KR20210048459A - Method for non-destructive inspection based on transmission image - Google Patents

Abstract

The present invention relates to a transmission image based non-destructive inspection method, or a method for providing a function thereof. Specifically, when there are transmission images of electronic components or industrial components on which an electronic circuit board is mounted, a user can perform various inspections on these transmitted images. Thus, the method for providing a non-destructive inspection function allows the user to easily inspect defects according to various situations or objects. The present invention includes a step of receiving a user input and setting a template for inspection.

Description

Transmission image-based non-destructive inspection method {METHOD FOR NON-DESTRUCTIVE INSPECTION BASED ON TRANSMISSION IMAGE}

The present invention relates to a transmission image-based non-destructive inspection method, or a method for providing the function thereof. Specifically, when transmission images of an electronic component or industrial component on which an electronic circuit board is mounted exist, the transmission images are targeted. The present invention relates to a method for providing a non-destructive inspection function that enables a user to easily perform defect inspection according to various situations or objects by allowing a user to perform various inspections.

In order to effectively inspect electronic parts or various industrial parts on which electronic circuit boards are mounted, in particular, inspection technologies using transmission images such as X-rays have existed in the prior art to inspect the internal state of the parts. Demand is constantly increasing.

Meanwhile, representative examples of electronic parts or industrial parts include Ball Grid Array (BGA), Land Grid Array (LGA), Quad Flat Package (QFP), and Quad Flat Non-lead Package (QFN). For inspection, a technology capable of automatically recognizing a large number of objects and effectively inspecting internal voids, solder areas, and location-based defects is essential.

On the other hand, the demand to inspect the inside of an object in 3D using an X-ray tomography image is also increasing, but due to various artifacts (ring artifact, beam hardening, scattering, cone beam, cupping artifact, etc.), the actual inspection is inspected. It is limited to the level of performing the inspection by directly setting the inspection area from the cross-sectional image in the direction he wants to examine.

Accordingly, it is currently necessary to develop an inspection method that efficiently sets an inspection area for both inspection using a two-dimensional transmission image and an X-ray tomography image and satisfies all inspection factors required by the industry.

The present invention was derived by focusing on such a problem, and invented to provide additional technical elements that can not be easily invented by those of ordinary skill in the art, as well as being able to solve the technical problems of salpin from the above. Became.

Republic of Korea Patent Publication No. 10-1962076 (2019.03.19.)

The present invention is to provide an environment capable of performing a non-destructive inspection function based on a 2D transmission image or a tomography image of an arbitrary object. The object of the present invention is to enable quick and accurate defect detection compared to the prior art by allowing the method to be selected.

In particular, the present invention enables an inspector to accurately identify defects according to parameters or templates set in advance even for objects that cannot easily identify defects using conventional methods, such as having a complex structure or having an inconsistent thickness. It is aimed at.

Meanwhile, the technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems that are not mentioned will be clearly understood by those skilled in the art from the following description.

In order to solve the above problem, the non-destructive inspection method according to the present invention is characterized in that iteratively performs non-destructive inspection based on target images-the target images are respective transmission images of a plurality of objects. And, the non-destructive inspection method comprises a step of receiving a user input and setting an inspection template, wherein the setting of the inspection template comprises: non-destructive inspection within transmission images of the objects Designating a region of interest to proceed with; And an object area in the ROI-the object area is an area consisting of sets of pixels having similar values within a preset range-or setting a plurality of parameters for identifying a defective area; have.

In addition, the non-destructive testing method according to another embodiment of the present invention is characterized in that sequential non-destructive testing is performed based on target images-the target images are a plurality of tomographic images obtained by tomography of a specific object, The non-destructive inspection method includes: receiving a user input and setting an inspection template, wherein the setting of the inspection template comprises: a region of interest in which the non-destructive inspection is to be performed within the tomographic images of the object. Specifying; And an object area within the ROI-the object area is an area consisting of sets of pixels having similar values within a preset range-or setting a plurality of parameters for identifying a defective area; have.

In addition, the non-destructive inspection method according to another embodiment of the present invention is characterized in that the non-destructive inspection is performed on the basis of a target image-the target image includes a plurality of objects having a similar shape, and the non-destructive inspection method Including a step of receiving a user input and setting an inspection template, wherein the setting of the inspection template includes: designating a region of interest in which a non-destructive inspection is to be performed within the tomographic images of the object; Designating a sub-ROI including at least one object therein as included in the ROI; And setting a plurality of parameters for identifying a defective area or an object area within the sub-ROI, wherein the object area is an area consisting of sets of pixels having similar values within a preset range. I can.

According to the present invention, there is an effect of providing an environment in which an inspector can finely set various parameters and templates for a non-destructive inspection, and also enables a non-destructive inspection to be performed by selecting an appropriate inspection method according to an object or situation.

In addition, according to the present invention, even if an object has a complex structure, it is possible to quickly and accurately detect defects in the presence of a 2D transmission image.

In addition, according to the present invention, there is an effect of being able to accurately identify defects without errors even for tomography images, which have previously been difficult to detect defects.

1 is a schematic diagram of a general process of non-destructive testing and a device required for non-destructive testing.
2 is a sequence diagram illustrating a method of executing the non-destructive testing function provided according to the present invention.
3 shows a display state of the first user interface.
4 is a diagram illustrating a configuration of a masking area on a first user interface.
5 is a diagram illustrating a state in a first user interface that appears when an inspection execution icon is clicked to execute a non-destructive inspection.
6 and 7 illustrate a first user interface displayed when the user moves the pointer and places the pointer at an arbitrary point on the target image.
8 shows an embodiment of a target image on which a non-destructive inspection has been performed according to a first inspection method.
9 shows an embodiment of a target image on which a non-destructive inspection has been performed according to a second inspection method.
10 shows an embodiment of a target image on which a non-destructive inspection has been performed according to a third inspection method.
11 shows a second user interface displayed.

Details of the objects and technical configurations of the present invention, as well as operational effects thereof, will be more clearly understood by the following detailed description based on the accompanying drawings in the specification of the present invention. An embodiment according to the present invention will be described in detail with reference to the accompanying drawings.

The embodiments disclosed herein should not be construed or used as limiting the scope of the present invention. It is natural for those skilled in the art that the description including the embodiments of the present specification has various applications. Accordingly, any of the embodiments described in the detailed description of the present invention are exemplary for better describing the present invention, and it is not intended that the scope of the present invention be limited to the embodiments.

The functional blocks shown in the drawings and described below are only examples of possible implementations. In other implementations, other functional blocks may be used without departing from the spirit and scope of the detailed description. Further, although one or more functional blocks of the present invention are represented as individual blocks, one or more of the functional blocks of the present invention may be a combination of various hardware and software configurations that perform the same function.

In addition, the expression that includes certain elements is an expression of “open type” and simply refers to the existence of the corresponding elements, and should not be understood as excluding additional elements.

Furthermore, when a component is referred to as being “connected” or “connected” to another component, it should be understood that it may be directly connected or connected to the other component, but other components may exist in the middle. do.

Prior to the full description, the transmission image-based non-destructive inspection function will be described with reference to FIGS. 1 and 2 through which systems and processes are provided.

First, FIG. 1 shows schematic configurations for non-destructive inspection based on a transmission image. Non-destructive testing refers to a method of inspecting an internal condition without destroying an object such as a circuit, and the methods include radiographic examination, ultrasonic flaw detection, magnetic flaw detection, and electromagnetic induction testing. In general, since objects such as circuits have uniform quality and are not completely intact throughout, it is necessary to detect defects by inspecting whether the object satisfies the required conditions sufficiently, and the present invention is in the process of finding such defects. For. Referring to FIG. 1, the non-destructive test presupposes analyzing images of a transmission image. Therefore, in order to utilize the present invention, a transmission image of an object is necessarily required. In this case, the object means an object to be identified whether or not there is a defect, and the type of the object may include electronic components or industrial parts, for example, circuit boards, fine pipes, panels, and the like. In addition, the generation of the transmission image may be performed according to various existing methods, such as the aforementioned radiographic examination, ultrasonic examination, magnetic examination, electromagnetic induction examination, etc. However, in this detailed description, the X-ray apparatus is used to aid understanding of the invention. It will be described on the premise that a transmission image of the object has been generated using (50).

After the transmission image(s) is generated, the transmission images are converted into a computer-readable image and then transmitted to the non-destructive testing apparatus 100 according to the present invention. In this case, the image in a computer-readable form will be referred to as target image(s) in this detailed description.

With respect to the non-destructive testing apparatus 100, the configuration is assumed to include a central processing unit (CPU) and a memory when viewed from the side of the apparatus. The central processing unit may also be referred to as a controller, a microcontroller, a microprocessor, and a microcomputer. In addition, the central processing unit may be implemented by hardware, firmware, software, or a combination thereof. When implemented using hardware, an application specific integrated circuit (ASIC) or a digital signal processor (DSP) , DSPD (digital signal processing device), PLD (programmable logic device), FPGA (field programmable gate array), etc., when implemented using firmware or software, modules, procedures or functions that perform the above functions or operations, etc. Firmware or software may be configured to include. In addition, the memory includes ROM (Read Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, SRAM (Static RAM), It may be implemented as a hard disk drive (HDD), a solid state drive (SSD), or the like. In addition, the non-destructive testing apparatus 100 may further include a communication device for transmitting and receiving data to and from an external terminal or an external local server in addition to a central processing unit and a memory.

In addition, the non-destructive testing apparatus 100 is required to have a display. In other words, in order to provide the non-destructive testing function according to the present invention, a means for displaying interfaces to the user is required, and if the interface can be visually displayed to the user regardless of the type of such display, all devices are one type of display. Can be.

On the other hand, when looking at the non-destructive testing device 100 in terms of functionality, the device can perform image processing on a plurality of target image(s) using the hardware described above, and in this process, the user can A user interface may be provided to enable input, that is, a user interface capable of input so that a user can execute an arbitrary function. That is, the non-destructive testing apparatus 100 may provide an object or a test result of a plurality of objects to an inspector by performing various inspection methods on a plurality of target image(s). A process in which the non-destructive testing device 100 provides a specific non-destructive testing function will be described with reference to the accompanying drawings.

FIG. 2 schematically shows a procedure in which the non-destructive test is performed, assuming that the non-destructive test function provided according to the present invention is executed by the user.

Referring to FIG. 2, the first step performed is a step (S101) of photographing a transmission image of an object, and this step may be performed by the X-ray apparatus 50. The x-ray apparatus 50 may be connected to the above-described non-destructive testing apparatus 100 through a network, and may perform X-ray imaging of a specific object according to an input of an inspector who operates the non-destructive testing apparatus 100. On the other hand, after the imaging is completed by the x-ray apparatus 50, a transmission image of the object may be generated, which is converted to the format of the target image(s) to the non-destructive inspection apparatus 100 and then transmitted to perform the next step. You can proceed. In this case, the transmitted images are already computer-readable by themselves, and thus may not need to be converted into separate target image(s), and in this case, the transmitted image may be treated as the same as the target image.

After the transmission image capturing step (S101), on the non-destructive testing apparatus 100, the test template may be set by receiving an input from the tester. (S102) The inspection template may be understood as a set of multiple parameter values input by an inspector in order to effectively identify a defect in an arbitrary target image. The inspection template may be set based on parameter values input by the inspector through input devices (keyboard, mouse, etc.) connected to the non-destructive inspection apparatus 100, and the types of these parameters include inspection tools, inspection purposes, and interest. Region position, region of interest shape, region of interest size, pre-processing method, extraction method, defect determination criteria, etc. can be included, and furthermore, alignment and correction of the target image, position of the target image, rotation, magnification, and actual size conversion per pixel Ratio, etc. may be further included. Meanwhile, the step S102 may include processes in which a user inputs various parameters through a first user interface that will be described later in this detailed description. The first user interface will be described in more detail from the following drawings.

Returning to FIG. 2 again, after step S102, the non-destructive inspection apparatus 100 loads the target image(s) for the object (S103) to prepare for full-fledged defect identification, and after the target image(s) is loaded, the non-destructive test Execution and result report generation (S104) is performed, and finally, the generated result report is output on the screen so that the inspector can see it. (S105)

Hereinafter, a first user interface will be described in detail among the methods for providing a non-destructive inspection function according to the present invention with reference to FIGS. 3 to 7. For reference, the first user interface includes menus for newly creating a template for inspection in step S102 as seen in FIG. 2 above.

The method of providing a non-destructive inspection function according to the present invention is basically based on a state in which the first user interface is displayed on a non-destructive inspection device having a display, and the inspection template creation menu is selected by a user who wants to create an inspection template. When it is, displaying a first user interface including a plurality of icons, receiving a user input while the first user interface is displayed, and generating and setting a test template according to the received user input. Includes.

3 shows the appearance of the first user interface, in which the main area 100 in which images and markers required for editing the inspection template are displayed, and the user creates the inspection template It contains a number of icons necessary for it. For reference, it is understood that the arrangement positions of the icons described below, the shape of the icons shown in the drawings, etc. are all described and illustrated as one embodiment to aid understanding of the invention, and these arrangements and shapes may be variously changed. .

Referring to FIG. 3, first, icons for changing the type of the user interface may be displayed on the leftmost side of the first user interface. That is, icons corresponding to the automatic inspection execution mode (ADR), inspection review mode (Review), and inspection template editing mode (Template) may be displayed within the reference numeral 200, and each of the icons can be selected by the user. If there is, it can be displayed on the screen in the form of a different user interface. 3 illustrates a state in which a template icon for inspection is clicked and a first user interface corresponding to the icon is displayed. On the other hand, around the icons in which the selection input was made by the user, a color different from the icon in which there was no selection input are displayed as a background, so that the user can recognize which icon is being selected.

In addition, a setting icon (Setting) may be disposed at the lower left of the first user interface, and when a setting icon is clicked by a user, a menu for language setting and inspection tool setting may be displayed. At this time, the menu for setting the inspection tool may include a menu for changing the color of the dragged line, the thickness of the line, the transparency of the dragged line or the area set by the drag when designating the region of interest.

On the other hand, in the first user interface, functions necessary for the user to create and edit the inspection template may be displayed for each group. For example, inspection templates such as 300 and 350 are newly created. A template tool group including icons for creation and storage, and functions that can set various settings for an arbitrary inspection template such as 400 and 450 are iconized to display a group of inspection tools displayed as a group. . In addition, the first user interface may further include an area in which parameters can be set (parameter input area) as shown in reference numeral 500. Hereinafter, each icon included in the template tool group and the inspection tool group will be described.

First, when looking at the template tool group, the tool group includes a target image import icon (Img.Import) for importing the target image, and a capture icon for capturing the result image of non-destructive testing and saving it in an arbitrary image format. , Result save icon (Save Result) to save the results of non-destructive testing in a file format that can be read in review mode (preferably XML), icon (New) to create and edit a new test template, and have already been saved. An icon (Import) for importing an existing inspection template or a component of the inspection template (the component is preferably created as an XML file), and an image change icon (Img. Change), a simulator icon that checks whether automatic inspection is possible by simply executing the automatic inspection mode based on the inspection template currently being edited, a save icon to save the edited inspection template, and non-destructive inspection Alignment icon (Alignment) to determine whether to automatically align the target images during execution, inspection icon (Inspect) that performs non-destructive testing on the target image in the current setting state during editing, and settings of the currently editing inspection tool. The clear icon (Clear) that initializes only the test result while maintaining it, the reset icon (Reset) that initializes all the test tool settings and test results, or the pixel unit of the image can be arbitrarily changed (preferably to the actual SI unit). A calibration icon (Calibration) may be included. On the other hand, in relation to the simulator icon, when the simulator icon is selected (clicked) by the user, a guide window may be displayed so that the user can designate a target folder on the non-destructive testing device, and when an arbitrary target folder is selected The entire image in the selected folder can be inspected based on the settings of the inspection template currently being edited. At this time, the test result may not be saved. In addition, in relation to the alignment icon, when the alignment icon is selected (clicked) by the user, the non-destructive inspection device asks the user whether to use automatic alignment, whether the alignment standard is the entire target image or a user-defined area. Items that ask whether to limit to only, and other items necessary for alignment (e.g., after alignment, calculate the sum of the difference in the density of each pixel between the target image and the target image that is the standard for the test template after alignment, and the value is If it is less than the reference value, the test is performed, and if it is exceeded, a value to be determined as an alignment error) may be displayed to the user through a separate guide window.

On the other hand, in the first user interface, pointer information 310 representing a position value of a pointer (preferably a mouse pointer) as an (x,y) coordinate value and displaying a pixel brightness value at a point where the corresponding pointer is located. May be displayed more. In the user interface according to the present invention, since the pointer information 310 can be provided to the user in real time whenever the position of the pointer changes, there is a slight difference in brightness in the target image, so that it is difficult for a person to identify it with the naked eye. (E.g., in a transmission image, when an arbitrary part is overlapped, the color displayed in the image may change due to the overlap. If the overlapping part is very thin, a slight color difference may occur.) While checking the pointer information 310 above, it is possible to input various parameters of the inspection template or to designate regions of interest.

Referring back to FIG. 3, test tool groups 400 and 450 and a parameter input area 500 may be displayed on the right side of the screen.

Reference numeral 400 denotes icons presented to allow the user to select an inspection method, and includes a first inspection method (void; when a region of interest is designated, inspects for defects in the relevant region of interest), and a second inspection method. (NGA; if the target images to be inspected are tomographic images of an arbitrary object, an atypical object in the region of interest is detected and defects therein are inspected), or a third inspection method (GA; objects with a fixed pattern) Icons that can be recognized and selected for defect inspection through comparison between the objects or between the objects may be included. The first to third inspection methods will be described in more detail in the description of FIGS. 8 to 10 which will be described later.

Meanwhile, when any one of the first to third inspection methods is selected, a menu for selecting a detailed inspection method (Inspection Method) may be further displayed. For example, assuming that the first inspection method is selected, the original inspection method (Original Inspection Method) below the icons of the reference numeral 400 is recognized as an object and the corresponding region of interest or defect within the object. Search method), Masking Region for Inspection (Masking Region for Inspection; Independently designating a separate masking area within the region of interest, separate preprocessing parameters and segmentation parameters can be set for the masking area, and considering the masking area) Any one of a method of finding defects within the region of interest) or a method of excluding masking regions (Masking Region for Exclusion; a method of designating a masking region to exclude inspection among regions of interest, and finding defects only in regions of interest other than the excluded region) A menu may be displayed to allow you to select. For reference, FIG. 4 shows an embodiment in which a masking area is further designated on a target image displayed in the main area of FIG. 3. Although mentioned for a while in the above description, the masking area designated on the interface according to the present invention can be defined in two ways. One is to distinguish an area in which configurations such as circuits are overlapped within the target image, and within the overlapped area. The parameters for identifying defects of the device may be defined as a masking area as an individually configurable area, and the other may be defined to exclude defect inspection. 4, similar to the case of designating the ROI, the masking area may be designated by a user through a dragging input, similar to when the ROI is designated, or a plurality of By specifying arbitrary points, an inner polygonal shape defined by the corresponding points can also be designated as a masking area.

Meanwhile, under the icon 400 for selecting the inspection method, icons 450 for determining whether to designate the region of interest as a rectangle or a polygon may be further provided in designating the region of interest. The main area 100 of FIG. 1 shows a state in which a region of interest (ROI) is designated within the target image (Img), and the region of interest (ROI) is an area directly designated by the user by clicking a rectangle icon. Is shown.

FIG. 5 is a diagram illustrating a state in which the target image in the main area is inspected when the inspection icon (Inspect) mentioned in FIG. 3 is clicked. After clicking the inspection icon (Inspection), compared with FIG. 1, the color of the square designating the ROI first changes in the target image to indicate to the user that the inspection has been executed. In the object areas identified by, outlines for identifying the corresponding object areas are displayed so that the user can indicate which area has been classified as the object area. For reference, as mentioned above for a while, the target image to be inspected when clicking the inspection icon (Inspect) is the reference image displayed on the main area when creating and editing the current inspection template, and the inspection icon ( Inspect) The inspection by clicking may be performed on a one-time basis, preferably with respect to the reference image. That is, the inspection icon (Inspect) on the first user interface may be used to check whether the non-destructive inspection is appropriately performed based on a region of interest designated by the user himself or parameters set by the user himself on the reference image.

6 shows the area information 1002 displayed when the user places the pointer 1001 in the object area on the main area of FIG. 5. In the first user interface of the present invention, the user wants to see When the pointer 1001 is positioned on the area, as shown in FIG. 6, the size information of the area (preferably, the size information indicated by the size of the pixel or the number of pixels), and the number of defects identified in the object area, or Area information such as the area ratio of the defect (the area ratio occupied by the defect among the object areas) may be displayed. In addition, depending on which area the pointer 1001 is positioned on, an area including the location may be marked with an outline of a different color to distinguish the corresponding area. In FIG. 6, a blue line is enclosed outside the object area.

On the other hand, FIG. 7 shows a state in which outlines 1003 of different colors are displayed to distinguish the corresponding defect area when the pointer 100 is positioned on the defect in the object area.

As described above, information on the corresponding area may be provided according to the point at which the user places the pointer 1001 on the main area after the inspection is performed on the first user interface, and an outline to distinguish the area is separately displayed. Can be done.

Hereinafter, with reference to FIGS. 8 to 10, what kinds of parameters are set in the parameter input area (reference numeral 500) when the first to third test methods and the respective test methods are selected. I will explain whether it can be done.

<First inspection method>

First, the first inspection method is to repeatedly identify whether a defect exists in the corresponding regions when a single region of interest or a plurality of regions of interest exist. For example, a target image photographed on circuit boards manufactured in large quantities In other words, it is a method of repetitively identifying whether or not there is a defect in each of the target images obtained by photographing a large amount of circuit boards produced according to the circuit diagram design of the same configuration and the same design.

Assuming that the transmission image of the object already exists, the first inspection method may begin with the non-destructive inspection apparatus 100 firstly receiving an input from a user and setting a template for inspection. The inspection template setting in the first inspection method is mainly to designate a so-called region of interest for which areas of the object to be inspected by the user, and to perform defect inspection in the corresponding area. It can also be understood as setting various parameters on how to proceed. At this time, the region of interest (ROI) refers to the region in which the inspector wants to identify whether there is a defect. For example, if one circuit board is completely captured as one target image, it is necessary to examine the defect within the target image. There may be areas that do not need to be viewed (e.g., areas on the substrate that are not equipped with elements) and areas that need to be checked for defects. In this step, the image processing speed and operation are performed by setting the areas that need to be inspected in advance. It was to increase the efficiency of the company.

Meanwhile, when the first inspection method is selected, the user can input various parameter values in the parameter input area (reference numeral 500 in FIG. 3 ). When looking at the parameters that can be set, first, parameters related to image preprocessing (preprocessing parameters; referred to as first group parameters for convenience) include the number of smoothing the target image, the pixel value of the kernel size, and the target image. A value for whether to correct the brightness value of the periphery within the region, the approximate size of the region (object region) with a significant pixel value within the region of interest, a value for whether the object region in the region of interest is brighter or darker than the peripheral region, the region of interest A value for a range in which the object region to be searched within is included in the intensity value of each pixel may be included. The above parameters related to image preprocessing are used in various ways prior to full-fledged defect identification in order to more accurately find the sub-region of interest or object region to be searched from within the region of interest set by the inspector for the target image loaded for defect inspection. The values required for preprocessing have been set. In addition, among the parameters that can be set by the user, parameters (segmentation parameters; referred to as second group parameters for convenience) for how to identify the object area or the defect area within the region of interest designated by the user may be included. The types of these second group parameters include a threshold value that can be a reference value for identifying an object area or a defect area or when defining a sub-ROI, and the minimum required to recognize it as an object area or a sub-ROI. The minimum area, the shape of the sub-ROI, the number of rows and spacings of the sub-ROIs listed when a plurality of sub-ROIs are defined and arranged, and the number and spacing of columns may be included.

Among the above parameters, the threshold value is a value used as a standard to determine whether an object area exists in the region of interest designated by the inspector, and whether a defect exists in the object area, and the threshold value represents the brightness value of an arbitrary pixel. It may be an intensity value, or may be a gray value or an RGB value according to the intention of the designer. Throughout this detailed description, unless otherwise specified, a description will be made on the premise that a threshold value is an intensity value representing a brightness value of a pixel to aid understanding of the invention. On the other hand, the threshold value may be able to identify various defects or other significant areas based on the value by receiving a fixed threshold value directly from the inspector. When input is received, it can also be used to identify defects or significant areas based on an automatic threshold. When the inspection is performed based on the automatic threshold value, the inspection is performed according to a threshold value arbitrarily selected by the non-destructive inspection apparatus 100 according to the present invention based on the distribution of image brightness values. In addition, the threshold value may be a reference value used when determining an arbitrary range in identifying a defect area or an object area. For example, the brightness value of each pixel in the target image is constant based on the threshold value. When having a difference value within a range, the pixels may be identified as a defective area or an object area.

On the other hand, in relation to the shape of the sub-region of interest among the above parameters, when the object region is identified within the region of interest designated by the inspector by performing a future inspection, the center value found based on the coordinates of the object region (X-axis , A sub-region of interest can be defined based on a center point found based on an intensity value of the coordinates in the object region or the coordinates obtained by adding all the Y-axis coordinates for each axis and then calculating the average. The shape of may be determined by inputting the second group parameter together by the inspector. The shape of the sub-ROI may include, for example, [Outline], [Circle], and [Rectangle]. When selecting [Outline], the sub-ROI is displayed in its shape along the object area identified based on the threshold. It will be defined, and when selecting [Circle] or [Rectangle], it will be defined as a circle or a rectangle based on the above-mentioned center value or center point. When defining a sub-region of interest in a circle or a rectangle, a diameter or width/height value may be further input from the user.

On the other hand, among the parameters that can be set, good or bad determination condition parameters (evaluation parameters; referred to as third group parameters for convenience) for the identified object area may be further included. For example, if object areas that should exist in different sub-interest areas are interconnected, the determination as a short circuit can be set through parameter input, and must exist without touching any sub-interest areas. If the object area to be performed touches the boundary of the sub-interest area, it can be determined through parameter input to determine that it is also defective. In addition, when a plurality of objects are identified, whether to determine whether to determine a small amount of payment or an excessive amount of payment based on a relative area may be set through parameter input.

8 is a diagram for helping understanding of various areas, defect areas, threshold values, etc. on the target image on the premise that inspection is performed on an arbitrary target image according to the first inspection method.

FIG. 8A shows a state in which the non-destructive testing apparatus 100 recognizes three regions of interest as A1, A2, and A3 in a target image based on a test template set by a user. (For reference, in this embodiment, a description will be made on the premise that the user has designated A1, A2, and A3 as the region of interest. However, the above A1, A2, and A3 are object The regions of interest may be identified as regions.) The regions of interest may be, for example, regions of a circuit board in which coating layers are formed of different materials, and thus may be directly designated by the inspector as regions of interest. For reference, the regions of interest may be set based on a randomly set coordinate system or a position of a pixel in the target image.

On the other hand, in the process of creating and editing the inspection template, when the regions of interest are designated, a threshold value specific to the region of interest, that is, a threshold value that is a reference for determining whether a defect exists in the region of interest in the future, is further set. Explained that it can be. In (a) of FIG. 8, these threshold values are indicated by g1, g2, and g3, and the threshold values may preferably be intensity values representing brightness values of a pixel in each ROI, or It may be a gray value or an RGB value according to the designer's intention. In this case, the threshold values mentioned are not necessarily limited to those mentioned above, and they are values that can be read by a computer, and there is no limitation of the type as long as each pixel can be distinguished. Regarding the utilization of the above thresholds, the above thresholds are defined or identified as defects in the non-destructive testing apparatus 100 having a threshold value exceeding a preset range compared to the threshold value within each region of interest. It can be used to make it happen. For example, if the threshold value in the region of interest A1 is g1, and there are pixels with values exceeding the range of -10% to 10% compared to g1, the set of pixels can be identified as a defective region. have. Alternatively, when the threshold value in the region of interest A1 is g1, it may be implemented to identify all pixels having a value different from that of g1 as a defective region.

Meanwhile, FIG. 8B is a diagram for explaining a defect identification step when the masking area is further included. In the above, it has been described that a masking area may also be designated by a user. Although the masking area is, for example, an area that exists on the same layer on the circuit board, some areas have an obstacle above the corresponding circuit board, and thus may have a darker threshold than an area without obstacles when generating a transmission image. This refers to a region in which the threshold value is changed due to the presence of an obstacle, etc., among regions within the same region of interest. In relation to this, FIG. 8B shows a region of interest A4 and two masking regions M1 and M2 existing in the region of interest. The masking regions may have a threshold value unique to each of the masking regions. As can be seen in FIG. 8B, a unique threshold value such as gm1 and gm2 may be set for each masking region. Regarding the utilization of the intrinsic thresholds of the masking area, if pixels exceeding a preset range compared to the threshold g4 in the A4 area are recognized as defects and are defined as D4, the masking area M1, the ROI A4, and For the defect D5 that exists over the masking area M2, it can be implemented to recognize a defect as a defect by identifying the defect areas based on gm1, g3, and gm2, respectively, and seeing the connection relationship between the defect areas within each area. If, unlike FIG. 8B, if the defect exists only inside the regions M1 and M2, and there is no connectivity defect in the region other than the masking region, the defects will naturally be recognized as separate.

On the other hand, in the embodiment of FIG. 8 above, only the embodiment of determining whether there is a defect by directly comparing pixel values in regions of interest with a threshold value has been described, but the defect can also be identified by the following method. That is, in performing the non-destructive inspection, the non-destructive inspection apparatus 100 may extract at least one object area within each ROI, and perform image processing to identify a defect candidate area within the corresponding object area. , After image processing, defect candidate regions are identified, and further, the non-destructive inspection may be finally performed according to the first inspection method through a process of determining defects. The object area may be defined as an area composed of sets of pixels having similar values in each ROI, and preferably, may be defined as an area having an arbitrary area inside a closed curve. For example, based on (a) of FIG. 8, a total of three defective areas and A1 areas of interest excluding the defective area may exist in the area of interest A1 as a result. In the step of extracting the object area, the defective area , Prior to being classified as a normal region, a region of sets of pixels having similar values in the closed curve is first identified as each object region. The process of identifying defect candidates is the process of identifying candidates to be determined as defects among the previously extracted object areas. In this process, parameters set for each area in the step of setting the inspection template in advance are used. Can be. For example, if a number of pixels having a threshold value exceeding a preset range are included among the object areas, the corresponding object area will be classified as a defect candidate, and an object area that does not may be excluded from the defect candidate. Finally, in the process of determining defect candidates as defects, defect candidates that satisfy the secondary requirements for the defect candidates are determined as defects. If there is a defect candidate that does not satisfy the secondary requirements, the defect candidate is not a defect. It is determined to be.

<Second inspection method>

The second inspection method is to identify defects in a plurality of tomography images when tomography is performed on an object.For example, tomography is used to see whether a defect exists inside an object having an arbitrary volume. When a plurality of tomographic images are generated by taking a photograph, it is an inspection method to determine whether there is a defect in each tomographic image by processing the images in order of the upper tomographic images.

The second inspection method may first be started from the step of setting the inspection template by receiving an input from the user by the non-destructive inspection apparatus 100. The creation and editing of the inspection template in the second inspection method is based on the assumption that the inspection template has been created by the inspector designating an arbitrary region of interest through dragging, etc., how to determine the object region or the sub region of interest including the above object region. It may consist of processes of receiving and storing parameter values for identifying or defining. For reference, the region of interest in the second inspection method, especially when considering that a plurality of tomography images will all have object regions of different shapes, and when considering the circumstances in which the alignment of each tomography image cannot always be accurately achieved, It can be understood to mean an area that always includes an object area that the inspector wants to see.

Many of the parameters that can be directly set by the user when the second inspection method is selected may overlap with the parameters in the first inspection method, but once again summarized as follows.

Parameters related to preprocessing (preprocessing parameters; referred to as first group parameters for convenience) include the number of smoothing the target image and the pixel value of the kernel size, the smoothing number of the object region itself, and the kernel size pixel value, A value for whether to correct the peripheral brightness value within the target image, a value for whether to correct the peripheral brightness value to detect a defect region within the object region, the approximate size of the object region or defect region, and an object within the region of interest Value of whether the area is brighter or darker than the periphery, the value of the range of the object area to be searched for in the region of interest based on the intensity value of each pixel, the defect within the object area based on the intensity value The detection range of the (void) region may be included.

In addition, among the parameters that can be set, parameters (segmentation parameters; referred to as second group parameters for convenience) for how to identify the object region or the defect region within the region of interest may also be included. The types of these parameters may include a threshold value for identifying an object area, a threshold value for identifying a defect area, a minimum area for recognizing an object area, a minimum/maximum area for recognizing a defect area, and the like. Meanwhile, the threshold value for identifying the object area or the threshold value for identifying the defective area may be set to a fixed threshold value or an automatic threshold value similar to that mentioned in the first inspection method.

On the other hand, among the parameters that can be set, good or bad determination condition parameters (evaluation parameters; referred to as third group parameters for convenience) for the identified defective area may be further included. These include, for example, the total area of the object area and the number of object areas that fall into the normal category, the area of the minimum/maximum defective area that falls into the normal category, the minimum/maximum value of the total area of the defect area that falls into the normal category, and The number, area, and size of defects may be included.

9A, 9B, and 9C show tomography images obtained by tomography of a certain object, that is, target images. In this case, the region of interest covers all the outermost parts of the object region. It can be seen that it is set as a dotted rectangular area, which is an existing area.

After image preprocessing, the non-destructive inspection apparatus 100 extracts each tomographic image, that is, an object region of the target image. The object area in the second inspection method may be defined as an area composed of sets of pixels having similar values within each ROI as in the first inspection method, and preferably, an area having an arbitrary area inside the closed curve. Can be defined as In FIGS. 9A to 9C, it can be seen that the respective object regions are marked as O1 to O7. That is, the object area in the second inspection method means that areas to be recognized as meaningful parts within the ROI are identified based on the brightness value or intensity value, and this object area extraction is performed by each tomographic image (target image). Defect inspection must be performed for each tomographic image (target image) in that they all contain significant areas of different shapes, in other words, all tomography images (target images) contain atypical object areas. I understand that.

In the process of extracting the object area, the threshold value previously set in the inspection template may be used. For example, when the values of each pixel are compared with each other and the difference is within a preset range, that is, mutual pixel values If it is determined that this is similar, the object region can be extracted by treating it as one region, and otherwise treating it as a different region.

Meanwhile, after the object region is extracted from the region of interest, the non-destructive inspection apparatus may perform pre-processing for each object region, and after performing the pre-processing, a step of identifying a defect may proceed. In the step of identifying defects within the object area, as described in the first inspection method, pixels having a value exceeding a preset range from the threshold value of each object area may be identified as defects, or As described in the first inspection method, a method of identifying defect candidates and determining defects therefrom can also be used. Since the method of identifying the defect in the object area is similar to that of the first inspection method, a detailed description will be omitted here. For reference, FIG. 9A illustrates an example in which defects D1 and D2 exist in the O1 object area, defects D3 and D4 in the O2 object area, and defects D5 and D6 in the O3 object area.

<3rd inspection method>

The third inspection method is to identify defects among the above objects by performing an operation based on a relative area when a plurality of objects having similar shapes exist in one target image.

The inspection template setting process in the third inspection method is similar to the inspection template setting process in the first inspection method described above. That is, even in the third inspection method, a template for inspection may be set by receiving inputs from the inspector for the first group parameters, the second group parameters, and the third group parameters, respectively.

For reference, in the third inspection method, in order to identify sub-ROIs within the ROI, parameters for how to define the sub-ROI may be input. The sub-region of interest refers to an area that the examiner wants to examine, in particular, within the region of interest. Due to the fact that a number of objects having similar shapes are arranged in the target image in the third inspection method, the location where each object is located is determined. It means that an individual region of interest is set around the center. On the other hand, in some cases, a very large number of objects may be included in the target image. In this case, allowing the user to designate and set sub-interest areas of individual objects can be very cumbersome. According to the non-destructive testing apparatus 100, by first receiving an input on how to define the sub-ROI from the user, first defines the sub-ROI, and the non-destructive testing device 100 defines the previous sub-ROI The shape of the sub-ROI to be identified in the future is defined with reference to and then the sub-ROI can be identified and determined within the ROI by referring to the defined shape of the sub-ROI.

For example, when editing the inspection template, the user can select the shape of the sub-region of interest as [Outline], [Circle], and [Rectangle]. If [Rectangle] is selected, the shape of the sub-region of interest is [ It can have a square] by default. In addition, in the step of identifying and determining the sub-ROI, after the non-destructive inspection apparatus 100 first identifies the object area within the ROI, the center value of each object area found based on the coordinates of each object area ( A step of obtaining a coordinate obtained by adding all X-axis and Y-axis coordinates for each axis and then calculating the average) or a center point calculated based on the intensity values of coordinates in the object areas (center point considering coordinates and intensity) May be included, and after the central value or the central point is obtained, sub-interest regions according to the standard (diameter, width/height, etc.) of the shape previously set by the inspector may be generated and determined around a corresponding one point. All values used in the calculation in the above process may be input by the user through the parameter input area 500.

Hereinafter, a third inspection method will be further described with reference to FIG. 10. For example, assuming that a plurality of solders exist on an object (circuit board), a plurality of circular regions may exist in the region of interest, as shown in FIG. After first identifying circular areas as object areas, the center value or center point is calculated for each object area, and based on the calculated center value or center point, the shape previously set by the inspector, that is, [rectangle] Sub-regions of interest can be created and determined. In the target image to be actually inspected, each of the solders may not necessarily have a perfect circle as shown in FIG. 10, and the sub-region of interest generated by this may not necessarily have the same size. Furthermore, as shown in FIG. 10, the alignment state of the sub-ROIs, that is, the spacing between the sub-ROIs, may not exactly match.

Meanwhile, after a plurality of sub-ROIs are determined in the target image, the non-destructive inspection apparatus 100 may identify the defective region in each of the sub-ROIs. The method of identifying defects in the third inspection method is partially different from the previous first and second inspection methods. Specifically, defects are identified based on the relative area between the object areas in each sub-region of interest or defect areas in the object area. Identify. In the third inspection method, the process of identifying defects in each sub-ROI includes: extracting a defect area within the sub-ROI(s), calculating at least one of the average area of the object areas or the average area of the defect area. And identifying a sub-ROI having an object area in which the area ratio of the object area or the defect area to the average area exceeds a preset range as a defect.

First, the step of extracting the defective area from the object area in the sub-ROI may be defined as an area composed of sets of pixels having similar values even in the object areas, and preferably, an area having an arbitrary area inside the closed curve. It can be defined as an area. In FIG. 10, object areas in the sub-ROI are indicated by a symbol of _{O n} , and defect areas are indicated by a symbol of _{d n in some object areas.}

Next, in the step of calculating at least one of the average area of the extracted object areas or the average area of the defect areas in each of the object areas, the calculated average area is whether or not the plurality of object areas in the future have defects of overpayment or overpayment. It is a value used to determine whether or not, that is, a reference value. In calculating such a reference value, in general, it can be obtained by summing all object areas or areas of defect areas and dividing by the number of object areas or defect areas. There may be defects with very large area values, very small values, or no area values at all.In this step, the areas of the object areas or defect areas are respectively calculated and sorted in descending or ascending order. One feature is to exclude from the average area calculation the object areas having the largest difference from the median value, for example, the median value (the value existing in the center of the listed object area values). have. For example, referring to FIG. 10, there are a total of 9 object areas (O1 to O9) in the target image, of which, when calculating the average area, the O3 object area having the largest area and the smallest The O7 object area having an area can be excluded. In this way, when calculating the average area, the object area that has the greatest difference from the median value is excluded in order to increase the reliability of the average area, that is, the reference value, which will be a reference for future defect identification. For reference, object areas that are excluded when calculating the average area are referred to as exclusion candidates for convenience. Meanwhile, in the above embodiment, the embodiment in which the area of each object area is compared based on the [median value] is mentioned, but it is understood that this is not necessarily limited thereto. That is, the non-destructive inspection apparatus 100 may determine object areas or defect areas to be excluded when calculating the average area according to the inspection template set by the inspector or directly input by the inspector. For example, it can be implemented to calculate the average area while excluding the area values of object areas or defect areas included in the top 3% when viewed in descending order, and area values included in the top 2% when viewed in ascending order, or It can be implemented to be used for the average area calculation only for areas having a distribution value of a certain level (number) or more from the area distribution of the fields, or when viewed in ascending or descending order, the average area calculation is performed while excluding a predetermined number of areas. It can also be implemented to do so.

On the other hand, the non-destructive inspection apparatus 100 compares the average area of the object areas or the average area of the defect areas with the area of each object area or defect area to determine that defects exist in object areas exceeding a preset range. Identify. At this time, in this step, of course, object areas that have been excluded during the average area calculation will be identified as defects, and in addition, object areas having a large difference compared to the average area will be identified as defects.

Referring to an embodiment with reference to FIG. 10, the non-destructive inspection apparatus 100 uses O3 (the object area having the largest area) and O7 (the object area having the smallest area) when calculating the average area of the object areas. You can calculate the average area for the area values of O1, O2, O4 to O6, O8, and O9, excluding the area value of ), and after the average area is calculated, the corresponding value is calculated for each object area (O1 to O9). Compared with, it is possible to determine a soldering defect/overpayment defect. On the other hand, when calculating the average area, the defective areas D1 and D2 may be further considered, and when determining a defect, a determination as to how large a portion of the defective area occupies in each object area may also be made. That is, an arbitrary object area may be determined to be within the normal range relative to the average area, but at the same time, the proportion of the defective area existing therein may exceed a preset value (eg, 60%) and may be determined as a defect. .

The first to third inspection methods have been described above.

11 is a view showing a display state of a second user interface according to the present invention. The second user interface is displayed when the user clicks the automatic test menu, that is, the icon on the left side of the screen in FIG. Reference numeral 202), a current inspection progress status (reference numeral 203), and a result (reference numeral 204) may be displayed.

In the plurality of icons 202 necessary for the automatic test execution, an icon for executing or stopping the automatic test (Start of Stop), an icon for setting a macro for automatic test (Setting), and an icon for initializing the test list ( Clear All), an icon to initialize the inspection statistics (Reset Stat) may be included.

In addition, in the area indicating the current inspection progress status 203, a character “Inspecting” may be displayed to indicate that the current automatic inspection mode is activated, whereas the character “Stop” indicates that the automatic inspection mode is deactivated. Can be marked to inform.

On the other hand, in the area displaying the result 204, the number of errors including the number of normal inspection results (OK), the number of defective persons (NG), and alignment errors may be displayed. A list of images that have been inspected by the inspection mode may be displayed.

A method for providing a non-destructive inspection function based on an abnormal transmission image and a storage medium for executing this method have been described. Meanwhile, the present invention is not limited to the specific embodiments and application examples described above, and various modifications are implemented by those of ordinary skill in the art without departing from the gist of the present invention claimed in the claims. Of course, these modifications should not be understood as being distinguished from the technical idea or perspective of the present invention.

50 X-ray devices
100 non-destructive testing device

Claims (3)

As a non-destructive testing method,
The non-destructive inspection method is characterized in that performing repetitive non-destructive inspection based on target images-the target images are respective transmission images of a plurality of objects,
The non-destructive inspection method includes: receiving a user input and setting a template for inspection;
The step of setting the inspection template,
Designating a region of interest in which non-destructive testing is to be performed within transmission images of the objects; And
An object area in the ROI-the object area is an area consisting of sets of pixels having similar values within a preset range-or setting a plurality of parameters for identifying a defective area;
Containing,
Non-destructive testing method. As a non-destructive testing method,
The non-destructive testing method is characterized in that sequential non-destructive testing is performed based on target images-the target images are a plurality of tomographic images obtained by tomography of a specific object,
The non-destructive inspection method includes: receiving a user input and setting a template for inspection;
The step of setting the inspection template,
Designating a region of interest in which non-destructive testing is to be performed within the tomographic images of the object; And
An object area within the ROI-the object area is an area consisting of sets of pixels having similar values within a preset range-or setting a plurality of parameters for identifying a defective area;
Containing,
Non-destructive testing method. As a non-destructive testing method,
The non-destructive inspection method is characterized in that the non-destructive inspection is performed based on a target image-the target image includes a plurality of objects having a similar shape,
The non-destructive inspection method includes: receiving a user input and setting a template for inspection;
The step of setting the inspection template,
Designating a region of interest in which non-destructive testing is to be performed within the tomographic images of the object;
Designating a sub-ROI including at least one object therein as included in the ROI; And
Setting a plurality of parameters for identifying a defective area or an object area within the sub-ROI, wherein the object area is an area consisting of sets of pixels having similar values within a preset range;
Containing,
Non-destructive testing method.

Images