KR20230056813A - System for outer inspection of inspecting object using machine vision algorithm based on artificial intelligent - Google Patents

Abstract

The present invention relates to an appearance inspection system for a to-be-inspected object using artificial intelligence-based machine vision algorithm, which can perform a defect inspection more precisely and effectively. The appearance inspection system comprises: a lighting device including first to fourth lighting modules; a line scan camera device for acquiring a two-dimensional image; and a machine vision control device for performing an external inspection of a corresponding to-be-inspected object.

Description

Appearance inspection system of inspected object using artificial intelligence-based machine vision algorithm

The present invention relates to a system for inspecting the appearance of a subject using an artificial intelligence (AI)-based machine vision algorithm.

In general, with the development of technology, it has become possible to produce articles of various types through various methods. For example, conventionally, an article having a desired shape can be produced through cutting or grinding. Alternatively, a so-called plastic injection method in which a liquid material, for example, liquid plastic is injected into a metal mold made of various materials and then cooled and injected to form a part, has also been utilized.

On the other hand, in order to cope with the increase in population and the growth of the market, manufacturers who manufacture products are faced with the task of producing a variety of products at low cost. In particular, as the use cycle of various products gradually shortens, the time required to equip mass production facilities for products has also become shorter.

Here, along with rapid mass production, another challenge given to manufacturers is to maintain the yield of the produced article. At the time when products were produced by handicraft, workers directly performed and controlled all production work, so despite the design of the same product, the quality of the product was consistently maintained by various environmental factors such as individual worker's skill or fatigue. It was practically impossible to maintain.

On the other hand, uniformity in manufacturing has been able to be obtained to a large extent since the era of mass automatic production by machines, but inspections for various mechanical errors and defects occurring during the production and assembly process of each item are still performed by workers. The method of visually inspecting each worker is used, which is not only expensive, but also consistent standards for judging product quality are applied due to the skill level or fatigue of each worker or the difference in individual standards, just like the problems of the handmade product production era. There are still issues that may not be addressed.

In addition, consumers' standards for quality are also increasing day by day, and manufacturers are also making efforts to eliminate defective products in the production and assembly processes of products, intermediate processes and final assembly processes. As part of these efforts, in the assembly process, high and uniform quality of the final product is ensured by using various inspection equipment to determine whether the product satisfies the standard quality, and removing intermediate assembly products judged to be of poor quality in advance. Maintaining and obtaining a high yield is one subject.

Looking at the method of inspecting intermediate goods or final products generated in the manufacturing process as described above in more detail, the method currently used in the industrial field is to take a video of the item to be inspected using a photographing equipment, and then take a picture. There is a method in which the workers who perform the inspection of the images are manually checked one by one.

However, as pointed out above, this method not only incurs excessive costs for inspection, but also the quality is easily affected by various external factors such as individual worker's skill, fatigue, and working environment, as in the production process of conventional products. The speed at which it is performed is also not sufficient, which is likely to affect the productivity of the article.

Therefore, in order to solve these problems, it is required to introduce an inspection system capable of automatically inspecting an item according to the same standard, and to introduce an inspection system capable of more precisely and effectively inspecting an item. do.

Domestic Patent No. 10-1605229 (Announced on March 21, 2016)

The present invention has been devised to solve the above problems, and an object of the present invention is to scan the surface image of the subject in line units through a line scan camera device in a state where a plurality of lights are sequentially irradiated to the subject to be inspected 2 After obtaining an image with a dimensional image, the entire image with the acquired two-dimensional image of the subject is used to perform an external inspection of the subject using an artificial intelligence (AI)-based machine vision algorithm. A system for inspecting the exterior of a subject using artificial intelligence-based machine vision algorithms that not only automatically and efficiently inspects the exterior of a corpse, but also enables more precise and effective inspection of defects on the subject. is in providing

In order to achieve the above object, one aspect of the present invention is a first lighting module for radiating light to the first surface and upper surface of an object to be inspected, and a second illumination module for radiating light to the second surface and upper surface of the object to be inspected. A lighting device having a lighting module, a third lighting module for radiating light to a third surface and an upper surface of the object to be inspected, and a fourth lighting module for radiating light to a fourth surface and an upper surface of the object to be inspected; It is installed at a certain distance perpendicular to the traveling direction of the subject, and each light output from the first to fourth lighting modules of the lighting device is sequentially irradiated to the subject, a line scan camera device for obtaining a two-dimensional image by scanning an upper surface image; and controlling the operation of the first to fourth lighting modules so that each light output from the first to fourth lighting modules of the lighting device is sequentially irradiated to the target object, and the first to fourth lighting modules of the lighting device are sequentially irradiated. Controlling the operation of the line scan camera device so that each light output from the lighting module is sequentially irradiated to the object to be inspected so that the object to be inspected is photographed in line units, and An entire image with a 2D image is provided, and using this, a 2D image of the subject for each lighting module is separated, and the 2D images of the subject for each of the separated lighting modules are overlapped into one combined image. After generating, artificial intelligence-based including a machine vision control device that performs an external inspection of the subject using a predetermined artificial intelligence (AI)-based machine learning method based on the generated one combined image. It is to provide a system for inspecting the appearance of a subject using a machine vision algorithm.

Here, the inspected object is preferably made of a polyhedron-shaped article having a plurality of faces.

Preferably, a test object transfer device for seating the test object and transporting the test object in one direction may be further included.

Preferably, one combined image generated by the machine vision controller may consist of a 2.5-dimensional (2.5D) image.

Preferably, the external appearance inspection of the corresponding inspected object performed by the machine vision control device is determined as a good product or a defective product, and the defective product may be classified and determined by at least one surface state among scratches, dents, foreign matter, and marking states. there is.

Preferably, the machine vision control device performs an external inspection of the subject by using a pre-set rule-based machine vision algorithm based on the generated one combined image, and then inspects the appearance of the subject. Based on the non-defective/defective product result data for the external inspection, the artificial intelligence learning model for the inspected object is created by repeatedly learning each good/defective product feature item for the subject through a predetermined artificial intelligence (AI)-based machine learning method. can be built

Preferably, the set rule base-based machine vision algorithm is at least one of pattern matching, point/line/plane fitting, edge, color, and location algorithms. It can be done with one algorithm.

Preferably, the machine vision control device receives an entire image having a 2D image of the subject to be inspected obtained from the line scan camera device, and separates the 2D image of the subject for each lighting module using this image. separation module; an image recombination module for generating a single combined image by overlapping two-dimensional images of a corresponding object under examination for each lighting module separated from the image separation module; Based on one combined image generated from the image recombination module, through a predetermined artificial intelligence (AI)-based machine learning method, iteratively learns each characteristic item of a good/defective product for the subject to inspect the appearance of the subject An artificial intelligence learning module that builds an artificial intelligence learning model; Using the artificial intelligence learning model for the external inspection of the inspected object built from the artificial intelligence learning module, the surface characteristics of the inspected object are inferred and analyzed for each good/defective product characteristic item, and the inference and analysis Based on the surface characteristic information of the inspected object, the inspected object appearance judgment module automatically determines the inspected object as a defective product by classifying the inspected object into at least one surface state of a good product or scratch, scratch, foreign matter, and marking state; and controlling the operation of the first to fourth lighting modules of the lighting device so that each light output from the first to fourth lighting modules of the lighting device is sequentially irradiated to the corresponding object to be inspected. It may include a control module for controlling the operation of the line scan camera device so that the object to be inspected is photographed line by line in a state in which each light output from the fourth lighting module is sequentially irradiated to the object to be inspected.

Preferably, the artificial intelligence (AI)-based machine learning method applied to the artificial intelligence learning module is selected from among neural networks, support vector machines (SVMs), multi-layer perceptions (MLPs), and deep learnings. It can consist of at least one artificial intelligence learning method.

Preferably, each of the first to fourth lighting modules of the lighting device may include a plurality of LED light emitting channels connected in parallel to each other including at least one LED light emitting element.

Preferably, the machine vision control device sets a target output level value of each LED light emitting channel provided in the first to fourth lighting modules of the lighting device according to a user's specific input signal output according to the user's request; A user setting module for setting output channel selection and order of each LED light emitting channel for high-speed switching control, and clock generation for generating high-speed switching operation signals; Based on the received output level target value of each LED light emitting channel set from the user setting module, an output voltage target value of each LED light emitting channel is generated in order to adjust the brightness of each LED light emitting element provided in each LED light emitting channel. an output level control module that outputs; and the output voltage target value of each LED light emitting channel output from the output level control module in synchronization with the setting information and clock information of the output channel selection and sequence of each LED light emitting channel set from the user setting module. A high-speed switching control module for controlling high-speed switching of each LED light emitting channel provided in the first to fourth lighting modules may be further included.

Preferably, the user setting module may output a specific input signal of a corresponding user according to a request of the corresponding user based on a graphical user interface (GUI).

Preferably, the high-speed switching control module may perform high-speed control by using a hardware programming-based Field Programmable Gate Array (FPGA) for a high-speed switching function.

Preferably, the high-speed switching control module detects an idle period of a photographing operation for a machine vision camera device installed near the lighting device, and inputs a photographing operation signal of the machine vision camera apparatus for a predetermined photographing operation reference time or longer. An automatic reset function may be performed to determine an idle state when it is not, remove a state during a photographing operation and initialize it to an initial input state so that the first state is maintained after the determined idle state.

Preferably, the machine vision control device sets a set value for an output level target of each LED light emitting channel set from the user setting module, a set value of output channel selection and order of each LED light emitting channel, and clock generation. A storage module for storing at least one set value among the set values for may be further included.

Preferably, the user's specific input signal output according to the user's request may be provided through an external terminal or server.

Preferably, the external terminal or server transmits a specific input signal of the user for user setting of fast switching LED lighting control according to the user's request through a pre-installed application related to user setting for controlling fast switching LED lighting by wire. Alternatively, it may be wirelessly transmitted to the user setting module.

Preferably, the high-speed switching control module, in order to protect each LED light emitting element from damage due to the continuous turning on of each LED light emitting element provided in each LED light emitting channel of the first to fourth lighting modules, the first to fourth lighting modules When the high-speed switching operation signal for each LED light emitting channel of the first to fourth lighting modules is changed, each LED light emitting channel of the first to fourth lighting modules is controlled to perform high-speed switching only for a predetermined operation standard time or less. can do.

Preferably, the entire image having the two-dimensional image of the subject obtained from the linescan camera device, the two-dimensional image of the subject separated for each lighting module, the generated combined image, and the corresponding subject A storage device for storing external appearance test result information may be further included.

Preferably, the machine vision control device includes an entire image having a two-dimensional image of the subject obtained from the line scan camera device, a two-dimensional image of the subject separated for each lighting module, and the generated one A two-dimensional image of the subject separated by each lighting module, including the entire image having a two-dimensional image for each subject using the combined image and the external inspection result information of the subject, the generated combined image, and The external inspection result information of the object to be inspected can be controlled to be stored in the storage device by making a database (DB).

Preferably, the entire image having the two-dimensional image of the subject obtained from the linescan camera device, the two-dimensional image of the subject separated for each lighting module, the generated combined image, and the corresponding subject A display device for displaying the external inspection result information on the display screen may be further included.

Preferably, the machine vision control device includes an entire image having a two-dimensional image of the subject obtained from the line scan camera device, a two-dimensional image of the subject separated for each lighting module, and the generated one A two-dimensional image of the subject separated by each lighting module, including the entire image having a two-dimensional image for each subject using the combined image and the external inspection result information of the subject, the generated combined image, and An operation of the display device may be controlled to display external inspection result information of the object to be inspected on the display screen.

Preferably, the entire image having the two-dimensional image of the subject obtained from the linescan camera device, the two-dimensional image of the subject separated for each lighting module, the generated combined image, and the corresponding subject A communication device for transmitting the external inspection result information of the wired or wireless communication method may be further included.

Preferably, the machine vision control device includes an entire image having a two-dimensional image of the subject obtained from the line scan camera device, a two-dimensional image of the subject separated for each lighting module, and the generated one The operation of the communication device may be controlled to transmit the combined image and external inspection result information of the subject to be inspected to an external terminal or server in a wired or wireless communication method.

Preferably, the external terminal or server includes the entire image having the two-dimensional image of the subject transmitted from the communication device through a pre-installed application related to the appearance inspection of the subject, as well as the image of the subject separated for each lighting module. A two-dimensional image of the subject separated by each lighting module, including the entire image having a two-dimensional image for each subject using the two-dimensional image, the generated combined image, and the external inspection result information of the subject, A service may be provided so that the generated combined image and the result information of the external examination of the object to be searched are searched or displayed on a display screen.

According to the system for inspecting the appearance of a subject using the artificial intelligence-based machine vision algorithm of the present invention as described above, the subject is sequentially irradiated with a plurality of lights, and the subject is line-by-line through a line scan camera device. After obtaining an image with a 2-dimensional image by scanning the surface image of the object, using the entire image with the acquired 2-dimensional image of the object to be inspected, an artificial intelligence (AI)-based machine vision algorithm is used to determine the appearance of the object to be inspected. By performing the inspection, there is an advantage in that not only can an external inspection of the subject to be inspected be performed automatically and efficiently, but also a defect inspection of the subject to be inspected can be performed more precisely and effectively.

1 is an overall block configuration diagram for explaining a system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm according to an embodiment of the present invention.
2 is a perspective view for explaining the disposition of a lighting device and a linescan camera device applied to an embodiment of the present invention.
3 is a specific block configuration diagram for explaining first to fourth lighting modules of a lighting device applied to an embodiment of the present invention.
4 is a detailed block configuration diagram for explaining a machine vision control device applied to an embodiment of the present invention.
5 illustrates an entire image having a 2D image of a subject obtained through a linescan camera device applied to an embodiment of the present invention and a 2D image of a subject for each lighting module separated through a machine vision control device. It is a concept for
6 is a diagram illustrating a process of generating a single combined image by overlapping two-dimensional images of an object under examination for each lighting module separated through a machine vision control device applied to an embodiment of the present invention.

The above objects, features and advantages will be described later in detail with reference to the accompanying drawings, and accordingly, those skilled in the art to which the present invention belongs will be able to easily implement the technical spirit of the present invention. In describing the present invention, if it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

Terms including ordinal numbers, such as first and second, may be used to describe various components, but the components are not limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise.

The terms used in the present invention have been selected from general terms that are currently widely used as much as possible while considering the functions in the present invention, but these may vary depending on the intention of a person skilled in the art or precedent, the emergence of new technologies, and the like. In addition, in a specific case, there is also a term arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the invention. Therefore, the term used in the present invention should be defined based on the meaning of the term and the overall content of the present invention, not simply the name of the term.

When it is said that a certain part "includes" a certain component throughout the specification, it means that it may further include other components without excluding other components unless otherwise stated. In addition, terms such as "...unit" and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented as hardware or software or a combination of hardware and software. .

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the embodiments of the present invention exemplified below may be modified in many different forms, and the scope of the present invention is not limited to the embodiments described below. The embodiments of the present invention are provided to more completely explain the present invention to those skilled in the art.

Combinations of each block of the accompanying block diagram and each step of the flowchart may be performed by computer program instructions (execution engine), and these computer program instructions are executed by a processor of a general-purpose computer, special-purpose computer, or other programmable data processing equipment. Since it can be mounted, the instructions executed through the processor of a computer or other programmable data processing equipment create means for performing the functions described in each block of the block diagram or each step of the flowchart. These computer program instructions may also be stored in a computer usable or computer readable memory that can be directed to a computer or other programmable data processing equipment to implement functionality in a particular way, such that the computer usable or computer readable memory The instructions stored in are also capable of producing an article of manufacture containing instruction means for performing the functions described in each block of the block diagram or each step of the flow chart.

In addition, since the computer program instructions can be loaded on a computer or other programmable data processing equipment, a series of operational steps are performed on the computer or other programmable data processing equipment to create a computer-executed process to generate a computer or other programmable data processing equipment. It is also possible that instructions performing possible data processing equipment provide steps for executing the functions described in each block of the block diagram and each step of the flow chart.

In addition, each block or each step may represent a module, segment, or portion of code that includes one or more executable instructions for executing specified logical functions, and in some alternative embodiments may refer to blocks or steps. It should be noted that it is also possible for functions to occur out of order. For example, two blocks or steps shown in succession may in fact be performed substantially concurrently, or the blocks or steps may be performed in the reverse order of their corresponding functions, if necessary.

1 is an overall block configuration diagram for explaining a system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm according to an embodiment of the present invention, and FIG. 2 is a lighting device applied to an embodiment of the present invention and It is a perspective view for explaining the arrangement state of the linescan camera device, FIG. 3 is a specific block configuration diagram for explaining the first to fourth lighting modules of the lighting device applied to an embodiment of the present invention, and FIG. 4 is the present invention. 5 is a detailed block configuration diagram for explaining a machine vision control device applied to an embodiment of the present invention, and FIG. It is a conceptual diagram for explaining a two-dimensional image of an inspection subject for each lighting module separated through a machine vision control device, and FIG. 6 is an inspection subject for each lighting module separated through a machine vision control device applied to an embodiment of the present invention. It is a diagram showing a process of generating a single combined image by overlapping two-dimensional images of .

1 to 6, the system for inspecting the appearance of a subject using an artificial intelligence (AI)-based machine vision algorithm according to an embodiment of the present invention includes a lighting device 100 and a line scan camera device 200. ), a machine vision control device 300, and a power supply device 400. In addition, the system for inspecting the appearance of a subject using an artificial intelligence (AI)-based machine vision algorithm according to an embodiment of the present invention includes a transfer device 500, a storage device 600, a display device 700, and a communication device. The device 800, an external terminal and/or a server 900 may be further included. Meanwhile, since the components shown in FIGS. 1 to 6 are not essential, the system for inspecting the appearance of a subject using an artificial intelligence (AI)-based machine vision algorithm according to an embodiment of the present invention includes more components than those shown in FIGS. may have or may have fewer components.

Hereinafter, the components of the system for inspecting the appearance of a subject using an artificial intelligence (AI)-based machine vision algorithm according to an embodiment of the present invention will be described in detail.

As shown in FIG. 2, the lighting device 100 is installed on the upper side of the object 1 to be inspected, and the plurality of surfaces (eg, first to fourth surfaces, etc.) and the upper portion of the object 1 to be inspected. A first lighting module 110 for irradiating light to the first surface (eg, left side) and upper surface of the object to be inspected (1), which performs a function of irradiating light to the surface, and the object to be inspected (1) ) The second illumination module 120 for radiating light to the second surface (eg, the right side) and the upper surface, and the third surface (eg, the front surface) and the upper surface of the object to be inspected (1) to radiate light A third lighting module 130 and a fourth lighting module 140 for radiating light to a fourth surface (eg, a back surface) and an upper surface of the subject 1 are provided.

In addition, although the first to fourth lighting modules 110 to 140 of the lighting device 100 are not shown in the drawing, the four side surfaces and It is preferable that the upper surface be irradiated with light at a predetermined distance from each side of the inspected object 1 and installed inclined at a predetermined angle.

In addition, each of the first to fourth lighting modules 110 to 140 of the lighting device 100 includes a plurality of LED light emitting channels CH1 to CHn connected in parallel to each other, as shown in FIG. Each of the plurality of LED light-emitting channels CH1 to CHn includes at least one LED light-emitting element LED1 to LEDn.

In addition, when there are two or more LED light emitting elements (LED1 to LEDn) provided in each LED light emitting channel (CH1 to CHn) forming the first to fourth lighting modules (110 to 140) of the lighting device 100, each The LED light emitting devices (LED1 to LEDn) may be connected in series and/or in parallel in an array form.

That is, the first to fourth lighting modules 110 to 140 of the lighting device 100 receive the operating voltage output from the high-speed switching control module 380 of the machine vision control device 300, and each LED light emitting channel (CH1) to CHn) of each LED light emitting element (LED1 to LEDn) can be operated. In this case, the operating voltage may be applied to the anode of the first LED light emitting element LED1, and the cathode of the first LED light emitting element LED1 may be connected to the anode of the second LED light emitting element LED2.

On the other hand, the object to be inspected 1 is preferably made of a polyhedron-shaped article having a plurality of faces. For example, the test subject 1 in one embodiment of the present invention is applied as a battery (eg, secondary battery) in the form of a flat rectangular parallelepiped, but is not limited thereto, and has a plurality of surfaces (eg, the left side, Any object (or article) in the form of a polyhedron having right side, front side, back side, plane, bottom side, etc.) is applicable.

It is preferable that the first to fourth lighting modules 110 to 140 of the lighting device 100 have a plurality of LEDs (Light Emitting Diodes) in an array form so that surface light can be irradiated to the subject 1, It is not limited thereto, and may include, for example, at least one of a fluorescent lamp, an incandescent lamp, a halogen lamp, a mercury lamp, a neon tube lamp, a sodium lamp, a metal halide lamp, or an EL (Electro-Luminescent) lamp.

In addition, the first to fourth lighting modules 110 to 140 of the lighting device 100 are, for example, liquid crystal displays (LCDs), light emitting diode displays (Light Emitting Diodes, LEDs), thin film transistor liquid crystal displays (Thin Film Transistor-Liquid Crystal Display (TFT LCD), Organic Light Emitting Diode (OLED), Flexible Display, Plasma Display Panel (PDP), Surface Alternate Lighting (ALiS), Digital Light source processing (DLP), silicon liquid crystal (LCoS), surface conduction electron emission display (SED), field emission display (FED), laser TV (quantum dot laser, liquid crystal laser), optoelectric liquid display (FLD), interferometer including at least one of a modulator display (iMoD), a thick film dielectric electric (TDEL), a quantum dot display (QD-LED), a telescopic pixel display (TPD), an organic light emitting transistor (OLET), and/or a laser fluorescence display (LPD) It may be made by, but is not limited thereto, and may include anything as long as it is possible to irradiate surface light to the object to be inspected (1).

As shown in FIG. 2, the linescan camera device 200 is installed at a predetermined distance perpendicular to the direction of travel of the subject 1, and the first to fourth lighting modules of the lighting device 100 ( 110 to 140) in a state where each light output from the object to be inspected 1 is sequentially irradiated, the image of the upper surface of the object to be inspected 1 is scanned in units of lines to obtain a two-dimensional image. Performs the function of acquiring images.

That is, in the line scan camera device 200, a corresponding object to be inspected is in a state in which one of the plurality of first to fourth lighting modules 110 to 140 provided in the lighting device 100 irradiates light in a corresponding order. The imaging surface of (1) is photographed, and the line scan image of the corresponding inspected object 1 is transmitted to the machine vision control device 300.

The line scan camera device 200 does not acquire images in units of areas, but through a single line, that is, a one-dimensional linear structure CMOS (Complementary Metal Oxide Semiconductor) or CCD (Charge Coupled Device) image sensor. As a camera of a method of obtaining a two-dimensional image by scanning in units of lines, in order to generate an image of the subject 1 by the line scan camera device 200, the line scan camera device 200 or the corresponding subject The body 1 is moved at a constant speed, and the entire image of the subject 1 is generated as an image by combining images taken of a portion of the subject 1 at the moment of movement.

The machine vision control device 300 is a device in charge of overall control of a system for inspecting the appearance of a subject using an artificial intelligence (AI)-based machine vision algorithm according to an embodiment of the present invention, particularly of the lighting device 100. The operation of the first to fourth lighting modules 110 to 140 of the lighting device 100 is performed so that each light output from the first to fourth lighting modules 110 to 140 is sequentially irradiated to the corresponding object 1. In addition to controlling, in a state where each light output from the first to fourth lighting modules 110 to 140 of the lighting device 100 is sequentially irradiated to the corresponding subject 1, the subject 1 is line-by-line. It performs a function of controlling the operation of the line scan camera device 200 to be photographed as .

In addition, the machine vision control device 300 receives the entire image having the two-dimensional image of the corresponding object 1 obtained from the linescan camera device 200 and uses it to detect the corresponding object 1 for each lighting module. It separates the two-dimensional image of and performs a function of generating one combined image by overlapping the two-dimensional images of the object under examination for each of the separated lighting modules.

In addition, the machine vision control device 300 performs a function of performing an external inspection of the subject 1 by using a predetermined artificial intelligence (AI)-based machine learning method based on the generated combined image. do.

In addition, the machine vision control device 300 performs an external inspection of the subject 1 by using machine vision based on a preset rule base based on the generated combined image. After that, based on the result data of the non-defective and/or defective products of the exterior inspection of the subject (1) performed above, a predetermined artificial intelligence (AI)-based machine learning method is used to obtain a good product for the subject (1). And/or iterative learning for each characteristic item of a defective product may perform a function of constructing an artificial intelligence learning model 330 for the corresponding inspected object 1 .

At this time, one combined image generated by the machine vision controller 300 is preferably composed of, for example, a 2.5-dimensional (2.5D) image. In this specification, a 2.5-dimensional (2.5D) image represents a two-dimensional image representing a three-dimensional effect based on a cognitive illusion. Cognitive illusions are optical illusions that occur in the cognitive process based on unconscious reasoning and appear because human cognition tends to comprehensively judge the whole thing.

In addition, the rule base-based machine vision algorithm set in the machine vision control device 300 is, for example, pattern matching, point/line/plane fitting, edge, color, and / or location (Location) algorithm is preferably made of at least one algorithm.

In addition, it is preferable that the external inspection of the subject 1 performed by the machine vision control device 300 is, for example, a good product and/or a defective product. It is preferable to classify and determine by at least one surface state among the marking states.

In addition, the machine vision control device 300 includes the entire image having the two-dimensional image of the subject 1 obtained from the linescan camera device 200 and two images of the subject 1 separated for each lighting module. From the dimensional image, the generated combined image, the external inspection result information of the subject 1, the artificial intelligence learning model built from the artificial intelligence learning module 330, and/or the subject appearance determination module 340 2 of the corresponding inspected subject (1) separated for each lighting module, including the entire image having a two-dimensional image for each inspected subject, using at least one of the result information of good and / or defective products of the determined corresponding inspected subject (1) Among the dimensional image, the generated combined image, the external inspection result information of the subject 1, the built artificial intelligence learning model, and/or the good and/or defective product determination result information of the subject 1 A function of controlling at least one piece of information to be converted into a database (DB) and stored in the storage device 600 may be performed.

In addition, the machine vision control device 300 includes the entire image having the two-dimensional image of the subject 1 obtained from the linescan camera device 200 and two images of the subject 1 separated for each lighting module. The dimensional image, the generated combined image, and/or the external inspection result information of the subject 1, the artificial intelligence learning model built from the artificial intelligence learning module 330, and/or the subject appearance determination module ( It may perform a function of controlling the operation of the display module 700 to display at least one of result information of good and/or defective products of the subject 1 determined from step 340 on the display screen.

In addition, the machine vision control device 300 includes the entire image having the two-dimensional image of the subject 1 obtained from the linescan camera device 200 and two images of the subject 1 separated for each lighting module. From the dimensional image, the generated combined image, the external inspection result information of the subject 1, the artificial intelligence learning model built from the artificial intelligence learning module 330, and/or the subject appearance determination module 340 At least one of the result information of the determined good product and/or defective product of the subject 1 is transmitted to an external terminal and/or server 900 through a wired and/or wireless communication method through the communication network 10. It can perform a function of controlling the operation of the communication module 800 to be transmitted.

Various embodiments described herein may be embodied in a recording medium readable by a computer or similar device using, for example, software, hardware, or a combination thereof.

According to the hardware implementation, the embodiments described herein include application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), and field programmable gate arrays (FPGAs). , processors, controllers, micro-controllers, microprocessors, and electrical units for performing functions. In some cases, such embodiments may be implemented by the machine vision controller 300 .

According to software implementation, embodiments such as procedures or functions may be implemented together with a separate software module to perform at least one function or operation. The software code may be implemented by a software application written in any suitable programming language. Also, the software code may be stored in the storage device 600 and executed by the machine vision controller 300 .

As shown in FIG. 4, the machine vision control device 300 includes an image separation module 310, an image recombination module 320, an artificial intelligence learning module 330, an object shape determination module 340, and It consists of a control module 350 and the like. In addition, the machine vision control device 300 applied to an embodiment of the present invention includes a user setting module 360, an output level control module 370, a high-speed switching control module 380, and/or a storage module 390, etc. may further include. Meanwhile, since the components shown in FIG. 4 are not essential, the machine vision control device 300 applied to an embodiment of the present invention may have more or fewer components.

Hereinafter, components of the machine vision control device 300 applied to an embodiment of the present invention will be described in detail.

The image separation module 310 receives the entire image having the 2D image of the subject 1 obtained from the line scan camera device 200 and uses it to obtain a 2D image of the subject 1 for each lighting module. It performs the function of separating images (see Fig. 5).

The image recombination module 320 performs a function of generating one combined image by overlapping the two-dimensional images of the subject 1 for each lighting module separated from the image separation module 310 into one (see FIG. 6). .

The artificial intelligence learning module 330 obtains a good product for the subject 1 through a predetermined artificial intelligence (AI)-based machine learning method based on one combined image generated by the image recombination module 320 and/or It performs the function of constructing an artificial intelligence learning model for the external inspection of the subject (1) by repeatedly learning each characteristic item of a defective product.

In addition, the artificial intelligence (AI)-based machine learning method applied to the artificial intelligence learning module 330 is, for example, a neural network, a support vector machine (SVM), a multi-layer perception (MLP), and / or deep learning ( Deep Learning) is preferably made with at least one artificial intelligence learning method.

The inspection subject appearance determination module 340 uses an artificial intelligence learning model for the external inspection of the inspection subject 1 built from the artificial intelligence learning module 330 to determine whether the inspection subject 1 is a good product and/or a defective product. Inferring and analyzing the surface characteristics of the subject (1) for each feature item, and based on the inferred and analyzed surface characteristic information of the subject (1), good quality and / or scratches for the subject (1), It performs a function of automatically determining a defective product by classifying it into at least one surface state among scratch, foreign matter, and/or marking state.

In addition, the control module 350 is a module in charge of overall control of the machine vision controller 300 applied to an embodiment of the present invention, and in particular, the first to fourth lighting modules 110 to 140 of the lighting device 100 Performs a function of controlling the operation of the first to fourth lighting modules 110 to 140 of the lighting device 100 so that each light output from ) is sequentially irradiated to the corresponding subject 1 .

In addition, the control module 350 controls the target object 1 in a state where each light output from the first to fourth lighting modules 110 to 140 of the lighting device 100 is sequentially irradiated to the corresponding target object 1. ) performs a function of controlling the operation of the line scan camera device 200 to be photographed in units of lines.

In addition, the control module 350 may perform all control functions of the machine vision controller 300 applied to the above-described embodiment of the present invention.

Various embodiments described herein may be embodied in a recording medium readable by a computer or similar device using, for example, software, hardware, or a combination thereof.

According to the hardware implementation, the embodiments described herein include application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), and field programmable gate arrays (FPGAs). , processors, controllers, micro-controllers, microprocessors, and electrical units for performing functions. In some cases, such embodiments may be implemented by control module 350 .

According to software implementation, embodiments such as procedures or functions may be implemented together with a separate software module to perform at least one function or operation. The software code may be implemented by a software application written in any suitable programming language. Also, the software code may be stored in the storage module 390 and executed by the control module 350 .

Additionally, the user setting module 360 is connected to the control module 350 and operates under the control of the control module 350, and each LED light emitting channel (CH1 to CH1 to CHn) output level target value setting, output channel selection and order setting of each LED light emitting channel (CH1 to CHn) for high-speed switching control, and clock generation for generating high-speed switching operation signals. perform the function of

In addition, the user setting module 360 preferably outputs a specific input signal of the corresponding user according to the user's request based on a graphical user interface (GUI) through a separate display screen and switch input, for example. do.

At this time, it is preferable that the user's specific input signal output according to the user's request is provided through an external terminal and/or the server 900 .

The output level control module 370 is connected to the control module 350, and according to the control of the control module 350, output level target values of each LED light emitting channel (CH1 to CHn) set from the user setting module 360 are set. Based on this, in order to adjust the brightness of each LED light emitting element (LED1 to LEDn) provided in each LED light emitting channel (CH1 to CHn), a target value for the output voltage of each LED light emitting channel (CH1 to CHn) is generated and output. perform the function of

The high-speed switching control module 380 is connected to the control module 350, and according to the control of the control module 350, the output channel selection and order of each LED light emitting channel (CH1 to CHn) set from the user setting module 360 The first to fourth lighting modules ( 110 to 140) performs a function of controlling high-speed switching of each LED light emitting channel (CH1 to CHn) provided.

In addition, it is preferable that the high-speed switching control module 380 performs high-speed control by using a hardware programming-based Field Programmable Gate Array (FPGA) for the high-speed switching function.

In addition, the high-speed switching control module 380 continuously turns on each LED light emitting element (LED1 to LEDn) provided in each LED light emitting channel (CH1 to CHn) of the first to fourth lighting modules 110 to 140. In order to protect each LED light emitting element (LED1 to LEDn) from being damaged, a preset operation is performed when a high-speed switching operation signal for each LED light emitting channel (CH1 to CHn) of the first to fourth lighting modules (110 to 140) is changed. It is possible to perform a function of controlling each LED light emitting channel (CH1 to CHn) of the first to fourth lighting modules (110 to 140) to perform a high-speed switching operation only for a certain operation standard time less than the standard time (preferably, about 1 ms). there is.

That is, the high-speed switching control module 380 applies a plurality of semiconductor elements (eg, buffers, resistors, capacitors, etc.) for a certain time of about 1 ms or less at the moment when the operation signal is changed. It may be provided so as to operate only on the user, so as to protect the damage of the lighting due to the continuous turning on of the lighting due to the user's negligence. At this time, among the plurality of semiconductor elements, the capacitor preferably serves to prevent the input of the operating signal from being maintained, and the resistor serves to maintain the operating signal in a low state. Preferably, the maximum pulse input time of the hardware can be determined by adjusting the values of the capacitor and the resistor.

In addition, the high-speed switching control module 380 detects an idle period of a photographing operation for the machine vision camera device 200 installed near the lighting device 100, so that a photographing operation signal of the corresponding machine vision camera device 200 is generated. Determining the idle state when there is no input for a certain period of time, that is, a preset shooting operation reference time, and automatically resetting the state during the shooting operation to be reset to the initial input state so that the initial state is maintained after the determined idle state ( Reset) function.

That is, by applying the first to fourth LED light emitting channels (CH1 to CH4) of the first to fourth lighting modules (110 to 140) and the machine vision camera device 200, which are four-sided lighting, the object being inspected is transferred. In the case of a machine vision system for photographing a corpse 1, as shown in FIG. 2, when the subject 1 moving while controlling lighting is sequentially photographed and line images taken when the light of the same position is turned on are collected , the effect of obtaining an image obtained by photographing four sides in a stationary state is obtained.

In this type of image capturing, the object to be imaged can be imaged during transport without stopping, so that the production yield can be improved significantly because image capturing and reading can be performed during transport without additional processes at the production site.

The above-described high-speed switching control module 380 pulses according to the output voltage target value of each LED light emitting channel CH1 to CHn of the first to fourth lighting modules 110 to 140 output from the output level control module 370. A pulse width modulation (PWM) control signal is generated, and a plurality of driving switches (eg, high-current semiconductors such as FET or IGBT) are turned on/off (ON) according to the generated pulse width modulation (PWM) control signal /OFF) By outputting a high-speed switching control signal for controlling the high-speed switching operation, the delay in response speed is minimized and provided in each LED light emitting channel (CH1 to CHn) of the first to fourth lighting modules (110 to 140). It can perform a function of high-speed control of the output current supplied to each of the LED light emitting elements (LED1 to LEDn).

At this time, in relation to high-speed control, in an embodiment of the present invention, an example of a machine vision system that uses a four-sided lighting device 100 and applies a line scan camera device 200 to photograph an object being transported , but this is just one application case for better understanding, and the range of high-speed image capture can be freely set by software settings without being limited by the type of lighting or the number of channels.

On the other hand, in the case of each LED light emitting element (LED1 to LEDn) provided in each LED light emitting channel (CH1 to CHn) of the first to fourth lighting modules 110 to 140, high power in the form of a pulse in a short time It is possible to flow a lot of current by applying a lot of current, and when a lot of current flows instantaneously, the brightness of each LED light emitting element (LED1 to LEDn) also has the effect of increasing as much as the current flowing rate.

In addition, as a method for supplying high current to each LED light emitting element (LED1 to LEDn), a capacitor for charging is provided on the output side, and the output voltage of the DC-DC converter is changed by setting control of the user setting module 360. A high current circuit may be configured by configuring a high voltage circuit capable of supplying sufficient current to each of the LED light emitting elements (LED1 to LEDn). This method can be used as a function of supplying sufficient current to the output side or adjusting brightness as needed.

And, the storage module 390 is connected to the control module 350, and according to the control of the control module 350, for the output level target of each LED light emitting channel (CH1 to CHn) set from the user setting module 360 It performs a function of storing at least one set value among set values, set values for output channel selection and order of each LED light emitting channel (CH1 to CHn), and/or set values for clock generation.

Such a storage module 390 may be, for example, a flash memory type, a hard disk type, a multimedia card micro type, or a card type memory (eg SD or XD memory). etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), A readable storage medium of at least one of a magnetic memory, a magnetic disk, and an optical disk may be included.

In addition, the power supply device 400 includes each of the aforementioned devices, that is, the lighting device 100, the line scan camera device 200, the machine vision control device 300, the inspection object transfer device 500, and the storage device 600. ), display device 700, and / or communication device 800 perform a function of supplying power necessary for operation, etc., commercial alternating current (AC) power (eg, AC 220V or 380V) for continuous power supply etc.) is preferably implemented to be converted into direct current (DC) and/or alternating current (AC) power, but is not limited thereto, and may be implemented by including a conventional portable battery.

In addition, the power supply device 400 may include a power management unit (not shown) that protects components from external power shock and outputs a constant voltage. The power management unit may include an electro static damage (ESD) protector, a power detector, a rectifier, and a power circuit breaker.

Here, the ESD protector is configured to protect electric components from static electricity or sudden power shock. The power detector is configured to send a cut-off signal to the power circuit breaker when a voltage outside the allowable voltage range is introduced, and transmit a step-up or step-down signal to the rectifier according to a voltage change within the allowable voltage range. The rectifier is configured to perform a step-up or step-down rectification operation according to the signal of the power detector so that a constant voltage is supplied by minimizing fluctuations in the input voltage. The power breaker is configured to cut off power supplied from a battery according to a cutoff signal transmitted from the power detector.

Additionally, the inspection object transfer device 500 is connected to the machine vision control device 300, and performs a function of seating and transporting the corresponding object 1 in one direction under the control of the machine vision control device 300. do. Since such an inspection object transport device 500 can be applied by selecting and applying a two-dimensional transport device of various structures, including a conveyor roller or a conveyor belt widely used in the industrial field, the present invention is specific. A description of the structure and its action will be omitted.

The storage device 600 is connected to the machine vision control device 300, and stores the two-dimensional image of the subject 1 obtained from the linescan camera device 200 under the control of the machine vision control device 300. The two-dimensional image of the subject (1) separated by each lighting module, the generated combined image, the external inspection result information of the subject (1), and the artificial intelligence learning module (330) It performs a function of storing at least one information of a good product and/or a defective product result information of the object 1 determined from the built artificial intelligence learning model and/or the object shape determination module 340 .

In addition, the storage device 600 may perform a function of storing and maintaining at least one program code executed through the machine vision control device 300 and at least one data set used by the program code.

The storage device 600 may be, for example, a flash memory type, a hard disk type, a multimedia card micro type, or a card type memory (eg SD or XD memory). etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), It may include at least one type of storage medium among a magnetic memory, a magnetic disk, and an optical disk.

The display module 700 is connected to the machine vision control device 300, and displays a two-dimensional image of the subject 1 obtained from the linescan camera device 200 under the control of the machine vision control device 300. The two-dimensional image of the subject (1) separated by each lighting module, the generated combined image, the external inspection result information of the subject (1), and the artificial intelligence learning module (330) Performs a function of displaying on the display screen at least one of the result information of the good product and/or the defective product of the subject 1 determined from the built artificial intelligence learning model and/or the subject shape determination module 340 .

Such a display device 700 includes, for example, a liquid crystal display (LCD), a light emitting diode display (LED), a thin film transistor-liquid crystal display (TFT LCD), an organic light emitting diode (Organic Light Emitting Diode, OLED), Flexible Display, Plasma Display Panel (PDP), Surface Alternate Lighting (ALiS), Digital Light Source Processing (DLP), Liquid Crystal in Silicon (LCoS), Surface Conduction type electron emission display (SED), field emission display (FED), laser TV (quantum dot laser, liquid crystal laser), optoelectric liquid display (FLD), interferometric modulator display (iMoD), thick film dielectric electric (TDEL), It may include at least one of a quantum dot display (QD-LED), a telescopic pixel display (TPD), an organic light emitting transistor (OLET), a laser fluorescence display (LPD), and a 3D display, but is not limited thereto. As long as it can display at least one of external image, text and/or document data or display external image information of the subject 1, it can include anything.

The communication device 800 is connected to the machine vision control device 300, and transmits a two-dimensional image of the subject 1 obtained from the linescan camera device 200 under the control of the machine vision control device 300. The two-dimensional image of the subject (1) separated by each lighting module, the generated combined image, the external inspection result information of the subject (1), and the artificial intelligence learning module (330) At least one of the result information of good and/or defective products of the subject 1 determined from the built artificial intelligence learning model and/or the subject appearance determination module 340 is transmitted through a wired and/or communication network 10. Alternatively, it performs a function of transmitting to an external terminal and/or server 900 in a wireless communication method.

In addition, the external terminal and/or the server 900 may perform a function of transmitting a specific input signal of the corresponding user to the user setting module 360 according to the request of the corresponding user.

That is, the external terminal and/or server 900 provides the user's information for user setting of high-speed switching LED lighting control according to the user's request through a pre-installed application related to user setting for controlling high-speed switching LED lighting. A specific input signal may be transmitted to the user setting module 360 wired and/or wirelessly through the communication network 10 .

In addition, the external terminal and/or the server 900 includes the entire image having the two-dimensional image of the subject 1 transmitted from the communication device 800 through a pre-installed application related to the external appearance inspection of the subject 1 and each lighting module. A two-dimensional image of the subject (1) separated by each other, the generated combined image, the external inspection result information of the subject (1), an artificial intelligence learning model built from the artificial intelligence learning module 330, and / or Each lighting module including the entire image having a two-dimensional image for each subject using at least one of the result information of good and / or defective products of the subject 1 determined from the subject shape determination module 340 The two-dimensional image of the subject (1) separated by each other, the generated combined image, the external inspection result information of the subject (1), the built artificial intelligence learning model, and / or the subject (1) ) performs a function of providing a service so that at least one information of good and/or defective product determination result information is displayed on the display screen.

At this time, the communication network 10 is a communication network that is a high-speed backbone network of a large-scale communication network capable of providing large-capacity, long-distance voice and data services, and provides Internet or high-speed multimedia services through Wi-Fi, WiGig, It may be a next-generation wireless communication network including Wibro (Wireless Broadband Internet), World Interoperability for Microwave Access (Wimax), and the like.

The Internet is a TCP / IP protocol and various services existing in the upper layer, that is, HTTP (Hyper Text Transfer Protocol), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), It means a worldwide open computer network structure that provides SNMP (Simple Network Management Protocol), NFS (Network File Service), NIS (Network Information Service), etc., and an external terminal and / or server 900 is a user setting module ( 360) and/or the communication device 800 are provided. Meanwhile, the Internet may be wired or wireless Internet, or may be a core network integrated with a wired public network, a wireless mobile communication network, or a portable Internet.

If the communication network 10 is a mobile communication network, it may be a synchronous mobile communication network or an asynchronous mobile communication network. As an example of the asynchronous mobile communication network, a wideband code division multiple access (WCDMA) communication network may be used. In this case, although not shown in the drawings, the mobile communication network may include, for example, a Radio Network Controller (RNC) and the like. Meanwhile, although the WCDMA network is taken as an example, it may be a next-generation communication network such as a cellular-based 3G network, an LTE network, a 4G network, and a 5G network, and other IP-based IP networks. The communication network 10 serves to mutually transfer signals and data between the external terminal and/or server 900 and the user setting module 360 and/or the communication device 800.

On the other hand, the communication network 10 uses, for example, Bluetooth, ZigBee, Beacon, UWB (Ultra Wideband), RFID (Radio Frequency Identification), and/or infrared (IR) communication. A communication method may also be used.

The external terminal and/or server 900 is composed of at least one mobile terminal device among a smart phone, a smart pad, or a smart note that communicates through wireless Internet or mobile Internet. In addition, personal PCs, notebook PCs, Palm PCs, mobile game machines (Mobile play-stations), DMB (Digital Multimedia Broadcasting) phones with communication functions, tablet PCs, iPads, etc. ), a user setting module 360 and/or a user interface for accessing the communication device 800, which may comprehensively mean all wired and wireless home appliances/communication devices.

Although a preferred embodiment of the system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm according to the present invention has been described above, the present invention is not limited thereto, and the claims and detailed description of the invention and the appended It is possible to carry out various modifications within the scope of the drawings, and this also belongs to the present invention.

100: lighting device,
200: line scan camera device,
300: machine vision control device,
400: power supply,
500: test subject transfer device,
600: storage device,
700: display device,
800: communication device,
900: External terminal or server

Claims (21)

A first lighting module for radiating light to the first surface and upper surface of the subject, a second lighting module for radiating light to the second surface and upper surface of the subject, and a third surface and upper surface of the subject to be inspected. A lighting device having a third lighting module for radiating light, and a fourth lighting module for radiating light to a fourth surface and an upper surface of the object to be inspected;
It is installed at a certain distance perpendicular to the traveling direction of the subject, and each light output from the first to fourth lighting modules of the lighting device is sequentially irradiated to the subject, a line scan camera device for obtaining a two-dimensional image by scanning an upper surface image; and
Controlling the operation of the first to fourth lighting modules so that the respective lights output from the first to fourth lighting modules of the lighting device are sequentially irradiated to the target object, and the first to fourth lights of the lighting device In a state in which each light output from the module is sequentially irradiated to the corresponding target object, the operation of the line scan camera device is controlled so that the corresponding target object is photographed in line units, and 2 An entire image with a dimensional image is provided, and using this, a 2D image of the object under test is separated for each lighting module, and a combined image is obtained by overlapping the 2D images of the object under test for each of the separated lighting modules. After generating, an artificial intelligence-based machine including a machine vision control device that performs an external inspection of the object to be inspected using a predetermined artificial intelligence (AI)-based machine learning method based on the generated single combined image. A system for inspecting the appearance of a subject using a vision algorithm.
According to claim 1,
The inspection subject's external appearance inspection system using an artificial intelligence-based machine vision algorithm, characterized in that the inspection subject is made of a polyhedron-shaped article having a plurality of faces.
According to claim 1,
The external inspection system of the inspected subject using an artificial intelligence-based machine vision algorithm, characterized in that further comprising a subject transport device for seating the inspected object and transporting it in one direction.
According to claim 1,
One combined image generated by the machine vision control device is a 2.5-dimensional (2.5D) image, characterized in that the appearance inspection system of the subject using an artificial intelligence-based machine vision algorithm.
According to claim 1,
The external appearance inspection of the inspected object performed by the machine vision control device is determined as a good product and a defective product, and the defective product is classified into at least one surface state of scratch, dent, foreign matter, and marking state. Characterized in that A system for inspecting the appearance of objects using artificial intelligence-based machine vision algorithms.
According to claim 1,
The machine vision control device performs an external inspection of the subject using a pre-set rule-based machine vision algorithm based on the generated one combined image, and then performs an external inspection of the subject. Based on the good/defective product result data for the inspected object, it is repeated learning for each non-defective/defective product characteristic item through a preset artificial intelligence (AI)-based machine learning method to build an artificial intelligence learning model for the inspected object. A system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm.
According to claim 6,
The preset rule base-based machine vision algorithm is at least one of pattern matching, point/line/plane fitting, edge, color, and location algorithms Appearance inspection system of the subject using an artificial intelligence-based machine vision algorithm, characterized in that consisting of.
According to claim 1,
The machine vision control device,
An image separation module for receiving an entire image having a two-dimensional image of the object to be inspected obtained from the line scan camera device and separating the two-dimensional image of the object to be inspected for each lighting module by using the received image;
an image recombination module for generating a single combined image by overlapping two-dimensional images of a corresponding object under examination for each lighting module separated from the image separation module;
Based on one combined image generated from the image recombination module, through a predetermined artificial intelligence (AI)-based machine learning method, iteratively learns each characteristic item of a good/defective product for the subject to inspect the appearance of the subject An artificial intelligence learning module that builds an artificial intelligence learning model;
Using the artificial intelligence learning model for the external inspection of the inspected object built from the artificial intelligence learning module, the surface characteristics of the inspected object are inferred and analyzed for each good/defective product characteristic item, and the inference and analysis Based on the surface characteristic information of the inspected object, the inspected object appearance judgment module automatically determines the inspected object as a defective product by classifying the inspected object into at least one surface state of a good product or scratch, scratch, foreign matter, and marking state; and
The operation of the first to fourth lighting modules of the lighting device is controlled so that each light output from the first to fourth lighting modules of the lighting device is sequentially irradiated to the target object, and the first to fourth lighting modules of the lighting device are sequentially irradiated. Artificial intelligence comprising a control module for controlling the operation of the line scan camera device so that each light output from the fourth lighting module is sequentially irradiated to the corresponding target object so that the target object is photographed in line units. A system for inspecting the appearance of a subject using an intelligence-based machine vision algorithm.
According to claim 8,
The artificial intelligence (AI)-based machine learning method applied to the artificial intelligence learning module is at least one of a neural network, a support vector machine (SVM), a multi-layer perception (MLP), and deep learning. An external inspection system for an object to be inspected using an artificial intelligence-based machine vision algorithm, characterized in that it consists of an artificial intelligence learning method.
According to claim 8,
Each of the first to fourth lighting modules of the lighting device includes a plurality of LED light emitting channels connected in parallel to each other including at least one LED light emitting element,
The machine vision control device sets target output levels of each LED light emitting channel provided in the first to fourth lighting modules of the lighting device according to a user's specific input signal output according to the user's request, and controls high-speed switching. A user setting module for setting output channel selection and order setting of each LED light emitting channel, and clock generation for generating high-speed switching operation signals;
Based on the received output level target value of each LED light emitting channel set from the user setting module, an output voltage target value of each LED light emitting channel is generated in order to adjust the brightness of each LED light emitting element provided in each LED light emitting channel. an output level control module that outputs; and
Control of the lighting device according to the output voltage target value of each LED light emitting channel output from the output level control module in synchronization with the setting information and clock information of the output channel selection and order of each LED light emitting channel set from the user setting module. A high-speed switching control module for controlling high-speed switching of each LED light emitting channel provided in the first to fourth lighting modules is further included.
According to claim 10,
The user setting module outputs a specific input signal of the user according to the user's request based on a graphical user interface (GUI). inspection system.
According to claim 10,
The high-speed switching control module performs high-speed control using a hardware programming-based FPGA (Field Programmable Gate Array) for a high-speed switching function.
According to claim 10,
The high-speed switching control module detects an idle period of a photographing operation for a machine vision camera device installed near the lighting device, and when a photographing operation signal of the machine vision camera device is not input for a predetermined photographing operation reference time or longer. An artificial intelligence-based machine characterized in that it performs an automatic reset function that determines the idle state, removes the state during the shooting operation and initializes it to the initial input state so that it is maintained in the first state after the determined idle state. A system for inspecting the appearance of a subject using a vision algorithm.
According to claim 10,
At least one of the setting value for the output level target of each LED light emitting channel set from the user setting module, the setting value of the output channel selection and order of each LED light emitting channel, and the setting value for clock generation A system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm, characterized in that it further comprises a storage module for storing.
According to claim 10,
A system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm, characterized in that the user's specific input signal output according to the user's request is provided through an external terminal or server.
According to claim 15,
The external terminal or server transmits the user's specific input signal for user setting of fast switching LED lighting control wired or wirelessly according to the user's request through a pre-installed user setting related application for fast switching LED lighting control. An external inspection system of an object to be inspected using an artificial intelligence-based machine vision algorithm, characterized in that transmitted to the user setting module.
According to claim 10,
The high-speed switching control module, in order to protect each LED light emitting element provided in each LED light emitting channel of the first to fourth lighting modules from being damaged due to the continuous turning on of each LED light emitting element, the first to fourth light emitting elements When the high-speed switching operation signal for each LED light emitting channel of the lighting module is changed, each LED light emitting channel of the first to fourth lighting modules is controlled to operate at high speed only for a predetermined operation standard time or less A system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm.
According to claim 1,
The whole image with the 2D image of the subject obtained from the line scan camera device, the 2D image of the subject separated by each lighting module, the generated combined image, and the external inspection of the subject A storage device for storing result information is further included,
The machine vision control device includes an entire image having a two-dimensional image of the subject obtained from the line scan camera device, a two-dimensional image of the subject separated for each lighting module, the generated combined image, And a two-dimensional image of the subject separated by each lighting module, including the entire image having a two-dimensional image for each subject using the external inspection result information of the subject, the generated combined image, and the subject The appearance inspection system of the subject using an artificial intelligence-based machine vision algorithm, characterized in that for controlling the appearance inspection result information of the database (DB) to be stored in the storage device.
According to claim 1,
The whole image with the 2D image of the subject obtained from the line scan camera device, the 2D image of the subject separated by each lighting module, the generated combined image, and the external inspection of the subject A display device for displaying result information on a display screen is further included,
The machine vision control device includes an entire image having a two-dimensional image of the subject obtained from the line scan camera device, a two-dimensional image of the subject separated for each lighting module, the generated combined image, And a two-dimensional image of the subject separated by each lighting module, including the entire image having a two-dimensional image for each subject using the external inspection result information of the subject, the generated combined image, and the subject An appearance inspection system of an object to be inspected using an artificial intelligence-based machine vision algorithm, characterized in that for controlling the operation of the display device to display the appearance inspection result information on the display screen.
According to claim 1,
The whole image with the 2D image of the subject obtained from the line scan camera device, the 2D image of the subject separated by each lighting module, the generated combined image, and the external inspection of the subject A communication device for transmitting the result information in a wired or wireless communication method is further included,
The machine vision control device includes an entire image having a two-dimensional image of the subject obtained from the line scan camera device, a two-dimensional image of the subject separated for each lighting module, the generated combined image, and controlling the operation of the communication device so that the external terminal or server is transmitted to an external terminal or server in a wired or wireless communication method. system.
According to claim 20,
The external terminal or server includes a whole image having a 2D image of the subject transmitted from the communication device through a pre-installed application related to the appearance inspection of the subject, and a 2D image of the subject separated for each lighting module. , A two-dimensional image of the subject separated for each lighting module, including the entire image having a two-dimensional image for each subject using the generated combined image and the external inspection result information of the subject, the generated A system for inspecting the appearance of a subject using an artificial intelligence-based machine vision algorithm, characterized in that it provides a service so that one combined image and information on the appearance inspection result of the subject are searched or displayed on a display screen.

Images