Inspection equipment and inspection method

Abstract

The present invention provides an inspection device and inspection method capable of inspecting the bonding state of an object based on temperature image data of the object. [Solution] The inspection device includes an input unit that acquires temperature image data generated by an imaging device that photographs objects joined in a predetermined joining region in a time series, and a processor that inspects the joining state of the object based on the temperature image data. The processor calculates a correlation value between time series data for a reference point on the temperature image data and time series data for a predetermined point on the temperature image other than the reference point, within a predetermined period. The processor inspects the joining state based on the correlation values ​​calculated for each of the multiple points other than the reference point on the temperature image.

Description

Technical Field

[0001] The present disclosure relates to an inspection apparatus and an inspection method.

Background Art

[0002] Patent Document 1 discloses a non-destructive inspection method using infrared lock-in thermography for determining the quality of a joint in a lap joint where dissimilar metal materials are friction stir welded. This non-destructive inspection method determines the quality of the joint by comparing the amount and phase of the acquired infrared rays with a predetermined determination value set in advance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] An object of the present disclosure is to provide an inspection apparatus and an inspection method capable of inspecting the joint state of an object based on temperature image data of the object.

Means for Solving the Problems

[0005] An inspection apparatus according to an aspect of the present disclosure includes an input unit that acquires temperature image data generated by a photographing device photographing an object joined in a predetermined joint region in a time series, and a processor that inspects the joint state of the object based on the temperature image data. The processor calculates a correlation value between time series data of a reference point on the temperature image indicated by the temperature image data within a predetermined period and time series data of a predetermined point other than the reference point on the temperature image, and inspects the joint state based on the correlation values of each of the plurality of points calculated for each of the plurality of points other than the reference point on the temperature image.

[0006] An inspection method according to one aspect of the present disclosure is an inspection method for inspecting the bonding state of an object, which is executed by a processor based on temperature image data generated by photographing an object bonded in a predetermined bonding region in a time series, acquired from a photographing device, and includes: calculating a correlation value between time series data for a reference point on the temperature image data indicated by the temperature image data and time series data for a predetermined point other than the reference point on the temperature image within a predetermined period; and inspecting the bonding state based on the correlation values ​​for each of the multiple points calculated for each of the multiple points other than the reference point on the temperature image. [Effects of the Invention]

[0007] According to this disclosure, an inspection device and inspection method can be provided that can inspect the bonding state of an object based on temperature image data of the object. [Brief explanation of the drawing]

[0008] [Figure 1] Block diagram showing an example configuration of an inspection system according to one embodiment 1 of this disclosure. [Figure 2] Block diagram showing an example configuration of the inspection equipment of the inspection system shown in Figure 1. [Figure 3] Flowchart illustrating the operation of the inspection device according to Embodiment 1 [Figure 4] A graph showing an example of time-series data of temperature changes for points included in the junction region in a thermal image. [Figure 5] This graph shows an example of time-series data of temperature changes for points in a thermal image that are located in a region different from the junction region. [Figure 6A] A schematic diagram illustrating an analysis image generated based on predetermined analysis conditions. [Figure 6B] A schematic diagram illustrating an analysis image generated based on predetermined analysis conditions. [Figure 6C] A schematic diagram illustrating an analysis image generated based on predetermined analysis conditions. [Figure 6D] A schematic diagram illustrating an analysis image generated based on predetermined analysis conditions. [Figure 7] A schematic diagram illustrating a phase image generated using Fourier transform based on predetermined analysis conditions. [Figure 8] A schematic diagram including a set of images to explain the defect estimation process. [Figure 9] A flowchart illustrating the process of determining analysis conditions. [Figure 10A] Graphs showing an example of time series data where the time average of the time derivative for each of the time series data of temperature change shown in Figures 4 and 5 has been calculated. [Figure 10B] Figure 10A shows a graph illustrating an example of time series data that shows the difference between two time series data. [Figure 11] Flowchart illustrating the operation of the inspection device according to Embodiment 2 [Figure 12] Flowchart showing the details of the effective area ratio calculation process in Figure 11 [Figure 13] Schematic diagram including a set of images to explain the effective area ratio calculation process in Figure 11. [Modes for carrying out the invention]

[0009] The embodiments will be described below with reference to the drawings as appropriate. However, unnecessarily detailed explanations may be omitted. For example, detailed explanations of already well-known matters or redundant explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily verbose and to facilitate understanding by those skilled in the art. The inventors provide the accompanying drawings and the following explanation so that those skilled in the art can fully understand this disclosure, and do not intend to limit the subject matter described in the claims by means of these.

[0010] In this disclosure, when describing modifications or multiple embodiments, the differences from Embodiment 1 will be described primarily. In this case, components that are the same or equivalent as those in Embodiment 1 will be denoted by the same reference numerals in the modifications or other embodiments. Furthermore, descriptions that overlap with those in Embodiment 1 may be omitted in the modifications or other embodiments.

[0011] (Embodiment 1) [1-1. Configuration] [1-1-1. Configuration of the inspection system] FIG. 1 is a block diagram showing a configuration example of an inspection system 1 according to Embodiment 1 of the present disclosure. The inspection system 1 includes an inspection device 10, an infrared camera 17, an excitation source 18, a control box 15, a power supply 16, and an alarm device 19.

[0012] The inspection system 1 inspects the bonding state of the work 90 nondestructively using an active thermography method that applies excitation energy to the work 90, which is an inspection object, and takes a temperature image. In the present embodiment, the inspection system 1 applies excitation energy to the work 90 by the excitation source 18 and takes temperature images in time series using the infrared camera 17.

[0013] In the example of FIG. 1, the work 90 includes a transmissive material 91 that mainly transmits light irradiated from a light source for welding or soldering, and an absorptive material 92 that mainly absorbs the light. The transmissive material 91 and the absorptive material 92 are joined in the joining region 95 by the transmissive material 91 and / or the absorptive material 92 melted by the light from the light source. In the present embodiment, the inspection system 1 inspects the bonding state of the transmissive material 91 and the absorptive material 92 nondestructively and can estimate the presence or absence of defects in the joining.

[0014] The material of the work 90 is, for example, a resin such as polybutylene terephthalate (PBT), polyphenylene sulfide (PPS), or polyamide (PA). Alternatively, the material of the work 90 may be ceramics. The material of the work 90 may be a metal such as cold-rolled steel (SPCC steel) or aluminum.

[0015] The infrared camera 17 is an example of an imaging device that captures a time-series image of an imaging area including at least a portion of the workpiece 90 to generate multiple temperature image data (hereinafter sometimes simply referred to as "temperature images"). The infrared camera 17 generates temperature image data by converting infrared image data into a temperature distribution, but in this specification, the infrared image data before such conversion and the temperature image data after conversion may not be treated as particularly distinguished. The infrared camera 17 includes, for example, an infrared sensor that detects infrared radiation having a wavelength of 3 μm to 15 μm. The frame rate of the infrared camera 17 is, for example, 50 Hz (or 50 fps), but is not limited thereto.

[0016] In this disclosure, “image” includes one or more moving images and one or more still images.

[0017] The excitation source 18 is an example of a heating device capable of heating the workpiece 90. The excitation source 18 is, for example, a light source such as a xenon lamp, a halogen lamp, or a laser light source, but is not limited to these, and can be any energy source capable of radiating energy.

[0018] The excitation source 18 can perform flash heating (pulse heating) by flash emission on the workpiece 90, step heating by heating in a step-like manner, or heating by continuously repeating these methods. The wavelength band of the light emitted by the excitation source 18 may be the same as or different from the wavelength band of infrared light detectable by the infrared camera 17. Figure 1 shows two excitation sources 18 as an example, but the number of excitation sources 18 is not limited to this, and may be one or three or more.

[0019] The power supply 16 supplies power to the infrared camera 17 and the excitation source 18. The control box 15 includes a control circuit that controls the power supply 16 based on control signals from the inspection device 10. The control box 15 may also control the emission method, emission period, emission time, etc., of the excitation source 18.

[0020] The notification device 19 notifies information to the outside. For example, the notification device 19 is controlled by the inspection device 10 and notifies the user of information indicating the inspection result of the internal state of the workpiece 90. The notification device 19 may include visual notification devices such as light sources such as LEDs, displays, and indicators. The notification device 19 may also include auditory notification devices such as speakers.

[0021] [1-1-2. Configuration of the inspection device] Figure 2 is a block diagram showing an example configuration of the inspection device 10 shown in Figure 1. The inspection device 10 comprises a processor 11, a storage device 12, and an interface 13.

[0022] The processor 11 consists of a CPU, MPU, etc., and controls the entire inspection device 10 by executing various programs stored in the storage device 12. The processor 11 controls the excitation source 18 via the control box 15, thereby controlling the start and stop of the heating output of the excitation source 18. The processor 11 also controls the shooting operations of the infrared camera 17, such as starting and stopping shooting. Furthermore, as described later, the processor 11 performs processing to inspect the bonding state of the workpiece 90 based on the temperature image data stored in the storage device 12.

[0023] The storage device 12 is a recording medium that stores various information, including programs and data necessary to realize the functions of the inspection device 10. The storage device 12 can be implemented as, for example, a semiconductor storage device such as flash memory or a solid-state drive (SSD), a magnetic storage device such as a hard disk drive (HDD), or other recording media, either alone or in combination thereof. The storage device 12 is not limited to an internal storage device installed in the same enclosure as the processor 11, but may also be an external storage device, a NAS (network-attached storage) type storage device, etc. The storage device 12 may also include volatile memory such as RAM.

[0024] Interface 13 connects the inspection device 10 to external devices such as the infrared camera 17, control box 15, notification device 19, and excitation source 18. Interface 13 may also be a communication circuit that performs data communication according to existing wired communication standards or wireless communication standards.

[0025] Interface 13 is an example of an input unit that connects the inspection device 10 and the infrared camera 17 in order to input temperature image data from the infrared camera 17 to the inspection device 10. Interface 13 is also an example of an output unit that connects the inspection device 10 to external devices such as the control box 15, notification device 19, and excitation source 18 in order to output information such as control signals from the processor 11 to these external devices. Such input and output units may be implemented as an integrated interface 13 that serves as both input and output, as shown in Figure 2, or they may be implemented as multiple interface circuits.

[0026] [1-2. Operation] [1-2-1. Overall Operation] Figure 3 is a flowchart illustrating the operation of the inspection device 10 according to Embodiment 1. Each process shown in this flowchart is executed, for example, by the processor 11 of the inspection device 10. In Embodiment 1, the material of the workpiece 90 is PBT.

[0027] In Figure 3, first, the processor 11 acquires the analysis conditions (S1). The analysis conditions are, for example, entered by the user and pre-stored in the storage device 12. The analysis conditions include, for example, parameters for the analysis time and the position of the reference point, which will be described later. The analysis time includes, for example, the start time of the analysis and the end time of the analysis. Alternatively, instead of the analysis time, the processor 11 may acquire a setting on which of the multiple temperature images generated by the infrared camera 17 will be used for the analysis. The processor 11 may be configured to determine the analysis conditions using the multiple temperature images acquired, as will be described in detail later.

[0028] Next, the processor 11 starts heating the workpiece 90 with the excitation source 18 and taking images of the workpiece 90 with the infrared camera 17 (S2).

[0029] The processor 11 acquires multiple temperature images generated by the infrared camera 17 capturing the workpiece 90 in a time series via the interface 13 (S3).

[0030] Next, the processor 11 analyzes the temperature image data based on the multiple temperature images acquired in step S3 using the zero-means normalized cross-correlation (ZNCC) method based on the analysis conditions. Specifically, the processor 11 uses the ZNCC method to calculate correlation values ​​that indicate the similarity between the time-series data of a predetermined reference point on the temperature image data and the time-series data of predetermined points other than the predetermined reference point, within a predetermined period set by the analysis conditions (S4). The processor 11 calculates correlation values ​​for each of the multiple points other than the predetermined reference point on the temperature image and generates an analysis image using the calculated multiple correlation values ​​(S5). The processor 11 plots each correlation value on a two-dimensional plane according to the position of each of the multiple points and assigns a predetermined brightness value determined based on the correlation value to each pixel to generate the analysis image.

[0031] Processor 11 can calculate the correlation value C by using equation (1).

number

[0032] An example of the analysis image generation process will be explained with reference to Figures 4 to 6D. Figure 4 is a graph showing an example of time-series data of temperature change for points included in the junction region in a temperature image. Figure 5 is a graph showing an example of time-series data of temperature change for points included in a region other than the junction region in a temperature image. Figures 6A to 6D are schematic diagrams illustrating analysis images 20a to 20d generated based on predetermined analysis conditions.

[0033] In this specification, the "joined area" is the area designated as the target of joining. While it is generally desirable for the entire joined area to be joined, depending on the joining quality, part or all of the joined area may not be joined. Furthermore, the portion of the joined area that is actually joined may include a portion where the joining strength is above a predetermined value and a portion where the joining strength is below a predetermined value. The processor 11 is configured to estimate the unjoined areas within the joined area based on the analysis image and detect them as defects. The unjoined areas within the joined area may, for example, be areas where the joining strength is estimated to be below a predetermined value.

[0034] Figure 4 shows an example of a time-series data graph of temperature change at a first point on the temperature image corresponding to point P1 on the analysis image 20a shown in Figure 6A. Figure 5 shows an example of a time-series data graph of temperature change at a second point on the temperature image corresponding to point P2 on the analysis image 20d shown in Figure 6D. The first and second points may be regions containing one pixel on the temperature image, or regions containing multiple pixels.

[0035] The vertical axis of the graphs in Figures 4 and 5 represents temperature (°C). The horizontal axis of the graphs in Figures 4 and 5 represents time (s). In the graphs shown in Figures 4 and 5, period T1 represents the period from 0.14s to 1s, and corresponds to the analysis conditions used to generate the analysis images 20a and 20d shown in Figures 6A and 6D. Period T2 represents the period from 1s to 2s, and corresponds to the analysis conditions used to generate the analysis image 20b shown in Figure 6B. Period T3 represents the period from 2s to 5s, and corresponds to the analysis conditions used to generate the analysis image 20c shown in Figure 6C.

[0036] In the graphs shown in Figures 4 and 5, time 0s indicates the timing when the workpiece 90 is heated by the excitation source 18. As shown in Figure 4, in Embodiment 1, the temperature at the first point on the temperature image rapidly decreases after the temperature rise due to the heating of the workpiece 90. Then, after a predetermined time has elapsed since heating, it rises again and then falls again. Period T1 includes the period of temperature decrease and subsequent rise. Period T2 includes the period of the subsequent rise. Period T3 includes the period of temperature decrease after the subsequent rise.

[0037] As shown in Figure 5, in Embodiment 1, the temperature at the second point on the temperature image differs from the temperature change at the first point. After the temperature rise due to heating of the workpiece 90, it drops sharply, and then gradually decreases while fluctuating slightly. Period T1 includes the period of temperature decrease.

[0038] Thus, the region on the temperature image corresponding to the junction region shows a different temperature change than the region on the temperature image corresponding to a region other than the junction region. In Embodiment 1, the processor 11 generates an analysis image by utilizing the difference in temperature changes between regions on the temperature image.

[0039] As described above, the processor 11 analyzes temperature image data using the ZNCC method and generates an analysis image. The ZNCC method is a pattern matching technique used to quantitatively calculate the similarity between a two-dimensional template image and a two-dimensional input image at each position in the input image. In the inspection apparatus 10 according to Embodiment 1, the processor 11 applies the ZNCC method to the time-series data of temperature changes. Specifically, the processor 11 calculates a correlation value between the time-series data for a predetermined reference point on the temperature image and the time-series data for points on the temperature image other than the predetermined reference point.

[0040] In this specification, the correlation value indicates the degree of similarity between the time-series data of a given reference point and the time-series data of other predetermined points other than the reference point over a predetermined period. The correlation value calculated using the ZNCC method is expressed in the range of -1 to 1. The higher the similarity between the two time-series data, the closer the correlation value is to 1. After calculating the correlation value for each of the multiple points other than the predetermined reference point, the processor 11 plots the multiple calculated correlation values ​​in two dimensions corresponding to their positions on the temperature image and generates an analysis image in which each correlation value is represented on a predetermined color scale. The processor 11 may also generate an analysis image in which each correlation value is represented on a grayscale. Each pixel in the analysis image has a brightness value determined based on the correlation value. For example, the processor 11 can set the brightness value of each pixel such that the pixel corresponding to a correlation value of 1 has the maximum brightness value, the pixel corresponding to a correlation value of -1 has the minimum brightness value, and the pixel corresponding to a correlation value between 1 and -1 has a brightness value of a gradation corresponding to that correlation value. The analysis image does not need to be generated in grayscale corresponding to correlation values ​​in the range of -1 to 1. The processor 11 can arbitrarily set upper and lower limits for the grayscale based on the maximum and minimum calculated correlation values. For example, the processor 11 may set the brightness value of each pixel such that the pixel corresponding to a correlation value of 1 has the maximum brightness value, and the pixel corresponding to a correlation value of 0 or less has the minimum brightness value. The processor 11 may also set the brightness value of each pixel such that the pixel corresponding to the maximum correlation value has the maximum brightness value, and the pixel corresponding to the minimum correlation value has the minimum brightness value. The upper and lower limits for the grayscale may be set by the user.

[0041] The analysis image 20a in Figure 6A is generated so that the grayscale corresponds to a correlation value between approximately 0.7 and approximately 1.0. The analysis image 20b in Figure 6B is generated so that the grayscale corresponds to a correlation value between approximately -0.4 and approximately 1.0. The analysis image 20c in Figure 6C is generated so that the grayscale corresponds to a correlation value between approximately 0.6 and approximately 0.95. The analysis image 20d in Figure 6D is generated so that the grayscale corresponds to a correlation value between approximately 0.8 and approximately 1.0. Compared to analysis images 20b and 20c, the boundary between white and black pixels is more clearly depicted in analysis image 20a. Compared to analysis image 20b, the boundary between white and black pixels is more clearly depicted in analysis image 20c. Also, compared to analysis images 20b and 20c, the boundary between white and black pixels is more clearly depicted in analysis image 20d. Therefore, it is estimated that the analyzed images 20a and 20d are more suitable for use in examining the fusion state than the analyzed images 20b and 20c. It is also estimated that the analyzed image 20c is more suitable for use in examining the fusion state than the analyzed image 20b.

[0042] As described above, the analysis images 20a to 20c are generated using the first point within the joined region corresponding to point P1 as the reference point. Therefore, in the analysis images 20a to 20c, the higher the correlation value of the pixels calculated for other predetermined points, the more it indicates that the region corresponding to those other predetermined points has a temperature change similar to that of the first point. Thus, in the analysis images 20a to 20c, regions with high correlation values, i.e., regions with white pixels, are presumed to be joined regions. Conversely, regions with low correlation values, i.e., regions with black pixels, are presumed to be regions different from joined regions.

[0043] The analysis image 20d is generated using a second point in a region different from the joined region, corresponding to point P2, as the reference point. Therefore, in the analysis image 20d, the higher the correlation value of the pixels calculated for the other predetermined point, the more it indicates that the region corresponding to the other predetermined point has a temperature change similar to that of the second point. Thus, in the analysis image 20d, regions with low correlation values, i.e., regions with black pixels, are presumed to be joined regions. Conversely, regions with high correlation values, i.e., regions with white pixels, are presumed to be regions different from the joined region.

[0044] In Embodiment 1, the greater the difference in brightness between a point included in the joining region and a point included in a region other than the joining region, the easier it is for the user or processor 11 to distinguish between the joined region and other regions in the analyzed image. For example, the processor 11 can calculate the contrast ratio using the brightness value Lmax corresponding to the highest correlation value and the brightness value Lmin corresponding to the lowest correlation value in the analyzed image as follows. Contrast ratio = (Lmax - Lmin) / (Lmax + Lmin) In the analyzed image 20a, the contrast ratio is 0.90.

[0045] For example, the processor 11 can perform a Fourier transform or discrete Fourier transform on temperature image data to generate a phase image showing the phase characteristics of the workpiece 90, and inspect the bonding state of the workpiece 90 based on the phase image. The processor 11 can generate a phase image suitable for inspecting the bonding state by using, for example, an analysis frequency of 0.4 Hz, 0.7 Hz, or 1.0 Hz. Figure 7 is a schematic diagram illustrating a phase image 20e generated using a Fourier transform based on predetermined analysis conditions. The phase image 20e shown in Figure 7 is generated so that the grayscale corresponds to a phase between approximately -0.7 and approximately 1.5. Each pixel of the phase image 20e has a brightness value determined based on the value of the phase (°). The phase image 20e shown in Figure 7 is generated using the first term of a Fourier series, and the analysis frequency of the Fourier transform is, for example, 1.0 Hz. The phase image 20e shown in Figure 7 is generated using temperature image data for a period of 0.3 seconds to 1 second. The contrast ratio of the phase image 20e is 0.85. The phase image is generated using temperature image data for a period determined by the analysis frequency.

[0046] The analysis image 20a and the phase image 20e, both created using the ZNCC method, are images generated for the same workpiece 90. As described above, the contrast ratio of the analysis image 20a is 0.90, and the contrast ratio of the phase image 20e is 0.85. Thus, the contrast ratio of the analysis image generated using the ZNCC method can be higher than that of the corresponding phase image. Therefore, the user or processor 11 can determine the bonding state more accurately by inspecting the bonding state using the analysis image generated by the ZNCC method than by using the phase image generated using the Fourier transform. Consequently, the user or processor 11 can accurately inspect the bonding state of the workpiece 90.

[0047] Next, the processor 11 performs a process to inspect the bonding state of the workpiece 90 based on the analysis image. In Embodiment 1, in order to inspect the bonding state, the processor 11 performs a process to estimate the presence or absence of bonding defects within the bonding area of ​​the workpiece 90 based on the analysis image. The defect estimation process performed by the processor 11 will be described with reference to Figures 3 and 8. Figure 8 is a schematic diagram including a set of images to explain the defect estimation process performed in the inspection device 10. In Embodiment 1, the processor 11 performs defect estimation processing related to the bonding of the workpiece 90 using the analysis image 20a.

[0048] First, the processor 11 acquires junction area mask information (S6). The mask information is the area in the analysis image 20a that corresponds to the junction area of ​​the workpiece 90. The mask information is determined based on drawings and design information and stored, for example, in the storage device 12. The mask information may also be determined based on the area of ​​the workpiece 90 that is heated by the excitation source 18. An example of mask information is a mask image M in which pixels corresponding to the junction area of ​​the analysis image 20a are set to white (pixel value 1) and other pixels to black (pixel value 0). If mask information is not already stored in the storage device 12, the processor 11 may generate mask information based on the analysis image 20a.

[0049] Next, the processor 11 converts the analysis image 20a into a grayscale image 21 having brightness values ​​from 0 to 255. The processor 11 superimposes mask information onto the grayscale image 21 and changes the brightness values ​​of the areas in the grayscale image 21 corresponding to the black areas on the mask information to zero (S7) to generate a joined region image 22. In other words, the processor 11 generates the joined region image 22 by changing the pixels in the grayscale image 21 that correspond to areas other than the joined region to black. In this specification, "analysis image" may include an image obtained by converting the analysis image to an arbitrary scale. The processor 11 may be configured to perform the processing from step S7 onward using the analysis image.

[0050] Next, the processor 11 binarizes the joined region image 22 based on predetermined conditions (S8) to generate a binarized image 23. The predetermined conditions may include, for example, that the brightness value of each pixel in the grayscale image 21 is greater than or equal to a predetermined threshold. The predetermined threshold is, for example, a brightness value of 160, but is not limited to this brightness value, and any value can be set depending on the object to be analyzed. The predetermined conditions may also include that the correlation value corresponding to each pixel in the joined region image 22 is greater than or equal to a predetermined threshold. In Embodiment 1, the processor 11 determines that pixels with a brightness value greater than or equal to the predetermined threshold satisfy the predetermined conditions and changes them to white, and determines that pixels with a brightness value less than the predetermined threshold do not satisfy the predetermined conditions and changes them to black.

[0051] The processor 11 determines whether or not there are defects in the joining of the workpiece 90 in the binarized image 23 (S9). In Embodiment 1, if a black pixel exists in the joining region, the processor 11 estimates the region corresponding to that pixel as a defect and determines that the joining of the workpiece 90 contains defects. If no black pixels exist in the joining region, the processor 11 determines that the joining of the workpiece 90 does not contain defects. The processor 11 may be configured to generate a binarized image 23 by changing pixels that satisfy the above predetermined conditions to black and pixels that do not satisfy the predetermined conditions to white, and to estimate the region corresponding to the white pixels as defects.

[0052] If the processor determines that the workpiece 90 contains a defect (S9: Yes), the processor 11 causes the notification device 19 to perform a notification operation (S10). This allows the user to know that the joint of the workpiece 90 contains a defect. The notification operation includes, for example, emitting a warning sound from a speaker or turning on or flashing a light source such as an LED. Alternatively, the processor 11 may cause the notification device 19, which is an example of a display, to display information indicating the inspection result, including the location of the defect in the analysis image.

[0053] The processor 11 may be configured to determine that the workpiece 90 contains defects if the area of ​​the region estimated to be defective is greater than or equal to a predetermined threshold. For example, the processor 11 may be configured to determine that the workpiece 90 contains defects if it determines that the ratio of the area of ​​the region estimated to be defective to the area of ​​white pixels in the mask image M, calculated based on the mask information, is greater than or equal to a predetermined threshold.

[0054] When the processor 11 executes step S10, it saves information indicating the inspection result to the storage device 12 (S11). The processor 11 also executes step S11 if it determines in step S9 that no defects are present (S9: No). The information stored in the storage device 12 is used, for example, to realize traceability. For example, the information stored in the storage device 12 is used to detect whether the state of the workpiece 90 has been altered during distribution due to damage, tampering, soiling, etc., and if so, to detect the extent of the alteration.

[0055] Even if the processor 11 determines in step S9 that there are no defects in the joint, it may display information indicating the inspection results, including an analysis image showing that there are no defects, on a display, which is an example of a notification device 19, before or after step S11.

[0056] [1-2-2. Determination of Analysis Conditions] The following describes the process by which the processor 11 determines the analysis conditions using the time series data shown in Figures 4 and 5, with reference to Figures 9 and 10A to 10B. Figure 9 is a flowchart illustrating the process of determining the analysis conditions by the inspection device 10. Figure 10A is a graph showing an example of time series data obtained by calculating the time average of the time derivatives for each of the time series data of temperature change shown in Figures 4 and 5. Figure 10B is a graph showing an example of time series data showing the difference between the two time series data shown in Figure 10A. The vertical axis of Figure 10A represents the time derivative of temperature (°C / s). The horizontal axis of Figure 10A represents time (s). The vertical axis of Figure 10B represents the time derivative of temperature (°C / s). The horizontal axis of Figure 10B represents time (s).

[0057] First, the processor 11 starts heating the workpiece 90 with the excitation source 18 and taking images of the workpiece 90 with the infrared camera 17 (S21). Then, the processor 11 acquires multiple temperature images generated by the infrared camera 17 taking time-series images of the workpiece 90 as temperature image data via the interface 13 (S22).

[0058] Next, the processor 11 performs predetermined data processing on the time-series data of points included in the junction region from the acquired temperature image data (S23). The processor 11 also performs predetermined data processing on the time-series data of points included in a region other than the junction region from the acquired temperature image data (S24). In Embodiment 1, for example, the processor 11 differentiates the time-series data of temperature change for point P1 shown in Figure 6A with respect to time and calculates its time average. The processor 11 differentiates the time-series data of temperature change for point P2 shown in Figure 6D with respect to time and calculates its time average.

[0059] Line a1 in Figure 10A represents the time series data obtained by differentiating the time series data of temperature change shown in Figure 4 and calculating the time average. In other words, line a1 represents the time series data obtained by calculating the moving average of the time derivative of the time series data of temperature change for point P1, which is included in the junction region. Line a2 represents the time series data obtained by differentiating the time series data of temperature change shown in Figure 5 and calculating the time average. In other words, line a2 represents the time series data obtained by calculating the moving average of the time derivative of the time series data of temperature change for point P2, which is included in a region different from the junction region.

[0060] In Embodiment 1, the processor 11 first calculates the derivative value obtained by differentiating the time series data shown in Figure 4 over time. Then, for each point in time, the processor 11 calculates the average value of the derivative value of the data over a predetermined time width as a moving average value for point P1. In Embodiment 1, the processor 11 calculates the moving average value using the average value of the derivative values ​​for 15 data points as the predetermined time width. The processor 11 similarly calculates the moving average value for point P2. By calculating the moving average value of the time derivative for the time series data of a specific point in the temperature image, the processor 11 can specifically grasp the increase or decrease in temperature at that specific point in the time series data. Comparing line a1 and line a2, it can be seen that there is a difference between the trend of temperature change at point P1 and the trend of temperature change at point P2 over a certain period after heating the workpiece 90.

[0061] Next, the processor 11 calculates the difference between the two calculated time series data (S25). Line b shown in Figure 10B shows the time series data obtained by subtracting the time series data shown on line a1 from the time series data shown on line a2. From line b, it can be seen that there is a difference between the temperature change trend at point P1 and the temperature change trend at point P2 during a certain period after heating the workpiece 90, for example, during period T4 shown in Figure 10B. Period T4 represents the period from time t1, when the difference value in the time series data of the difference value represented by line b becomes 0 or less, to time t2, when the difference value becomes 0 or greater.

[0062] The processor 11 determines the start time t1 and end time t2 of period T4 as analysis conditions (S26). Then, the processor 11 stores the determined analysis conditions in the storage device 12 (S27).

[0063] In this operation, the processor 11 can determine the analysis conditions using the acquired temperature images. The processor 11 may also determine the analysis conditions in advance using the acquired temperature images. The processor 11 may be configured to determine the analysis conditions using the acquired temperature images at the start of the inspection. In this case, the processor 11 may be configured to execute step S2 without executing step S1 shown in Figure 3, determine the analysis conditions based on the acquired data, and then execute subsequent processing.

[0064] [1-3. Effects, etc.] The inspection device 10 according to Embodiment 1 of this disclosure can achieve the following effects.

[0065] The inspection device 10 includes an input unit 13 that acquires temperature image data generated by an imaging device 17 capturing images of an object 90 joined in a predetermined joining area in a time series, and a processor 11 that inspects the joining state of the object 90 based on the temperature image data. The processor 11 calculates a correlation value between time series data for a reference point on the temperature image data and time series data for a predetermined point on the temperature image other than the reference point, within a predetermined period. The processor 11 then inspects the joining state based on the correlation values ​​calculated for each of the multiple points other than the reference point on the temperature image.

[0066] With this configuration, the processor 11 of the inspection device 10 can calculate a correlation value between the time-series data of a reference point and the time-series data of other points, based on the temperature image data captured for the object 90. By calculating the correlation value for a predetermined period set according to the conditions, the processor 11 can determine whether the temperature change of the reference point and the temperature change of the predetermined point are similar based on the correlation value. For example, if the reference point is included in a joining region, the processor 11 can estimate that the predetermined point for which a high correlation value is calculated is in a joined region, and estimate that the predetermined point for which a low correlation value is calculated is in a non-joined region. Also, if the reference point is included in a region different from the joining region, the processor 11 can estimate that the predetermined point for which a high correlation value is calculated is in a non-joined region, and estimate that the predetermined point for which a low correlation value is calculated is in a joined region. In this way, the processor 11 can calculate the similarity between the time-series data of two points as a correlation value and inspect the joining state based on this correlation value. Therefore, the inspection device 10 according to this disclosure can efficiently inspect the joining state of an object 90 joined in a predetermined joining area, and the accuracy of the inspection of the joining state can be improved.

[0067] Furthermore, the processor 11 generates an analysis image using correlation values ​​for each of the multiple points and inspects the joint state based on the analysis image. By operating in this manner, the inspection device 10 can generate an analysis image with a high contrast ratio between the brightness values ​​based on the correlation values ​​calculated for points corresponding to the joint region and the brightness values ​​based on the correlation values ​​calculated for regions other than the joint region. Since the processor 11 can inspect the joint state using an analysis image with a high contrast ratio, the accuracy of the inspection may be improved. In addition, since the analysis image is generated by visualizing the difference in temperature over time between the welded and unwelded parts of the object 90, it contains information that allows for visual identification of the size and position of the welded and unwelded parts of the object 90. Therefore, the processor 11 can detect this information through image processing, which may improve the accuracy of the inspection. Furthermore, since the user can visually distinguish between the welded and unwelded parts, it becomes easier to detect, for example, defects in the joint. For example, if an unwelded part is included in the joint region, the user or the processor 11 can estimate that the unwelded part is a defect in the joint. Thus, the inspection device 10 according to this disclosure makes it possible to efficiently inspect the bonding state of objects 90 that have been bonded in a predetermined bonding area.

[0068] Furthermore, the processor 11 determines a predetermined period based on the time-series data of a first point included in the bonding region and the time-series data of a second point included in a region different from the bonding region. By operating in this manner, the processor 11 can determine a period during which an analysis image suitable for inspecting the bonding state can be generated based on the acquired temperature image data. By generating an analysis image based on the determined period, the processor 11 can generate an analysis image with a high contrast ratio. Therefore, the inspection apparatus 10 according to this disclosure can efficiently inspect the bonding state of an object 90 that has been bonded in a predetermined bonding region.

[0069] Furthermore, the processor 11 calculates a moving average of a predetermined time width for the derivative of the time series data of the first point as the first time series data, and calculates a moving average of a predetermined time width for the derivative of the time series data of the second point as the second time series data. The processor 11 determines a predetermined period using the third time series data calculated from the difference between the first time series data and the second time series data. By operating in this manner, the processor 11 can automatically determine a period for which an analysis image suitable for inspecting the bonding state can be generated based on the acquired temperature image data. By generating an analysis image based on the determined period, the processor 11 can generate an analysis image with a high contrast ratio. Therefore, according to the inspection device 10 of this disclosure, the bonding state of an object 90 bonded in a predetermined bonding region can be efficiently inspected.

[0070] Furthermore, in the inspection device 10, the correlation value is determined using the zero-mean normalized cross-correlation (ZNCC) method. By determining the correlation value using the ZNCC method, the processor 11 normalizes the temperature waveform to a zero mean and calculates the correlation value. The processor 11 can eliminate the influence of the overall rise or fall in temperature, making it easy to compare waveforms with similar trends or patterns. In addition, the processor 11 can reduce the influence of noise contained in the time-series data of temperature changes, making it possible to calculate a correlation value based on the temperature waveform related to the junction. Therefore, even if a predetermined offset is included as noise in the time-series data of temperature changes, the processor 11 can calculate a correlation value based on the temperature waveform related to the junction. Moreover, by using the ZNCC method, the processor 11 can clearly compare each data even if the time-series data of temperature changes is obtained using different criteria. Therefore, the processor 11 can easily evaluate the correlation between data obtained from different sensors or under different conditions. Furthermore, while some matching methods may be greatly affected by outliers when calculating the correlation value, the ZNCC method calculates the correlation value by zero averaging. Therefore, by using the ZNCC method, the processor 11 can reduce the influence of outliers and calculate stable correlation values. Furthermore, since the correlation values ​​calculated using the ZNCC method are normalized to a range of -1 to 1, users can intuitively grasp the similarity of temperature changes. In addition, the processor 11 can easily determine the similarity based on the quantitatively calculated correlation values. Thus, by using the ZNCC method, the processor 11 can easily compare time-series data of temperature changes between different conditions.

[0071] Furthermore, the processor 11 determines whether the correlation value or brightness value for each of the multiple pixels located within the bonding region in the analyzed image satisfies predetermined conditions. The processor 11 estimates pixels that do not satisfy the predetermined conditions as bonding defects and estimates the presence or absence of defects within the bonding region as an inspection of the bonding state. By operating in this manner, the processor 11 can estimate the region corresponding to pixels associated with correlation values ​​or brightness values ​​that do not satisfy the predetermined conditions as an unbonded region. Also, the processor 11 can estimate the region corresponding to pixels associated with correlation values ​​or brightness values ​​that satisfy the predetermined conditions as a bonded region. In this way, the processor 11 can estimate the presence or absence of defects in the bonding for a predetermined pixel based on the correlation value calculated as the similarity between two time-series data points, or based on the brightness value corresponding to the correlation value. Therefore, the inspection device 10 according to this disclosure allows for easy estimation of the presence or absence of defects, and thus enables non-destructive and efficient inspection of the bonding state of an object 90 bonded in a predetermined bonding region.

[0072] The inspection method relating to this disclosure is a method for inspecting the bonding state of an object, which is executed by a processor 11 based on temperature image data generated by photographing an object bonded in a predetermined bonding region in a time series, acquired from a photography device 17. The inspection method includes calculating a correlation value between time series data for a reference point on the temperature image data indicated by the temperature image data within a predetermined period and time series data for a predetermined point on the temperature image other than the reference point. The inspection method also includes inspecting the bonding state based on the correlation values ​​calculated for each of the multiple points other than the reference point on the temperature image.

[0073] According to this method, the processor 11 can calculate the similarity between two time-series data points as a correlation value and inspect the bonding state based on this correlation value. Therefore, according to the inspection method of this disclosure, the bonding state of an object 90 bonded in a predetermined bonding region can be efficiently inspected.

[0074] (Embodiment 2) [2-1. Structure] An overview of the inspection device 10 according to Embodiment 2 will be described. The inspection device 10 according to Embodiment 2 can be configured in the same way as the inspection device 10 according to Embodiment 1.

[0075] [2-2. Operation] [2-2-1. Overall Operation] Generally, the joint strength of a workpiece is determined using destructive testing, which measures the force required to break the joint portion of the workpiece. However, destructive testing has the problem that the workpiece is destroyed for testing, making it difficult to reuse. In contrast, the inspection device 10 according to this embodiment can non-destructively predict the joint strength of the workpiece 90, that is, the joint strength indicating the strength of the bond between the permeable material 91 and the absorbent material 92.

[0076] Figure 11 is a flowchart illustrating the operation of the inspection device 10 according to Embodiment 2. Steps S1 to S5 in Figure 11 are the same as steps S1 to S5 in Figure 3.

[0077] When the processor 11 acquires an analysis image, it performs an effective area ratio calculation process (S31) to calculate the ratio of the area of ​​the effective bonding region to the area of ​​the bonding region of the workpiece 90 (hereinafter also referred to as the "effective area ratio"). The effective area ratio is calculated, for example, by dividing the area of ​​the effective bonding region (A1) in the analysis image by the total area of ​​the bonding region (A2).

[0078] In this specification, the "effective bonding area" refers to the area within the bonding area where the correlation value or brightness value is within a predetermined range. The predetermined range for the correlation value or brightness value is set according to the performance required for the workpiece 90 or the product using the workpiece 90. Details of the effective area ratio calculation process S31 will be described later.

[0079] Following step S31, the processor 11 predicts the bonding strength based on the effective area ratio obtained in the effective area ratio calculation process S31 (S32).

[0080] Then, the processor 11 determines whether the bond strength predicted in step S32 is below a threshold (S33). If the bond strength is below the threshold (S33: Yes), the processor 11 causes the notification device 19 to perform a notification operation (S34). This allows the user to know that the bond strength is below the threshold. The notification operation includes, for example, emitting a warning sound from a speaker or turning on or flashing a light source such as an LED. Alternatively, the processor 11 may display correlation information indicating the analysis results such as the bond strength and the determination result on a display, which is an example of the notification device 19.

[0081] Following step S34, the processor 11 saves information indicating the analysis results, such as the bonding strength and the judgment result, to the storage device 12 (S35). The processor 11 also executes step S35 if it determines in step S33 that the bonding strength is equal to or greater than the threshold (S33: No).

[0082] Even if the processor 11 determines in step S33 that the bonding strength is equal to or greater than the threshold, it may display information indicating the bonding strength, determination result, and other analysis results on a display, which is an example of a notification device 19, before or after step S35.

[0083] [2-2-2. Calculation process for effective area ratio] Figure 12 is a flowchart detailing the effective area ratio calculation process S31 shown in Figure 11. Figure 13 is a schematic diagram including a set of images to illustrate the effective area ratio calculation process S31.

[0084] As shown in Figure 12, in the effective area ratio calculation process S31, the processor 11 first obtains bonding area mask information from the storage device 12 (S41). The mask information is information that indicates the portion of the analysis image corresponding to the bonding area of ​​the workpiece 90.

[0085] Referring to Figure 13, an example of the process for generating such a mask image M will be explained. The processor 11 converts the analyzed image 20a into a grayscale image having brightness values ​​from 0 to 255, and segments the grayscale image to create a segmented image 31 by segmenting the portion corresponding to the joining region. The processor 11 may also be configured to generate the segmented image 31 using the analyzed image 20a. Next, the processor 11 generates a mask image M by setting the portion corresponding to the joining region of the segmented image 31 to white (pixel value 1) and the other portions to black (pixel value 0). The processor 11 stores the generated mask image M in the storage device 12.

[0086] Returning to Figure 12, after step S41, the processor 11 generates a joined region image 32 by excluding areas other than the joined region from the grayscale image (S42). For example, as shown in Figure 13, the processor 11 generates the joined region image 32 by setting the parts of the segment image 31 other than the joined region to black (pixel value 0). Alternatively, the processor 11 may generate the joined region image 32 by applying a masking process to the grayscale image using a mask image M. The processor 11 may also be configured to generate a segment image using the analyzed image and then execute the processing from step S42 onward.

[0087] Returning to Figure 12, after step S42, the processor 11 performs a binarization process on the joined region image 32 and generates an effective joined region image 33 in which pixels in the joined region image 32 whose brightness values ​​satisfy a predetermined condition are set to white (pixel value 1) (S43). In Embodiment 2, the processor 11 changes pixels in the joined region image 32 whose brightness values ​​are within a predetermined range to white. The effective joined region image 33 is an image showing the effective joined region. The predetermined range of brightness values ​​can be set by the user or the processor 11. In the example of Figure 13, the processor 11 changes pixels in the joined region image 32 with brightness values ​​above a predetermined threshold to white and changes pixels with brightness values ​​below the predetermined threshold to black to generate the effective joined region image 33. The processor 11 may be configured to change each pixel to white or black based on whether the correlation value corresponding to each pixel in the joined region image is within a predetermined range.

[0088] Following step S43, the processor 11 calculates the number of white pixels in the effective junction region image 33 as the area of ​​the effective junction region (S44).

[0089] The processor 11 calculates the area of ​​the joining region from the joining region mask information obtained in step S41 (S45). For example, the processor 11 calculates the area of ​​the joining region as the number of white pixels in the mask image M. Step S45 may be executed after step S41, and may be executed before step S42, unlike the example shown in Figure 12. Alternatively, since step S41 may be executed before step S45, step S41 may be executed after step S44, unlike the example shown in Figure 12.

[0090] The processor 11 calculates the ratio of the area of ​​the effective bonding region to the area of ​​the bonding region as the effective area ratio (S46). For example, the processor 11 calculates the effective area ratio (A1 / A2) by dividing the area of ​​the effective bonding region (A1) calculated in step S44 by the area of ​​the bonding region (A2) calculated in step S45. Alternatively, the processor 11 may calculate the effective area ratio as the ratio of the number of white pixels in the effective bonding region image 33 to the number of white pixels in the mask image M obtained in step S45.

[0091] After completing step S46, processor 11 completes the effective area ratio calculation process S31 and proceeds to step S32 in Figure 11.

[0092] [2-3. Effects, etc.] The inspection device 10 according to Embodiment 2 of this disclosure can achieve the following effects.

[0093] The inspection device 10 includes an input unit 13 that acquires temperature image data generated by a camera 17 that photographs an object 90 joined in a predetermined joining region in a time series, and a processor 11 that inspects the joining state of the object 90 based on the temperature image data. The processor 11 calculates a correlation value between time series data for a reference point on the temperature image data and time series data for a predetermined point on the temperature image other than the reference point, within a predetermined period. The processor 11 then inspects the joining state based on the correlation values ​​calculated for each of the multiple points other than the reference point on the temperature image. The processor 11 generates an analysis image using the correlation values ​​for each of the multiple points and inspects the joining state based on the analysis image. The processor 11 calculates an effective area ratio, which is the ratio of the area of ​​the effective joining region to the area of ​​the joining region, where the correlation value or brightness value for each of the multiple pixels in the analysis image satisfies predetermined conditions. The processor 11 estimates the joining strength of the object as an inspection of the joining state based on the calculated effective area ratio.

[0094] By operating in this manner, the processor 11 can easily calculate the area of ​​the effective bonding region based on correlation values ​​or brightness values ​​that satisfy predetermined conditions. Therefore, the processor 11 can calculate the ratio of the area of ​​the effective bonding region to the area of ​​the bonding region, and thus can easily estimate the bonding strength. Consequently, the processor 11 can non-destructively and efficiently predict the bonding strength of the object 90 based on temperature image data.

[0095] Furthermore, the processor 11 calculates the effective area ratio by calculating the ratio of the number of pixels in the effective bonding area to the number of pixels in the bonding area in the analyzed image. By operating in this way, the processor 11 can easily calculate the effective area ratio, and therefore can non-destructively and efficiently predict the bonding strength of the object 90 based on the temperature image data.

[0096] Furthermore, in the inspection device 10, the predetermined conditions include that when the reference point is included in the bonding region, the correlation value or brightness value for each of the multiple pixels is equal to or greater than the first threshold. The predetermined conditions also include that when the reference point is included in a region other than the bonding region, the correlation value or brightness value for each of the multiple pixels is equal to or less than the second threshold. With this configuration, the processor 11 can easily calculate the area included in the effective bonding region, whether the reference point used to calculate the correlation value is included in the bonding region or in a region other than the bonding region. Therefore, the processor 11 can non-destructively and efficiently predict the bonding strength of the object 90 based on the temperature image data.

[0097] (modified version) In the embodiment described above, the processor 11 generates an analysis image using the ZNCC method to calculate the similarity of time series data between two points. However, the evaluation method used to generate the analysis image is not limited to the ZNCC method. For example, the processor 11 may be configured to calculate the similarity between the time series data of a reference point and the time series data of other points as a correlation value using a matching method such as the difference square sum (SSD) method, the difference absolute sum (SAD) method, or the normalized cross-correlation (NCC) method. The processor 11 can use any of these matching methods to calculate correlation values ​​for each of the multiple points other than the reference point and generate an analysis image using these correlation values.

[0098] In the embodiment described above, the processor 11 generates an analysis image using the calculated correlation value and inspects the bonding state based on the analysis image. However, the inspection by the processor 11 is not limited to a method using an analysis image. For example, the processor 11 may inspect the bonding state of an object without generating an analysis image. The processor 11 may be configured to inspect the bonding state by associating the position on the temperature image with the correlation value and storing it in the storage device 12, determining whether the position associated with the correlation value is included in the bonding region, and determining whether the correlation value satisfies predetermined conditions. Thus, the processor 11 may be configured to inspect the bonding state based on a plurality of calculated correlation values ​​without creating an image.

[0099] In the embodiment described above, the processor 11 performs either a defect in the joint or a prediction of the joint strength. However, the processor 11 may be configured to perform either or both of the following: a defect in the joint and a prediction of the joint strength. For example, the processor 11 may be configured to decide in step S1 to perform either or both of the above processes as analysis conditions. When performing both of the above processes, the processor 11 may be configured to perform steps S6 to S11 and steps S31 to S35 sequentially after the execution of steps S1 to S6, or to perform them simultaneously.

[0100] In the above-described embodiment, the processor 11 determines in step S8 or S43 that a predetermined condition is met if the correlation value or brightness value of each pixel is equal to or greater than a predetermined threshold, but the predetermined condition is not limited thereto. For example, if the reference point used when calculating the correlation value is located in a region different from the joining region, the processor 11 may be configured to determine that a predetermined condition is met if the correlation value or brightness value of each pixel is less than or equal to a predetermined threshold.

[0101] (Summary of characteristics) As is clear from the above description, this disclosure includes the following embodiments. In the following, reference numerals are enclosed in parentheses solely to indicate their correspondence with the embodiments.

[0102] (Aspect 1) The inspection device (10) relating to this disclosure is An input unit (13) acquires temperature image data generated by a camera (17) that photographs an object (90) joined in a predetermined joining area in a time series, The system includes a processor (11) that inspects the bonding state of the object based on the temperature image data, The aforementioned processor, The correlation value is calculated between the time-series data for a reference point on the temperature image data shown in the temperature image data within a predetermined period and the time-series data for a predetermined point on the temperature image other than the reference point. The bonding state is inspected based on the correlation value calculated for each of the multiple points other than the reference point on the temperature image.

[0103] (Aspect 2) In the inspection apparatus (10) of Aspect 1, the processor (11) is An analysis image is generated using the correlation values ​​for each of the aforementioned points. The bonding state may be inspected based on the aforementioned analysis image.

[0104] (Aspect 3) In the inspection apparatus (10) of Aspect 2, the processor (11) may determine the predetermined period based on the time series data of a first point included in the junction region and the time series data of a second point included in a region different from the junction region.

[0105] (Aspect 4) In the inspection apparatus (10) of Aspect 3, the processor (11) is The first time series data is calculated as the first time series data by taking the moving average over a predetermined time range of the derivative of the time series data of the first point. The moving average of the predetermined time range for the derivative of the time series data of the second point is calculated as the second time series data. The predetermined period may be determined using a third time series data calculated from the difference between the first time series data and the second time series data.

[0106] (Aspect 5) In the inspection apparatus (10) of any of aspects 2 to 4, the correlation value may be determined using the zero-mean normalized cross-correlation method.

[0107] (Aspect 6) In the inspection apparatus (10) of any of aspects 2 to 5, the processor (11) is In the aforementioned analysis image, it is determined whether the correlation value or brightness value for each of the multiple pixels located within the joining region satisfies predetermined conditions. Pixels that do not satisfy the aforementioned predetermined conditions are presumed to be joining defects. As part of the inspection of the bonding state, the presence or absence of defects within the bonding region may be estimated.

[0108] (Aspect 7) In the inspection apparatus (10) of any of aspects 2 to 5, the processor (11) is The effective area ratio is calculated, which represents the ratio of the area of ​​the effective bonding region where the correlation value or brightness value for each of the multiple pixels in the analyzed image satisfies a predetermined condition, to the area of ​​the bonding region. Based on the calculated effective area ratio, the bonding strength of the object may be estimated as an inspection of the bonding state.

[0109] (Aspect 8) In the inspection apparatus (10) of Aspect 7, the processor (11) may calculate the effective area ratio by calculating the ratio of the number of pixels in the effective bonding area to the number of pixels in the bonding area in the analysis image.

[0110] (Aspect 9) In the inspection apparatus (10) of Aspect 7 or Aspect 8, the predetermined conditions are: If the reference point is included in the joining region, the correlation value or brightness value for each of the plurality of pixels is equal to or greater than the first threshold, If the reference point is located in a region different from the joining region, the correlation value or brightness value for each of the multiple pixels may be less than or equal to the second threshold.

[0111] (Aspect 10) An inspection method relating to the present disclosure is an inspection method for inspecting the bonding state of an object, which is executed by a processor (11) based on temperature image data generated by taking time-series images of an object (90) that has been bonded in a predetermined bonding area, acquired from a camera (17), To calculate the correlation value between time-series data for a reference point on the temperature image data shown in the temperature image data within a predetermined period and time-series data for a predetermined point on the temperature image other than the reference point, The bonding state is inspected based on the correlation value calculated for each of the multiple points other than the reference point on the temperature image, Includes.

[0112] In this specification, terms such as “First,” “Second,” etc., are used for illustrative purposes only and should not be understood as expressing or implying relative importance or ranking of technical features. Features designated as “First” or “Second” express or imply that they include one or more such features.

[0113] The inspection apparatus and inspection method described in this disclosure are realized through the cooperation of hardware resources, such as a processor and memory, and software (computer programs). [Industrial applicability]

[0114] This disclosure provides an inspection device and inspection method capable of inspecting the bonding state of an object based on temperature image data of the object, and is therefore suitably applicable in this type of industrial field. [Explanation of Symbols]

[0115] 1. Inspection System 10 Inspection equipment 11 processors 12 Storage device 13 Interfaces 15 Control Box 16 Power supply 17 Infrared Camera 18 Excitation source 19. Notification device 20a~20d Analyzed Images 20e Phase Image 90 Work 91 Transparent material 92 Absorbent material 95 Joint area

Claims

1. An input unit that acquires temperature image data generated by a camera that photographs objects joined in a predetermined bonding area over time, The system includes a processor that inspects the bonding state of the object based on the temperature image data, The aforementioned processor, The correlation value is calculated between the time-series data for a reference point on the temperature image data shown in the temperature image data within a predetermined period and the time-series data for a predetermined point on the temperature image other than the reference point. The bonding state is inspected based on the correlation value calculated for each of the multiple points other than the reference point on the temperature image. Inspection device.

2. The aforementioned processor, An analysis image is generated using the correlation values ​​for each of the aforementioned points. The bonding state is inspected based on the aforementioned analysis image. The inspection apparatus according to claim 1.

3. The inspection apparatus according to claim 2, wherein the processor determines the predetermined period based on time-series data of a first point included in the junction region and time-series data of a second point included in a region different from the junction region.

4. The aforementioned processor, The first time series data is calculated as the first time series data by taking the moving average over a predetermined time range of the value obtained by differentiating the time series data of the first point. The moving average of the predetermined time range for the derivative of the time series data of the second point is calculated as the second time series data. The predetermined period is determined using a third time series data calculated from the difference between the first time series data and the second time series data. The inspection apparatus according to claim 3.

5. The inspection apparatus according to claim 2, wherein the correlation value is determined using the zero-mean normalized cross-correlation method.

6. The aforementioned processor, In the aforementioned analysis image, it is determined whether the correlation value or brightness value for each of the multiple pixels located within the joining region satisfies predetermined conditions. Pixels that do not satisfy the aforementioned predetermined conditions are presumed to be joining defects. As an inspection of the bonding state, the presence or absence of the defect within the bonding region is estimated. The inspection apparatus according to any one of claims 2 to 5.

7. The aforementioned processor, The effective area ratio is calculated, which represents the ratio of the area of ​​the effective bonding region where the correlation value or brightness value for each of the multiple pixels in the analyzed image satisfies a predetermined condition, to the area of ​​the bonding region. Based on the calculated effective area ratio, the bonding strength of the object is estimated as an inspection of the bonding state. The inspection apparatus according to any one of claims 2 to 5.

8. The inspection apparatus according to claim 7, wherein the processor calculates the effective area ratio by calculating the ratio of the number of pixels in the effective bonding region to the number of pixels in the bonding region in the analyzed image.

9. The aforementioned predetermined conditions are: If the reference point is included in the joining region, the correlation value or brightness value for each of the plurality of pixels is equal to or greater than the first threshold, If the reference point is located in a region different from the joining region, the correlation value or brightness value for each of the plurality of pixels is less than or equal to the second threshold, The inspection apparatus according to claim 7.

10. An inspection method for inspecting the bonding state of an object, which is performed by a processor based on temperature image data generated by photographing an object bonded in a predetermined bonding region in a time series, acquired from a photography device, To calculate the correlation value between time-series data for a reference point on the temperature image data shown in the temperature image data within a predetermined period and time-series data for a predetermined point on the temperature image other than the reference point, The bonding state is inspected based on the correlation value calculated for each of the multiple points other than the reference point on the temperature image, Testing methods, including those mentioned above.

Images