 211 is separated from the present device, it may be a database connected by wire or wirelessly.

In the memory 211 , image characteristic information of a corresponding inspection target part for each photographing illumination and location characteristic information of the plurality of light source modules 105 , 106 and 112 may be linked and stored. Also, the memory 211 may store image characteristic information of a corresponding inspection target part for each photographing illumination and light color information and light amount information of the plurality of light source modules 105 , 106 , 109 , and 112 in association with each other. In this case, the parts to be inspected may be all parts for inspecting defects of parts, such as parts mounted on a board, equipment parts, and the like.

FIG. 10 is a diagram illustrating an example of a process of storing location characteristic data, light color data, and light amount data in the memory of FIG. 9 .

Referring to FIG. 10 , the server 10 electrically connected to the control unit 210 converts input values of photographing lighting condition data ID1 , which are metadata, and image characteristic data ID2 of a part to be inspected, to a part inspection model (M). Outputs the result values of position characteristic data OD1 (height, irradiation angle), light color data OD2, and light intensity data OD3 of the plurality of light source modules 105, 106, and 112 recommended by learning based on can do.

In this case, the photographing lighting condition data ID1 may include at least one of a brightness state, a glare state, a spectral distribution state, a luminance distribution state, and a shaded state for taking photographing lighting into consideration for photographing the part to be inspected. In addition, the image characteristic data ID2 of the part to be inspected may include at least one of a type, a standard, a height, a width, a volume, a three-dimensional surface, an edge point, and a reflectivity according to a material of the part to be inspected. Also, the light color data OD2 may be R (red), G (green), B (blue), or W (white). In addition, the light intensity data OD3 may be a light intensity value for forming the photographing illumination of the part to be inspected in an optimal state.

The parts inspection model M may be constructed to learn the photographing lighting condition data ID1 included in the input data and the image characteristic data ID2 of the parts to be inspected through correlation. The part inspection model M may be constructed and reinforced with a training data set using the photographing lighting condition data ID1 and the image characteristic data ID2 of the part to be inspected using a CNN algorithm or an RNN algorithm.

The server 10 determines the location characteristic data OD1 (height, irradiation angle), light color data OD2 (R (red), G (green), B (blue), W (white)) and light amount data OD3 may be transmitted to the memory 211 of the controller 210 . . The memory 211 includes the received positional characteristic data OD1 (height, irradiation angle) and light color data OD2 (R (red), G (green), B (blue), W (white)) and light intensity data (OD3) can be stored. At this time, the memory 211 stores the optimal image characteristic data of the corresponding inspection target part for each photographing lighting and the optimal positional characteristic data (height, irradiation angle) of the plurality of light source modules 105 , 106 and 112 in association. can be In addition, the memory 211 stores the optimal image characteristic data of the corresponding inspection target part for each photographing illumination, the optimal light color data and the optimal light intensity data of the plurality of light source modules 105 , 106 , and 112 are linked and stored. can be

When an image including the part A to be inspected is received from the camera 120 , the processor 212 may extract image characteristic information of the part A to be inspected from within the image based on artificial intelligence. Here, the processor 212 may extract image characteristic information of the part A to be inspected by using the object detection model. In this case, the image characteristic information may include at least one of a type, a standard, a height, a width, a volume, a three-dimensional surface, an edge point, and a reflectivity according to a material.

The processor 212 includes the location characteristic information ( height, irradiation angle) can be explored.

The processor 212 moves the first frame 103a to change the heights of the plurality of first light source modules 105 based on the position characteristic information (height, irradiation angle) found by the first transfer module 103 The first driving unit 221 may be driven, and the second driving unit 222 may be driven to adjust the tilting angle of the first angle adjusting member 103c to change the irradiation angles of the plurality of first light source modules 105 . can The first frame 103a may move the heights of the plurality of first light source modules 105 to an optimal target height. The first angle adjusting member 103c may adjust the irradiation angles of the plurality of first light source modules 105 to an optimal target irradiation angle. In this case, the first driving unit 221 and the second driving unit 222 may include at least one of a mechanical driving part and an electronic driving part to move the first frame 103a, and the first angle adjusting member 103c ) tilt angle can be adjusted.

The processor 212 moves the second frame 104a to change the heights of the plurality of second light source modules 106 based on the position characteristic information (height, irradiation angle) that the second transfer module 104 has searched for. It is possible to drive the third driving unit 223 and drive the fourth driving unit 224 for adjusting the tilting angle of the second angle adjusting member 104c to change the irradiation angles of the plurality of second light source modules 106 . can The second frame 104a may move the plurality of second light source modules 106 to an optimal target height. The second angle adjusting member 104c may adjust the irradiation angles of the plurality of second light source modules 106 to an optimal target irradiation angle. In this case, the third driving unit 223 and the fourth driving unit 224 may include at least one of a mechanical driving part and an electronic driving part to move the second frame 104a, and the second angle adjusting member 104c ) tilt angle can be adjusted.

The processor 212 is a fifth driving unit 225 that moves the third frame 111a to change the heights of the plurality of fourth light source modules 112 based on the location characteristic information (height) detected by the transfer module 111 . ) can be driven. The third frame 111a may move the heights of the plurality of fourth light source modules 112 to an optimal target height. The transfer module 111 may change the irradiation angle to the optimal target irradiation angle while changing the heights of the plurality of fourth light source modules 112 to the optimal target height along the longitudinal direction of the support 102b. In this case, the fifth driving unit 225 may include at least one of a mechanical driving part and an electronic driving part to move the third frame 111a and adjust the tilting angle of the second angle adjusting member 104c. have.

The processor 212 includes the light color information ( R (red), G (green), B (blue), W (white)) and light intensity information can be further explored.

The processor 212 is configured to configure Red LEDs in the plurality of first light source modules 105 based on the retrieved light color information (R (red), G (green), B (blue), W (white)) and light intensity information. The sixth driving unit 226 for turning on the corresponding LED device may be driven so that the light source is irradiated with the light color and light intensity of at least one of the device, the Green LED device, the Blue LED device, and the White LED device. The corresponding LED elements in the plurality of first light source modules 105 may irradiate the light source to the component A to be inspected with an optimum light color and optimum light intensity. In this case, the sixth driving unit 226 may include at least one of a mechanical driving component and an electronic driving component to turn on the corresponding LED device.

The processor 212 is configured to configure Red LEDs in the plurality of second light source modules 106 based on the retrieved light color information (R (red), G (green), B (blue), W (white)) and light intensity information. The seventh driving unit 227 for turning on the corresponding LED device may be driven so that the light source is irradiated with the light color and light intensity of at least one of the device, the Green LED device, the Blue LED device, and the White LED device. The corresponding LED elements in the plurality of second light source modules 106 may irradiate the light source to the part A to be inspected with an optimal light color and optimal light intensity. In this case, the seventh driving unit 226 may include at least one of a mechanical driving component and an electronic driving component to turn on the corresponding LED device.

The processor 212 performs the red LEDs in the plurality of fourth light source modules 112 based on the retrieved light color information (R (red), G (green), B (blue), W (white)) and light intensity information. The eighth driving unit 228 that turns on the corresponding LED device may be driven so that the light source is irradiated with the light color and light intensity of at least one of the device, the Green LED device, the Blue LED device, and the White LED device. The corresponding LED elements in the plurality of fourth light source modules 112 may irradiate the light source to the component A to be inspected with an optimum light color and optimum light intensity. In this case, the eighth driving unit 228 may include at least one of a mechanical driving component and an electronic driving component to turn on the corresponding LED device.

11 is a flowchart illustrating a component defect inspection method according to the present disclosure.

11 , the component defect inspection method may include a determination step (S1101), an extraction step (S1102), a search step (S1103), a control step (S1104), an investigation step (S1105), and a photographing step (S1106). have.

In the determining step, the processor 212 may determine whether an image including the part A to be inspected is received from the camera 120 ( S1101 ).

In the extraction step, when an image including the part A to be inspected is received through the processor 212, image characteristic information of the part A to be inspected may be extracted from the image based on artificial intelligence ( S1102). Here, the processor 212 may extract image characteristic information of the part A to be inspected by using the object detection model. In this case, the image characteristic information may include at least one of a type, a standard, a height, a width, a volume, a three-dimensional surface, an edge point, and a reflectivity according to a material.

In the search step, through the processor 212 , the plurality of light source modules 105 , 106 , 112 associated with the image characteristic information of the corresponding inspection target part A according to the photographing lighting condition extracted from the memory 211 . ) can be searched for location characteristic information (height, irradiation angle) (S1103). In addition, the search step, through the processor 212, the plurality of light source modules 105, 106 associated with the image characteristic information of the corresponding inspection target part (A) according to the photographing lighting conditions extracted from the memory 211 , 112), light color information (R (red), G (green), B (blue), W (white)) and light intensity information may be further searched (S1103).

The control step is, through the processor 212, the position (height) of the plurality of light source modules 105, 106, 112 based on the position characteristic information (height, irradiation angle) in which the transfer modules 103, 104, 111 are searched. , irradiation angle) can be controlled to change (S1104).

For example, in the control step, through the processor 212 , the first transfer module 103 changes the heights of the plurality of first light source modules 105 based on the searched position characteristic information (height, irradiation angle). It is possible to drive the first driving unit 221 to move the first frame 103a, and to adjust the tilting angle of the first angle adjusting member 103c to change the irradiation angle of the plurality of first light source modules 105 . The second driving unit 222 may be driven. The first frame 103a may move the heights of the plurality of first light source modules 105 to an optimal target height. The first angle adjusting member 103c may adjust the irradiation angles of the plurality of first light source modules 105 to an optimal target irradiation angle.

As another example, in the control step, through the processor 212 , the second transfer module 104 changes the height of the plurality of second light source modules 106 based on the searched position characteristic information (height, irradiation angle). may drive the third driving unit 223 for moving the second frame 104a to adjust the tilting angle of the second angle adjusting member 104c to change the irradiation angle of the plurality of second light source modules 106 . The fourth driving unit 224 may be driven. The second frame 104a may move the plurality of second light source modules 106 to an optimal target height. The second angle adjusting member 104c may adjust the irradiation angles of the plurality of second light source modules 106 to an optimal target irradiation angle.

As another example, in the control step, through the processor 212 , the transfer module 111 changes the height of the plurality of fourth light source modules 112 based on the searched position characteristic information (height) of the third frame. The fifth driving unit 225 that moves the 111a may be driven. The third frame 111a may move the heights of the plurality of fourth light source modules 112 to an optimal target height. The transfer module 111 may change the irradiation angle to the optimal target irradiation angle while changing the heights of the plurality of fourth light source modules 112 to the optimal target height along the longitudinal direction of the support 102b.

In the irradiation step, through the processor 212, based on the searched light color information and light intensity information, Red LED elements, Green LED elements, Blue LED elements, and White LED elements in the plurality of light source modules 105, 106, 112. The corresponding LED element may be turned on so that the light source is irradiated with at least one light color and light intensity of at least one of them. In the irradiation step, through the plurality of light source modules 105 , 106 , 112 , the light source may be irradiated to the part A to be inspected at different irradiation angles to form photographing lighting ( S1105 ).

12 is a diagram illustrating an arrangement structure of a red LED element, a green LED element, a blue LED element, and a white LED element electrically connected to the sixth and seventh driving units of FIG. 9 as an example.

Referring to FIG. 12 , the sixth driving unit 226 includes a four-division LED element 105a including a Red LED element, a Green LED element, a Blue LED element, and a White LED element in the plurality of first light source modules 105 and the electrical can be connected to The seventh driving unit 227 may be electrically connected to the quadruple LED element 106a including the Red LED element, the Green LED element, the Blue LED element, and the White LED element in the plurality of second light source modules 106 .

13 is a diagram illustrating a process of turning on the Red LED device through the sixth and seventh driving units of FIG. 9 as an example. 14 is a diagram illustrating a process of turning on a Red LED device and a Green LED device through the sixth and seventh driving units of FIG. 9 as an example.

Referring to FIG. 13 , the sixth driving unit 226 drives the Red LED element 105a1 so that the light source is irradiated with the light color and light intensity intensity of the Red LED element 105a1 in the plurality of first light source modules 105 . can The seventh driver 227 may drive the Red LED element 106a1 so that the light source is irradiated with the light color and light intensity of the Red LED element 106a1 in the plurality of second light source modules 106 . The red LED elements 105a1 and 106a1 may irradiate a light source to the part A to be inspected with an optimal light color and optimal light intensity.

Referring to FIG. 14 , the sixth driving unit 226 is configured such that the light source is irradiated with the light color and light intensity of the Red LED elements 105a1 and the Green LED elements 105a2 in the plurality of first light source modules 105 , the Red LED The device 105a1 and the Green LED device 105a2 can be driven. The seventh driving unit 227 is configured to irradiate the light source with the light color and light intensity of the Red LED element 106a1 and the Green LED element 106a2 in the plurality of second light source modules 106, the Red LED element 106a1 and the Green The LED element 106a2 can be driven. The red LED elements 105a1 and 106a1 and the green LED elements 105a2 and 106a2 may irradiate a light source to the part A to be inspected with an optimal light color and optimal light intensity.

On the other hand, the sixth driving unit 226 and the seventh driving unit 227 are various types of LEDs, such as a Green LED element, a Blue LED element, a White LED element, a Red LED element and a Blue LED element, a Green LED element, and a Blue LED element. device can be driven

15 is a view illustrating a process of turning on the Red LED device through the eighth driving unit of FIG. 9 as an example. 16 is a diagram illustrating a process of turning on a Red LED device and a Green LED device through the eighth driving unit of FIG. 9 as an example.

Referring to FIG. 15 , the eighth driving unit 228 drives the Red LED element 112a1 so that the light source is irradiated with the light color and light intensity of the Red LED element 112a1 in the plurality of fourth light source modules 112 . can Referring to FIG. 16 , the eighth driving unit 228 is configured to irradiate the light source with the light color and light intensity of the Red LED element 112a1 and the Green LED element 112a2 in the plurality of fourth light source modules 112 , the Red LED The device 112a1 and the Green LED device 112a2 may be driven. The red LED element 112a1 and the green LED element 112a2 may irradiate the light source to the part A to be inspected with an optimal light color and optimal light intensity.

Meanwhile, the eighth driving unit 228 may drive various types of LED devices such as a Green LED device, a Blue LED device, a White LED device, a Red LED device and a Blue LED device, and a Green LED device and a Blue LED device.

In the photographing step, the inspection target part A may be photographed through the camera 120 ( S1106 ). The camera 120 may photograph the inspection target part A under optimal photographing lighting conditions. The operator may check the image of the optimal inspection target part obtained by the camera 120 .

Therefore, the apparatus and method for inspecting parts defects using artificial intelligence-based image recognition according to the present disclosure can prevent ignoring defects of the parts to be inspected, and can accurately identify the defects of the parts to be inspected, It is possible to shorten the inspection time to confirm the defect of the part.

Meanwhile, in the present disclosure, as shown in FIG. 17 , a camera and a plurality of light source modules may be configured as an optical illumination system and an imaging system optimized for each display sample characteristic. In this case, the camera and the lens may be selected or designed in consideration of lighting arrangement, spacing, lighting quantity, lighting wavelength, lighting angle, and lighting distance. In addition, the present disclosure can visualize and measure a defect corresponding to a specific height by using a 2.5D optical system and using an algorithm called Photometric Stereo of Computer Vision. Here, the intaglio defect may be a dent, a scratch, etc., and the embossed defect may be a dust, a burr, or the like.

That is, in the present disclosure, as shown in FIGS. 18 to 22 , different materials were verified with a display sample and a metal material sample. For example, the display sample may be a display such as a mobile phone cover glass or a TV, and the metal sample may be a product such as an automobile part. The present disclosure can visualize and measure negative and positive defects through an optical illumination system and an imaging system optimized for each display sample characteristic. At this time, as shown in FIG. 20 , according to the present disclosure, information on convexity and surface roughness of the fine protrusion of 100 μm or less of the metal surface may be obtained. In addition, as shown in FIG. 21 , the present disclosure may obtain surface texture information after restoration and processing of the shape of the embossed character having a height of 2 mm or more of the metal surface. Also, as shown in FIG. 22 , the present disclosure may measure the degree of inclination using slope information.

At least one component may be added or deleted according to the performance of the components shown in FIGS. 1 to 10 . In addition, it will be readily understood by those of ordinary skill in the art that the mutual positions of the components may be changed corresponding to the performance or structure of the system.

11 shows that a plurality of steps are sequentially executed, but this is merely illustrative of the technical idea of the present embodiment, and those of ordinary skill in the art to which this embodiment belongs are essential characteristics of this embodiment 11 is not limited to a time-series order, since it will be variously modified and modified to execute by changing the order described in FIG. .

The disclosed embodiments have been described with reference to the accompanying drawings as described above. Those of ordinary skill in the art to which the present disclosure pertains will understand that the present disclosure may be practiced in other forms than the disclosed embodiments without changing the technical spirit or essential features of the present disclosure. The disclosed embodiments are illustrative and should not be construed as limiting.

120: camera 200: driving device
210: control unit 211: memory
212: processor 221: first driving unit
222: second driving unit 223: third driving unit
224: fourth driving unit 225: fifth driving unit
226: sixth driving unit 227: seventh driving unit
228: eighth driving unit

Claims (10)

camera;
a plurality of light source modules for irradiating light sources to parts to be inspected at different irradiation angles to form photographic lighting;
a transport module for moving the plurality of light source modules;
a memory in which image characteristic information of a corresponding inspection target part for each photographing light and location characteristic information of the plurality of light source modules are linked and stored; and
When an image including the part to be inspected is received from the camera, image characteristic information of the part to be inspected is extracted from the image based on artificial intelligence,
Searching for positional characteristic information of a plurality of light source modules associated with the extracted image characteristic information in the memory,
a processor for controlling the transfer module to change positions of the plurality of light source modules based on the searched position characteristic information; including,
The memory is
image characteristic information of the corresponding inspection target part for each photographing illumination and light color information and light intensity information of the plurality of light source modules are further linked and stored;
The processor is
Further searching for light color information and light intensity information of a plurality of light source modules associated with the extracted image characteristic information in the memory,
Based on the searched light color information and light intensity information, the light source is irradiated with the light color and light intensity of at least one of Red LED elements, Green LED elements, Blue LED elements, and White LED elements in the plurality of light source modules, turn on the LED element,
Further comprising a plate for seating the part to be inspected;
It is disposed along the periphery of the part to be inspected, and forms a movement path of the transfer module, further comprising a support provided on the plate to be curved toward the central axis of the plate;
The transfer module is
While being supported on the support, but moving in the longitudinal direction of the support,
a third frame elevating or descending along the guide rail included in the support; and
a third connecting member connecting the third frame and the plurality of fourth light source modules;
The plurality of fourth light source modules are adjusted in height and irradiation angle as they move in the longitudinal direction of the support by the transfer module,
The server electrically connected to the processor,
Input values of imaging lighting condition data, which are metadata, and image characteristic data corresponding to image characteristic information of parts to be inspected, into the parts inspection model;
Outputs result values of location property data corresponding to location property information of a plurality of light source modules that are learned and recommended based on the part inspection model, light color data corresponding to light color information, and light intensity data corresponding to light intensity information Part defect inspection device using artificial intelligence-based image recognition, characterized in that. delete delete delete delete delete delete delete delete In a component defect inspection method using artificial intelligence-based image recognition performed by a device,
determining, through the processor of the device, whether an image including the part to be inspected is received from the camera of the device;
when an image including the part to be inspected is received through the processor of the apparatus, extracting image characteristic information of the part to be inspected from within the image based on artificial intelligence;
searching for positional characteristic information of a plurality of light source modules associated with the extracted image characteristic information in a memory of the apparatus through a processor of the apparatus;
controlling, through the processor of the apparatus, the transfer module of the apparatus to change the positions of the plurality of light source modules based on the searched position characteristic information;
forming photographic lighting by irradiating light sources to the parts to be inspected at different irradiation angles through the plurality of light source modules; and
photographing the part to be inspected through the camera of the device; including,
The memory is
image characteristic information of the corresponding inspection target part for each photographing illumination and light color information and light intensity information of the plurality of light source modules are further linked and stored;
The processor is
Further searching for light color information and light intensity information of a plurality of light source modules associated with the extracted image characteristic information in the memory,
Based on the searched light color information and light intensity information, the light source is irradiated with the light color and light intensity of at least one of Red LED elements, Green LED elements, Blue LED elements, and White LED elements in the plurality of light source modules, turn on the LED element,
Further comprising a plate for seating the part to be inspected;
It is disposed along the periphery of the part to be inspected, and forms a movement path of the transfer module, further comprising a support provided on the plate to be curved toward the central axis of the plate;
The transfer module is
While being supported on the support, but moving in the longitudinal direction of the support,
a third frame elevating or descending along the guide rail included in the support; and
a third connecting member connecting the third frame and the plurality of fourth light source modules;
The plurality of fourth light source modules are adjusted in height and irradiation angle as they move in the longitudinal direction of the support by the transfer module,
The server electrically connected to the processor,
Input values of imaging lighting condition data, which are metadata, and image characteristic data corresponding to image characteristic information of parts to be inspected, into the parts inspection model;
Outputs result values of location property data corresponding to location property information of a plurality of light source modules that are learned and recommended based on the part inspection model, light color data corresponding to light color information, and light intensity data corresponding to light intensity information A method, characterized in that

Images