WO2024043006A1

WO2024043006A1 - Inspection system, inspection device, and inspection method

Info

Publication number: WO2024043006A1
Application number: PCT/JP2023/027859
Authority: WO
Inventors: 庸介入江
Original assignee: パナソニックＩｐマネジメント株式会社
Priority date: 2022-08-26
Filing date: 2023-07-28
Publication date: 2024-02-29

Abstract

This inspection device, which is used in an inspection system provided with an infrared camera, comprises: an input circuit for receiving image data; and a computation circuit for analyzing the image data. Using infrared rays emitted from a welding object after welding, the infrared camera acquires a plurality of temperature images each indicating the temperature of the welding object, and generates the image data which represents the plurality of temperature images. The plurality of temperature images are captured in time series in accordance with temperature changes in the welding object, and are reflective of changes in thermal conduction due to the internal structure of the welding object. The computation circuit: calculates an analysis image pertaining to the internal structure of the welding object on the basis of the image data; and, according to a difference in emissivity from respective surface parts of the welding object and on the basis of the contrast in at least one temperature image, extracts, in the analysis image, an inspection region corresponding to a welding region formed on the welding object by welding, and outputs the extracted inspection region.

Description

Inspection system, inspection device and inspection method

The present disclosure relates to an inspection system, an inspection device, and an inspection method for inspecting a workpiece after welding.

Patent Document 1 discloses a method of inspecting a laser welded part that is welded by irradiating a laser onto the surface of an object to be inspected after welding. In the inspection method of Patent Document 1, after welding, a laser welded part that has been cooled to ambient temperature is imaged with an infrared camera, and the difference in brightness in the obtained infrared image is used as an index representing the difference in infrared emissivity of the surface of the object to be inspected. The method includes a first process of detecting a weld area, and a second process of determining the quality of welding based on the brightness value of an infrared image within the weld area.

Patent No. 4140218

The present disclosure provides an inspection system, an inspection device, and an inspection method that can accurately and efficiently inspect a welded object after welding.

One aspect of the present disclosure provides an inspection device used in an inspection system equipped with an infrared camera. The infrared camera captures a plurality of temperature images, each of which indicates the temperature of the welded object, using infrared rays emitted from the welded object after welding, and generates image data representing the plurality of temperature images. The plurality of temperature images are taken in time series according to changes in the temperature of the object to be welded, and reflect changes in heat conduction due to the internal structure of the object to be welded. The inspection device includes an input circuit that receives image data and an arithmetic circuit that analyzes the image data. The arithmetic circuit calculates an analytical image regarding the internal structure of the workpiece based on the image data, and calculates an analysis image based on the emissivity difference from each part of the workpiece's surface in at least one of the plurality of temperature images. Based on the contrast, an inspection area corresponding to a welding area formed on the object to be welded by welding is extracted from the analysis image, and the extracted inspection area is output.

One aspect of the present disclosure provides an inspection method that includes the steps of input circuitry of a computer receiving image data generated by an infrared camera, and arithmetic circuitry of the computer analyzing the image data. The infrared camera captures a plurality of temperature images, each of which indicates the temperature of the welded object, using infrared rays emitted from the welded object after welding, and generates image data representing the plurality of temperature images. The plurality of temperature images are taken in time series according to changes in the temperature of the object to be welded, and reflect changes in heat conduction due to the internal structure of the object to be welded. The arithmetic circuit calculates an analytical image regarding the internal structure of the workpiece based on the image data, and calculates an analysis image based on the emissivity difference from each part of the workpiece's surface in at least one of the plurality of temperature images. Based on the contrast, an inspection area corresponding to a welding area formed on the workpiece by welding is extracted from the analysis image, and the extracted inspection area is output.

According to the inspection system, inspection device, and inspection method according to the present disclosure, it is possible to accurately and efficiently inspect a welded object after welding.

Diagram for explaining the configuration of an inspection system according to Embodiment 1 of the present disclosure A block diagram showing an example of the configuration of an inspection device in the inspection system of Embodiment 1. Flowchart illustrating the operation of the inspection device according to the first embodiment Diagram illustrating a temperature image captured by an infrared camera of the inspection system Diagram for explaining temporal changes in measured temperature in temperature images Diagram illustrating phase images before and after extracting the inspection area with the inspection device Flowchart illustrating inspection area extraction processing by the inspection apparatus of Embodiment 1 Diagram for explaining a temperature difference image in the inspection device of Embodiment 1 A diagram for explaining inspection area extraction processing in the inspection apparatus of Embodiment 1. Diagram for explaining an example of defect detection by the inspection system A diagram for explaining a temperature difference image in an inspection apparatus according to a modification of Embodiment 1.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings as appropriate. However, more detailed explanation than necessary may be omitted. For example, detailed explanations of well-known matters or redundant explanations of substantially the same configurations may be omitted. This is to avoid unnecessary redundancy in the following description and to facilitate understanding by those skilled in the art. The applicant provides the accompanying drawings and the following description to enable those skilled in the art to fully understand the present disclosure, and they are not intended to limit the subject matter recited in the claims. do not have.

(Aspects of this disclosure)
An inspection device according to a first aspect of the present disclosure is an inspection device used in an inspection system equipped with an infrared camera. The infrared camera captures a plurality of temperature images, each of which indicates the temperature of the welded object, using infrared rays emitted from the welded object after welding, and generates image data representing the plurality of temperature images. The plurality of temperature images are taken in time series according to changes in the temperature of the object to be welded, and reflect changes in heat conduction due to the internal structure of the object to be welded. The inspection device includes an input circuit that receives the image data, and an arithmetic circuit that analyzes the image data. The arithmetic circuit calculates an analytical image regarding the internal structure of the workpiece based on the image data, and calculates radiation from each part of the surface of the workpiece in at least one temperature image of the plurality of temperature images. An inspection area corresponding to a welding area formed on the object to be welded by welding is extracted from the analysis image based on a contrast according to the rate difference, and the extracted inspection area is output.

According to a second aspect of the present disclosure, in the inspection apparatus according to the first aspect, the calculation circuit corresponds to at least a peak temperature at which the temperature of the welded object is at a peak among the plurality of temperature images. The contrast is calculated from the first image.

According to a third aspect of the present disclosure, in the inspection apparatus according to the second aspect, the arithmetic circuit is configured to lower the temperature than the peak temperature captured before or after the first image in the plurality of temperature images. A third image showing a difference between a second image corresponding to temperature and the first image is calculated, and the contrast is calculated from the third image.

According to a fourth aspect of the present disclosure, in the inspection device according to any one of the first to third aspects, the arithmetic circuit performs Fourier transformation on the plurality of temperature images to calculate the analysis image. .

According to a fifth aspect of the present disclosure, in the inspection apparatus according to the fourth aspect, the analysis image is calculated by the Fourier transform based on a phase or amplitude defined according to a temperature change of the workpiece. be done.

According to a sixth aspect of the present disclosure, in the inspection apparatus according to any one of the first to fifth aspects, the arithmetic circuit performs an analysis of the workpiece based on the extracted analysis image of the inspection area. Detect internal defects.

According to a seventh aspect of the present disclosure, the inspection apparatus according to any one of the first to sixth aspects includes a recording medium for recording information, and the arithmetic circuit is configured to analyze an analysis image from which the inspection area has been extracted. Image data is output to the recording medium.

According to an eighth aspect of the present disclosure, the inspection apparatus according to any one of the first to seventh aspects includes an output circuit connected to an external device, and the arithmetic circuit is configured to control the inspection area by the output circuit. The image data of the extracted analysis image is output to the external device.

An inspection system according to a ninth aspect of the present disclosure includes an infrared camera that generates image data by capturing a temperature image indicating the temperature of the welded object using infrared rays emitted from the welded object after welding; to the inspection device according to any one of the eighth aspects.

According to a tenth aspect of the present disclosure, in the inspection system according to the ninth aspect, an excitation source that emits excitation energy toward the workpiece after welding to heat the workpiece, or an excitation source that heats the workpiece after welding; The infrared camera further includes a cooling device that cools the object to be welded, and the infrared camera captures the plurality of temperature images according to a temperature change due to heating or cooling of the object to be welded.

An inspection method according to an eleventh aspect of the present disclosure includes a step in which an input circuit of a computer receives image data generated by an infrared camera, and a step in which an arithmetic circuit of the computer analyzes the image data. The infrared camera captures a plurality of temperature images, each of which indicates the temperature of the welded object, using infrared rays emitted from the welded object after welding, and generates image data representing the plurality of temperature images. The plurality of temperature images are taken in time series according to changes in the temperature of the object to be welded, and reflect changes in heat conduction due to the internal structure of the object to be welded. The arithmetic circuit calculates an analytical image regarding the internal structure of the workpiece based on the image data, and calculates radiation from each part of the surface of the workpiece in at least one temperature image of the plurality of temperature images. An inspection area corresponding to a welding area formed on the object to be welded by welding is extracted from the analysis image based on a contrast according to the rate difference, and the extracted inspection area is output.

A program according to a twelfth aspect of the present disclosure is a program for causing the arithmetic circuit to execute the inspection method according to the eleventh aspect.

(Embodiment 1)
The inspection system and inspection device according to Embodiment 1 will be described below.

1. Configuration 1-1. Configuration of Inspection System FIG. 1 is a diagram for explaining the configuration of an inspection system 1 according to Embodiment 1 of the present disclosure. The inspection system 1 of this embodiment is applied to inspect the workpiece 11, which is an object to be welded, in-line or offline after welding the workpiece 11. In the inspection by the present system 1, the inspection device 20 detects internal defects 12 such as voids that may occur inside the workpiece 11 using, for example, active thermography. The active thermography method is a non-destructive inspection method in which the temperature change of the object is observed by excitation such as applying heat to the object to be inspected, and the internal structure of the object is estimated from the temperature change.

The inspection system 1 of FIG. 1 includes two excitation sources 18 and an infrared camera 17, for example for active thermography, a power supply 16 for supplying power to each excitation source 18 and infrared camera 17, and a control for controlling the operation of the power supply 16. It includes a box 15 and an inspection device 20. In this system 1, the control box 15 is connected to the power source 16 and the inspection device 20 so as to control the power supply to the excitation source 18 and the like by the power source 16 based on, for example, a control signal from the inspection device 20.

The infrared camera 17 of the present system 1 is arranged so that, for example, the welding area 30 to be welded on the workpiece 11 can be seen. The infrared camera 17 repeats an imaging operation at a predetermined period, for example, and generates image data representing a captured image. The infrared camera 17 is connected to, for example, the inspection device 20 so as to transmit image data. For example, the inspection device 20 controls the imaging operation of the infrared camera 17, collects and analyzes the captured image data.

The infrared camera 17 includes an infrared sensor that detects infrared rays having a wavelength of 3 μm to 15 μm, for example. The excitation source 18 is, for example, a halogen lamp, a xenon lamp, a laser light source, a vibrator that generates ultrasonic waves, or a coil that generates electromagnetic induction, but is not limited to these, and may be any configuration that emits energy. Further, when the excitation source 18 is a halogen lamp, a xenon lamp, or a laser light source, the excitation source 18 may pulse-heat the workpiece 11 by emitting flash light. The wavelength of the light emitted by the excitation source 18 may be different from the wavelength of the infrared rays detected by the infrared camera 17 as long as it is a wavelength that is highly absorbed by the object to be welded. Although two excitation sources 18 are illustrated in FIG. 1, the number of excitation sources 18 is not limited to two, and may be one or three or more.

1-2. Configuration of Inspection Device The configuration of the inspection device 20 in the present system 1 will be explained using FIG. 2. FIG. 2 is a block diagram showing a configuration example of the inspection device 20 in the inspection system 1 of this embodiment.

The inspection device 20 is configured as, for example, various types of computers. The inspection device 20 in FIG. 2 includes a CPU 21, a storage device 22, an input interface 23, and an output interface 24. Hereinafter, the interface will be abbreviated as "I/F".

The CPU 21 realizes the functions of the inspection device 20 by performing information processing in collaboration with software, for example. Such information processing is realized, for example, by the CPU 21 operating in accordance with instructions from the control program 25 stored in the storage device 22. For example, the CPU 21 may include an internal memory as a temporary storage area that holds various data and programs. The CPU 21 is an example of an arithmetic circuit in the inspection device of the present disclosure.

The arithmetic circuit of the inspection device 20 may include a circuit that performs arithmetic operations for information processing, and is not limited to a CPU. For example, the arithmetic circuit may be a hardware circuit such as a dedicated electronic circuit or a reconfigurable electronic circuit designed to realize various functions in the inspection device 20, or may be a circuit such as an MPU or an FPGA. may be configured.

The storage device 22 is a recording medium that records various information including programs and data such as the control program 25 necessary to realize the functions of the inspection device 20. The storage device 22 is realized by, for example, a flash memory, a semiconductor storage device such as a solid state drive (SSD), a magnetic storage device such as a hard disk drive (HDD), and other recording media alone or in combination. The storage device 22 may include volatile memory such as SRAM and DRAM. The above program may be provided via a communication network such as the Internet, or may be stored in a portable recording medium.

The input I/F 23 is an interface circuit that connects the inspection device 20 with external equipment such as the infrared camera 17 in order to input information such as image data from the infrared camera 17 into the inspection device 20. The input I/F 23 may be an input terminal that receives image data from the infrared camera 17, for example. The input I/F 23 may be a communication circuit that performs data communication according to a predetermined wired communication standard or wireless communication standard. The predetermined communication standards include IEEE802.3, USB, HDMI (registered trademark), IEEE802.11, IEEE1394, WiFi (registered trademark), Bluetooth (registered trademark), and the like.

The output I/F 24 is an interface circuit that connects the inspection device 20 and external equipment in order to output information such as control signals from the inspection device 20. Such external devices include, for example, the control box 15, the infrared camera 17, other information processing terminals such as a server device, and other output devices such as a display. For example, like the input I/F 23, the output I/F 24 may be a communication circuit that performs data communication according to a predetermined wired communication standard or wireless communication standard. The output I/F 24 may be, for example, various signal lines, output terminals, or connection terminals that output information to external equipment. Further, the input I/F 23 and the output I/F 24 may be realized by similar hardware, or may be an integrated input/output I/F.

2. Operation The operation of the inspection system 1 configured as described above will be explained below.

For example, as shown in FIG. 1, this system 1 excites the workpiece 11 after welding with an excitation source 18 and measures the temperature change of the workpiece 11 with an infrared camera 17. The infrared camera 17 detects infrared rays emitted from the workpiece 11, captures temperature images indicating the temperature of the workpiece 11 at a period such as a predetermined frame rate, and generates image data indicating each temperature image. The infrared camera 17 sequentially transmits such temperature image data to the inspection device 20, for example. The inspection device 20 receives temperature image data via, for example, the input I/F 23 and stores it in the storage device 22 .

The inspection device 20 of the present system 1 detects internal defects 12 in the workpiece 11 by analyzing temperature image data over a predetermined period as an inspection of the workpiece 11, and determines whether the welding quality is good or bad according to the detection result. Determine. In the following, an example will be described in which the workpiece 11 is inspected by pulse thermography in which the excitation source 18 is used to pulse-heat the workpiece 11. In the following explanation, in the thickness direction of the work 11 corresponding to the vertical direction in FIG. is sometimes called downward.

2-1. Overall Operation of Inspection Apparatus The overall operation of the inspection apparatus 20 in the inspection system 1 of this embodiment will be explained using FIGS. 3 to 6.

FIG. 3 is a flowchart illustrating the operation of the inspection device 20 according to this embodiment. The flowchart in FIG. 3 is started, for example, in synchronization with the timing of pulse heating by the excitation source 18. Each process shown in this flowchart is executed by, for example, the CPU 21 of the inspection device 20.

First, the CPU 21 acquires temperature image data of the workpiece 11 captured by the infrared camera 17, for example, from the storage device 22 (S1). FIG. 4 is a diagram illustrating a temperature image captured by the infrared camera 17 of the inspection system 1. FIG. 4A shows a temperature image N21 of the workpiece 11 at room temperature before being heated by the excitation source 18. FIG. 4(B) shows a temperature image N1 at a peak temperature at which the temperature in the temperature image becomes the highest due to heating while measuring the temperature change of the workpiece 11. FIG. 4(C) shows a temperature image N22 of the workpiece 11 in which the temperature in the temperature image decreases after heating and is in a normal temperature state, for example, as before heating.

FIG. 5 is a diagram for explaining temporal changes in measured temperature in a temperature image. In FIG. 5, the vertical axis shows the temperature (in degrees Celsius) measured by the infrared camera 17, and the horizontal axis shows time (in seconds). Further, in FIG. 5, the dotted line indicates a temperature change in a portion of the surface of the workpiece 11 where the emissivity is relatively low and the temperature measured by the infrared camera 17 is difficult to rise due to heating (#1). On the other hand, a solid line indicates a temperature change in a portion of the surface of the workpiece 11 that has a higher emissivity than the above-mentioned portion and whose measured temperature is likely to rise due to heating (#2). Emissivity indicates the ratio of the measured temperature in the temperature image to the actual value when the temperature of the surface of the workpiece 11 is actually measured, and in the range of "0" to "1", the higher the value, the more infrared rays are emitted. shows.

In the example of FIG. 4(A), the pre-heating temperature image N21 is captured at time t21, one frame before the start of heating by the excitation source 18. In the example of FIG. 4(B), the temperature image N1 of the peak temperature is captured at time t1 when the measured temperature is the highest during the measurement period. In the example of FIG. 4C, the temperature image N22 after heating is captured at time t22, 100 frames after the start of heating. In each of the temperature images N21, N1, and N22 in FIGS. 4A to 4C, the level of the measured temperature is expressed by the level of brightness, and the brighter the image, the higher the measured temperature.

As shown in FIGS. 4A to 4C, in comparison with the temperature images N21 and N22 before and after heating, the welding area 30 on the workpiece 11 appears brighter than other areas in the temperature image N1 at the peak temperature, The boundaries between the welding region 30 and other regions can be easily observed. A possible reason for this is that the welding region 30 undergoes a thermal history due to heating during welding, causing changes in shape and physical properties, making it easier for infrared rays to be emitted.

Returning to FIG. 3, the CPU 21 sets analysis conditions indicating conditions for analyzing the temperature image data (S2). The analysis conditions include, for example, a frame range of a temperature image used for analysis by the inspection device 20, an analysis frequency used for calculating a phase image, which will be described later.

The frame range of the analysis conditions is set, for example, to correspond to a predetermined period long enough to allow observation of temperature changes in the workpiece 11 due to pulsed heating from the excitation source 18. The analysis frequency is set to a frequency that is less than or equal to 1/2 of the sampling frequency (Nyquist frequency: a frequency that can be correctly detected by Fourier transform), and is set to a frequency that is equal to or lower than 1/2 of the sampling frequency (Nyquist frequency: a frequency that can be correctly detected by Fourier transform), and is set at a depth where internal defects 12, etc. of the workpiece 11, which is assumed to be the object to be inspected, exists. It will be set accordingly. For example, when detecting an internal defect 12 on the upper side (closer to the surface) of the workpiece 11, a higher frequency is used than when detecting an internal defect 12 on the lower side (farther from the surface). .

In step S2, the CPU 21 sets analysis conditions based on user input values input from an external input device via the input I/F 23, for example. Alternatively, the CPU 21 may set the analysis conditions based on setting values stored in advance in the storage device 22 or the like.

The CPU 21 analyzes the temperature image data in the frame range of the set analysis conditions (S2) using discrete Fourier transform (S3). For example, the CPU 21 calculates a complex function in the frequency domain using a discrete Fourier transform from a time function indicating a temperature change at each part of the surface of the workpiece 11 corresponding to each pixel of the temperature image. The analysis result of such a discrete Fourier transform of the temperature image data includes, for example, a phase and an amplitude defined by a complex function regarding the frequency, depending on the temperature change of the workpiece 11 over the frame range of the analysis conditions.

In this embodiment, the CPU 21 calculates the phase value at the analysis frequency of the set analysis condition from the complex function corresponding to each pixel of the temperature image based on the analysis result of step S3, and converts each phase value into a pixel value. A phase image is calculated (S4). The phase image is an example of an analysis image in the inspection device 20 of this embodiment, and for example, the difference in heat propagation depending on the internal structure of each part of the workpiece 11 is reflected as a phase difference between pixels for a period of the frame range. obtain.

After calculating the phase image (S4), the CPU 21 executes filter processing on the calculated phase image using, for example, various image processing filters (S5). Various image processing filters include, for example, a high-pass filter that emphasizes edges of an image, a Gaussian filter that smoothes an image, and the like. Through such filter processing, for example, edge enhancement and noise reduction can be performed in the phase image, thereby making it easier to perform the processing related to inspection using the phase image (S6 to S8). The CPU 21 stores, for example, a phase image to which filter processing has been applied, in an internal memory or the like.

FIG. 6 is a diagram illustrating phase images before and after the inspection region is extracted by the inspection device 20. The inspection area indicates an area corresponding to the welding area 30 on the workpiece 11 in the phase image, and is used for detecting an internal defect 12, which will be described later. FIG. 6(A) illustrates the phase image P1 after filter processing is performed (S5) and before the inspection region is extracted. In the example of FIG. 6(A), an internal defect 12 inside the workpiece 11 appears as a darker area than the surrounding area in a region corresponding to the welding region 30 in the phase image P1.

In the inspection apparatus 20 of this embodiment, the CPU 21 performs a process of extracting an inspection area corresponding to the welding area 30, for example, in the phase image P1 after filter processing (S5) (S6). FIG. 6(B) illustrates a phase image P1 in which the inspection region 30p has been extracted by such inspection region extraction processing (S6). For example, as shown in FIG. 6(B), in the extracted phase image P1, the background portion other than the inspection area 30p is masked. For example, after extracting the inspection region 30p, the CPU 21 may hold the analysis results of the temperature image such as the phase image P1 in an internal memory or the like. Details of the inspection area extraction process (S6) will be described later.

Next, the CPU 21 performs internal defect 12 detection processing, for example, based on the pixel value of the inspection area 30p in the phase image P1 (S7). In this embodiment, the CPU 21 determines, for example, the size and per unit area of the region corresponding to the internal defect 12 in the inspection region 30p in the phase image P1 after extracting the inspection region 30p as shown in FIG. 6(B). Detect the number of . Such defect detection processing (S7) is executed using, for example, various types of machine learning.

For example, before executing the process of this flowchart, welding is performed multiple times, and after each welding, multiple temperature images are sequentially captured. From a plurality of temperature images obtained in each welding process, for example, similarly to steps S3 to S6 above, an inspection area is extracted in the calculated phase image, and the size and unit of the area corresponding to the internal defect 12 within the inspection area are determined. Learning data annotated with the number of pieces per area is created. Then, based on the learning data, machine learning of the regression model is executed so as to input the pixel values of the inspection area extracted from the phase image and output the size and number of the internal defect 12 area. . In the defect detection process (S7), internal defects 12 can be detected using such a learned regression model.

After performing defect detection processing (S7), the CPU 21 determines whether the welding quality is good or bad based on the detection results (S8). For example, the CPU 21 compares the size of the region detected corresponding to the internal defect 12 and the number of defects per unit area with predetermined threshold values to determine whether the welding quality is good or not, that is, the workpiece 11 after welding. Determine whether the product is good or defective. For example, it may be determined that the product is non-defective when both the size and the number of detected regions are greater than or equal to each threshold value, and it may be determined that the product is defective when either of the detected regions is less than or equal to the threshold value. The predetermined threshold value can be set as appropriate in accordance with the required level of processing accuracy in welding the workpiece 11, taking into account, for example, the rate of overlooking and over-detecting defective products.

Based on the welding quality determination result, the CPU 21 determines whether the welding quality is poor (S9). If the welding quality is poor (YES in S9), the CPU 21 notifies, for example, an external device of the poor welding quality. For example, the CPU 21 outputs, via the output I/F 24, a signal notifying a defective welding quality to a control device of a production line including the welding process of the workpiece 11, or a terminal owned by the manager of the production line. The defect may be notified to the device or the like. This makes it possible to take measures, such as stopping the production line or excluding defective products, depending on the determination result of defective welding.

After notifying the welding quality defect (S10), the CPU 21 stores, for example, the analysis result held in the internal memory and the determination result of step S8 in the storage device 22 or the like (S11).

Furthermore, if the welding quality is not poor (NO in S9), the CPU 21 does not particularly execute the process of step S10, and stores, for example, the analysis result and the determination result (S11).

After that, the CPU 21 ends the processing of this flowchart. For example, in an in-line inspection using pulse thermography, the process shown in FIG. 3 may be started again in accordance with the timing of pulse heating the next workpiece 11.

According to the above processing, a phase image P1 as shown in FIG. The corresponding inspection area 30p is extracted (S6). Then, internal defects 12 are detected in the inspection area 30p from the phase image P1 after extracting the inspection area 30p (see FIG. 6(B)) (S7), and the quality of welding is determined based on the detection result. (S8).

As described above, by extracting the inspection area 30p (S6), for example, in the defect detection process (S7), internal defects are detected based only on the pixel values of the inspection area 30p corresponding to the welding area 30 in the phase image P1. 12 detections can be performed. As a result, for example, noise etc. in the phase image P1 before extraction is reduced to improve detection accuracy, and the processing load and amount of memory used in the inspection device 20 are reduced, improving processing speed and inspection costs. It is possible to reduce the amount of Moreover, in this way, it is possible to easily detect internal defects 12 in the inspection area 30p, and improve the determination accuracy, processing efficiency, etc. in the welding quality determination (S8) using the detection results. Thereby, the inspection device 20 can accurately and efficiently inspect the workpiece 11 after welding.

In the above process, an example was explained in which temperature image data is acquired from the storage device 22 of the inspection device 20 (S1). The temperature image data may be stored not only in the storage device 22 but also in, for example, an external server device. In this case, in step S1, the CPU 21 may acquire temperature image data from the server device or the like, for example, via the input I/F 23.

In the above processing, an example has been described in which a phase image is calculated from the analysis result of temperature image data (S4), and the processing from step S5 onwards is performed on the phase image. In step S4, the CPU 21 may calculate an amplitude image not only based on the phase image but also based on the amplitude of the complex function based on the analysis result. The processing after step S5 may be performed using an amplitude image instead of the phase image, or may be performed using both the phase image and the amplitude image.

Furthermore, in the above processing, an example has been described in which the CPU 21 executes the defect detection processing (S7) and the welding quality determination (S8). These processes are not limited to the above example, and the internal defect 12 may be observed, for example, by visual inspection of the welding region 30 of the phase image P1, and the quality of the welding may be determined based on the observation result. In this case, the CPU 21 may output, for example, the phase image P1 after extracting the inspection region 30p to an external output device or the like through the output I/F 24, and display the phase image P1.

2-2. Inspection Area Extraction Process The details of the inspection area extraction process in step S6 in FIG. 3 will be described with reference to FIGS. 7 to 10.

FIG. 7 is a flowchart illustrating the inspection area extraction process (S6) by the inspection device 20 of this embodiment. The process shown in the flowchart of FIG. 7 is started, for example, with the temperature image data acquired in step S1 of FIG. 3 and the phase image P1 after filter processing (S5) being held in the internal memory of the CPU 21, etc. .

First, the CPU 21 generates a temperature image N1 at a peak temperature (also referred to as a "peak temperature image N1"), for example, as shown in FIG. ) is acquired (S21).

Next, the CPU 21 acquires image data of a temperature image (also referred to as "low temperature image N2") at a temperature lower than the peak temperature (for example, room temperature) from the temperature image data corresponding to the temperature change (S22). The CPU 21 acquires, for example, a temperature image N21 of one frame before heating as shown in FIG. 4(A) as the low temperature image N2.

The CPU 21 calculates a temperature difference image N3 based on the image data of the peak temperature image N1 and the low temperature image N2 acquired in steps S21 and S22 (S23). The CPU 21 calculates a temperature difference image N3 by calculating the difference in pixel values (that is, brightness values) at corresponding positions between the peak temperature image N1 and the low temperature image N2. In the temperature difference image N3, for example, between the temperature images N1 and N2, a background region corresponding to a region where the temperature change on the workpiece 11 due to thermal excitation is smaller than that in the welding region 30, that is, a background component is suppressed.

FIG. 8 is a diagram for explaining a temperature difference image in the inspection device 20 of this embodiment. FIG. 8A illustrates a temperature difference image N3 calculated by subtracting each pixel value of the temperature image N21 before heating as the low temperature image N2 from each pixel value of the peak temperature image N1. In the temperature difference image N3 of FIG. 8(A), the contrast is emphasized and the boundary between the area corresponding to the welding area 30 and other areas becomes more clear than, for example, the peak temperature image N1 of FIG. 4(B). ing.

Further, FIG. 8B shows, as another example of calculating a temperature difference image, a temperature difference image N3b calculated by subtracting each pixel value of the peak temperature image N1 from each pixel value of the temperature image N21 before heating. is illustrated using a temperature scale different from that in FIG. 8(A). Also in the temperature difference image N3b of FIG. 8(B), the boundary between the area corresponding to the welding area 30 and other areas can be clearly observed, similar to the temperature difference image N3 of FIG. 8(A), for example. .

For example, after calculating the temperature difference image N3 (S23), the CPU 21 performs a binarization process on the temperature difference image N3 based on the contrast due to the emissivity difference in each part of the surface of the workpiece 11 (S24). The CPU 21 executes the binarization process using, for example, a brightness value histogram according to the emissivity of the temperature difference image N3. At this time, the CPU 21 sets pixels whose pixel values are greater than or equal to a predetermined threshold value to be white, and sets pixels whose pixel values are less than the threshold value to black. The predetermined threshold value is set using, for example, Otsu's binarization method, which is also known as the discriminant analysis method.

FIG. 9 is a diagram for explaining the inspection area extraction process in the inspection device 20 of this embodiment. FIG. 9(A) illustrates a temperature difference image N3 similar to FIG. 8(A) as a temperature image before performing the binarization process. FIG. 9(B) illustrates a binarized image M3 obtained by applying binarization processing to the temperature difference image N3 of FIG. 9(A). In the binarized image M3 of FIG. 9(B), a region 30m corresponding to the welding region 30 is set to white.

Based on the calculated binarized image M3, the CPU 21 creates a region mask for extracting the inspection region 30p from the phase image P1 as shown in FIG. 6(A) (S25). For example, in the binarized image M3 of FIG. 9(B), the CPU 21 sets the pixel value of an area 30m set to white as "1", with the position where the pixel value changes as a boundary, and the area set to black. A region mask having the pixel value of "0" may be created.

After creating the region mask (S25), the CPU 21 combines the region mask with the phase image P1 and extracts the inspection region 30p (S26). For example, the CPU 21 may multiply each pixel value of the region mask created from the above-described binarized image M3 by each pixel value at the corresponding position of the phase image P1. As a result, in the phase image P1, the pixel value at the position corresponding to the area 30m is not changed, while the pixel value in the background area other than the area becomes zero, and for example, the welding area as shown in FIG. A phase image P1 is obtained by extracting the inspection region 30p corresponding to 30.

For example, the CPU 21 outputs image data indicating the phase image P1 from which the inspection area 30p is extracted to the internal memory or the like as the extraction result of the inspection area 30p (S27). Note that the CPU 21 may output the extraction result of the inspection area 30p not only to the internal memory but also to an external output device of the inspection apparatus 20, for example, via the output I/F 24.

For example, after outputting the extracted inspection area 30p (S27), the CPU 21 ends the process of this flowchart and returns to step S7 in FIG. 3.

According to the above processing, a region mask is created based on the contrast according to the emissivity difference on the workpiece 11 in the temperature image (S24, S25), and the region mask is used for inspection according to the welding region 30 in the phase image P1. Region 30p is extracted (S26). As a result, for example, in processing such as defect detection by analyzing a temperature image (S3 to S7), the temperature image is used not only to calculate the phase image P1 (S4) but also to create a region mask. The inspection area 30p for performing defect detection processing (S7) can be extracted.

FIG. 10 is a diagram for explaining an example of defect detection by the inspection system 1. FIG. 10 shows an example in which defect detection is performed in a phase image different from that in FIG. 6. FIG. 10A illustrates the phase image P10 calculated from the temperature image analysis result in step S4 of FIG. FIG. 10(B) illustrates a phase image P11 obtained by applying filter processing to the phase image P10 of FIG. 10(A) in step S5. FIG. 10C illustrates a phase image P11 in which the inspection region 30p is extracted by combining region masks from the state of FIG. 10B in the inspection region extraction process (S6). In the examples shown in FIGS. 10A to 10C, the dark-looking region in the phase images P10 and P11 corresponding to the welding region 30 corresponds to the internal defect 12.

In the phase image P11 of FIG. 10(B), the visibility of the area corresponding to the welding area 30 is improved from the phase image P10 of FIG. 10(A) by filtering such as edge enhancement and smoothing. Furthermore, according to the phase image P11 in FIG. 10(C), for example, the area of the internal defect 12 can be searched only in the extracted inspection area 30p of the phase image P11, making it easier to perform the defect detection process (S7). can do.

In the above process, an example in which the discriminant analysis method is used in the binarization process (S24) has been described. The binarization process in step S24 is not limited to the above example, and various methods such as the P-tile method, mode method, dynamic threshold determination method, level slice method, Laplacian histogram method, differential histogram method, etc. are used. It's okay.

(Modified example)
In the above example, the image data of the temperature image N21 before heating is acquired as the low temperature image N2 (S22), and is used to calculate the temperature difference image N3 (S23). The low temperature image N2 is not limited to the temperature image N21 before heating, but may also be acquired, for example, a temperature image N22 after heating as shown in FIG. 4(C). FIG. 11 is a diagram for explaining a temperature difference image in an inspection apparatus according to a modification of the present embodiment.

FIG. 11(A) illustrates a temperature difference image N3a calculated using the temperature image N22 after heating instead of the temperature image N21 before heating in calculating the temperature difference image N3 shown in FIG. 8(A). FIG. 11(B) illustrates a temperature difference image N3ab calculated using the temperature image N22 after heating instead of the temperature image N21 before heating in calculating the temperature difference image N3b in FIG. 8(B). As shown in FIGS. 11(A) and (B), even if the temperature image N22 after heating is used as the low temperature image N2, it is similar to the temperature difference images N3 and N3b in FIGS. 8(A) and (B), for example. In addition, temperature difference images N3a and N3ab with enhanced contrast can be obtained.

3. Effects, etc. As described above, in this embodiment, the inspection device 20 is used in the inspection system 1 equipped with the infrared camera 17. The infrared camera 17 captures a plurality of temperature images each indicating the temperature of the workpiece 11 using infrared rays emitted from the workpiece 11 after welding (an example of an object to be welded), and generates image data indicating the plurality of temperature images. generate. The plurality of temperature images are taken in time series according to changes in the temperature of the workpiece 11, and reflect changes in heat conduction due to the internal structure of the workpiece 11. The inspection device 20 includes an input I/F 23 (an example of an input circuit) that receives image data, and a CPU 21 (an example of an arithmetic circuit) that analyzes the image data. Based on the image data, the CPU 21 calculates a phase image P1 as an example of an analysis image regarding the internal structure of the workpiece 11 (S3, S4), and calculates each part of the surface of the workpiece 11 in at least one temperature image among the plurality of temperature images. The inspection area 30p corresponding to the welding area 30 formed on the workpiece 11 by welding is extracted from the phase image P1 based on the contrast according to the emissivity difference from (S6), and the extracted inspection area 30p is output ( S6, S27).

According to the inspection apparatus 20 described above, the phase image P1 regarding the internal structure of the work 11 is generated by analyzing a plurality of temperature images according to the temperature change of the work 11 (S3, S4), and the phase image P1 is generated using the temperature images. In this step, the inspection area 30p is extracted (S6). Thereby, it is possible to easily inspect the workpiece 11 after welding based on the inspection area 30p corresponding to the welding area 30. For example, in the phase image P1, the welding quality regarding the internal defect 12 etc. can be inspected from the extracted inspection area 30p, and the influence of the background area other than the inspection area 30p can be suppressed to accurately and accurately inspect the workpiece 11 after welding. It can be done efficiently.

In the present embodiment, the CPU 21 calculates the contrast from the peak temperature image N1 (an example of the first image) corresponding to the peak temperature at which the temperature of the workpiece 11 is at least the peak among the plurality of temperature images (S21 , S24). As a result, in the plurality of temperature images, the peak temperature image N1 tends to have a large contrast due to the difference in emissivity of each part of the surface of the workpiece 11, so it is possible to easily calculate the contrast used for, for example, binarization processing.

In the present embodiment, the CPU 21 selects a low temperature image N2 (second image) corresponding to a temperature lower than the peak temperature, which is captured before or after the peak temperature image N1 (an example of the first image) among the plurality of temperature images. A temperature difference image N3 (an example of a third image) is calculated (S23), and a contrast is calculated from the temperature difference image N3 (S24). As a result, as shown in FIG. 8A, for example, in the temperature difference image N3, the contrast between the area corresponding to the welding area 30 and the background area is emphasized, and a region mask is created by binarization processing (S24 to S25). ) can be made easier.

As the low-temperature image N2 used in calculating the temperature difference image N3 (S23) described above, either the temperature image N21 before heating or the temperature image N22 after heating can be used, depending on the imaging period of the plurality of temperature images, for example. It may also be determined whether the Further, calculation of the temperature difference image N3 is not limited to the example using the above-mentioned peak temperature image N1 and low temperature image N2, and it is sufficient to obtain a difference between an image with a relatively high temperature and an image with a relatively low temperature.

In the present embodiment, the CPU 21 performs discrete Fourier transform, which is an example of Fourier transform, on the plurality of temperature images (S3), and calculates a phase image P1, which is an example of an analysis image (S4). Thereby, for example, from a plurality of temperature images that reflect changes in heat conduction due to the internal structure of the workpiece 11 due to temperature changes, a phase image in which such changes in heat conduction are reflected as a phase change is obtained.

In this embodiment, the phase image P1, which is an example of an analysis image, is calculated based on the phase defined by a complex function based on the analysis results of a plurality of temperature images according to the temperature change of the workpiece 11 by discrete Fourier transform. (S4). Further, the amplitude image, which is another example of the analysis image, includes an amplitude defined by the complex function according to the temperature change of the workpiece 11. According to such a phase image or an amplitude image, information regarding the internal structure of the internal defect 12 or the like can be obtained depending on the temperature change of the workpiece 11, for example.

In the present embodiment, the CPU 21 detects an internal defect 12, which is an example of a defect occurring inside the workpiece 11, based on the pixel value of the inspection area 30p in the phase image P1, as an example of the extracted analysis image of the inspection area 30p. Detect (S7). Thereby, the internal defect 12 can be detected more accurately and efficiently than, for example, by using the entire phase image P1. Furthermore, based on the detection results of such internal defects 12, it is possible to accurately and efficiently determine whether the welding quality is good or bad (S8).

In this embodiment, the inspection device 20 includes an internal memory of the CPU 21 as an example of a recording medium for recording information, and the CPU 21 outputs image data of the phase image P1 from which the inspection region 30p has been extracted to the recording medium (S27 ). The CPU 21 may output and store the extraction result of the inspection area 30p not only in the internal memory but also in the storage device 22, for example. Furthermore, in this embodiment, the inspection device 20 includes an output I/F 24 (an example of an output circuit) that connects to an external device such as an output device such as a display. In step S27, the CPU 21 may output the image data of the phase image from which the inspection region 30p has been extracted to the external device using the output I/F 24. For example, the extraction result of the inspection area 30p may be output to an output device such as a display or an information processing terminal such as a server device as an external device.

In the present embodiment, the inspection system 1 includes an infrared camera 17 that generates image data by capturing a temperature image indicating the temperature of the work 11 using infrared rays emitted from the work 11 after welding, and an inspection device 20. In the inspection system 1 of this embodiment, the inspection device 20 controls the imaging operation of the infrared camera 17, for example, and acquires image data of a temperature image from the infrared camera 17. Note that the inspection device 20 is not limited to the above example, and may receive image data of a temperature image by, for example, communicating data with an infrared camera external to the inspection system 1.

In the present embodiment, the inspection system 1 further includes an excitation source 18 that heats the work 11 by emitting excitation energy toward the work 11 after welding. The infrared camera 17 captures a plurality of temperature images N1 and N2 according to temperature changes due to heating of the workpiece 11. Thereby, for example, temperature image data can be acquired by cooling the workpiece 11 after welding, heating it with the excitation source 18, and capturing the image with the infrared camera 17 (S1 in FIG. 3).

The inspection method in this embodiment includes a step in which the input I/F 23 (an example of an input circuit) of the inspection device 20, which is an example of a computer, receives image data generated by the infrared camera 17, and a step in which the CPU 21 (arithmetic circuit example) includes steps (S3 to S7) of analyzing image data. The infrared camera 17 captures a plurality of temperature images each indicating the temperature of the workpiece 11 using infrared rays emitted from the workpiece 11 after welding (an example of an object to be welded), and generates image data indicating the plurality of temperature images. generate. The plurality of temperature images are taken in time series according to changes in the temperature of the workpiece 11, and reflect changes in heat conduction due to the internal structure of the workpiece 11. The CPU 21 calculates a phase image P1 (an example of an analysis image) regarding the internal structure of the workpiece 11 based on the image data (S3, S4), and calculates a phase image P1 (an example of an analysis image) regarding the internal structure of the workpiece 11, and calculates a phase image P1 (an example of an analysis image) regarding the internal structure of the workpiece 11, and Based on the contrast according to the emissivity difference from each part, an inspection area 30p corresponding to the welding area 30 formed on the workpiece 11 by welding is extracted from the phase image P1 (S6), and the extracted inspection area 30p is output. (S6, S27).

In this embodiment, a control program 25 is provided as an example of a program for causing the CPU 21 to execute the above inspection method. According to the above inspection method and program, the work 11 after welding can be inspected accurately and efficiently.

(Other embodiments)
In the first embodiment described above, an example has been described in which the inspection system 1 performs inspection using pulse thermography. The inspection system 1 of this embodiment is not limited to this, and may be applied, for example, to an inspection using lock-in thermography that excites the workpiece 11 at a predetermined cycle.

In each of the above embodiments, a peak temperature image N1 and a low temperature image N2 are acquired (S21, S22), a temperature difference image N3 is calculated (S23), and the temperature difference image N3 is calculated based on the contrast due to the emissivity difference in the temperature difference image N3. An example of performing binarization processing (S24) has been described. The inspection area extraction process (S6) in this embodiment is not limited to the above example. For example, if the contrast in the peak temperature image N1 is large enough to allow binarization processing, the processes in steps S22 and S23 are omitted. It's okay. In this case, in step S24, binarization processing may be performed based on the contrast within the peak temperature image N1.

In each of the above embodiments, an example has been described in which a temperature image is binarized (S24) before creating a region mask (S25). In this embodiment, for example, if the contrast in the temperature image before applying the binarization process is large enough to create a region mask, the process in step S24 may be omitted, and the temperature image is calculated based on the contrast of the temperature image. A region mask may be created.

In each of the above embodiments, the inspection system 1 was described which inspects the workpiece 11 after welding by capturing a temperature image after heating by the excitation source 18 with the infrared camera 17. In the present embodiment, the inspection system may include a cooling device that cools the welded workpiece 11 instead of the excitation source 18, for example, when the temperature of the workpiece 11 that is the object to be inspected is high. The cooling device may be, for example, a device that sprays low-temperature gas onto the workpiece 11. In this case as well, for example, temperature images before and after cooling can be captured by the infrared camera 17, and the workpiece 11 can be inspected using the plurality of temperature images in the same manner as when using the excitation source 18 described above.

As described above, in each of the above embodiments, the inspection system includes the excitation source 18 that heats the workpiece 11 by emitting excitation energy toward the workpiece 11 after welding (an example of an object to be welded), or the workpiece after welding. The apparatus further includes a cooling device for cooling 11. The infrared camera 17 captures a plurality of temperature images according to temperature changes due to heating or cooling of the workpiece 11. Note that the inspection system is not limited to the above example, and the inspection system may not particularly include an excitation source or a cooling device. For example, instead of heating the workpiece 11 that has been naturally cooled after welding with an excitation source, the heat generated during welding may be used to obtain temperature image data captured by the infrared camera 17 immediately after welding the workpiece 11. Good too. In this case as well, the workpiece 11 is inspected in the same way as when using the excitation source 18 etc. using, for example, a temperature image immediately after welding and a temperature image after the workpiece 11 has naturally cooled down to the ambient temperature after welding. It can be carried out.

The present disclosure is applicable to an inspection system, an inspection device, and an inspection method for inspecting a welded object after welding, and is particularly applicable to inspecting a welded object by analyzing an image captured by an infrared camera. .

Claims

An inspection device used in an inspection system equipped with an infrared camera,
The infrared camera captures a plurality of temperature images each indicating the temperature of the welded object using infrared rays emitted from the welded object after welding, and generates image data indicating the plurality of temperature images,
The plurality of temperature images are taken in time series according to temperature changes of the welding object, and reflect changes in heat conduction due to the internal structure of the welding object,
The inspection device includes:
an input circuit that receives the image data;
and an arithmetic circuit that analyzes the image data,
The arithmetic circuit is
calculating an analysis image regarding the internal structure of the welded object based on the image data;
A welding area formed on the workpiece by welding in the analysis image based on a contrast according to the emissivity difference from each part of the surface of the workpiece in at least one temperature image of the plurality of temperature images. Extract the inspection area corresponding to
An inspection device that outputs the extracted inspection area.
The inspection device according to claim 1, wherein the arithmetic operation circuit calculates the contrast from at least a first image of the plurality of temperature images that corresponds to a peak temperature at which the temperature of the workpiece reaches a peak.
The arithmetic circuit is
A third image showing a difference between the first image and a second image taken before or after the first image in the plurality of temperature images and corresponding to a temperature lower than the peak temperature. Calculate,
The inspection device according to claim 2, wherein the contrast is calculated from the third image.
The inspection device according to claim 1, wherein the arithmetic circuit performs Fourier transform on the plurality of temperature images to calculate the analysis image.
The inspection device according to claim 4, wherein the analysis image is calculated by the Fourier transform based on a phase or an amplitude defined according to a temperature change of the workpiece.
The inspection device according to claim 1, wherein the arithmetic circuit detects a defect occurring inside the workpiece based on the extracted analysis image of the inspection area.
Equipped with a recording medium for recording information,
The inspection apparatus according to claim 1, wherein the arithmetic circuit outputs image data of an analysis image from which the inspection area has been extracted to the recording medium.
Equipped with an output circuit to connect with external equipment,
The inspection apparatus according to claim 1, wherein the arithmetic circuit outputs image data of an analysis image from which the inspection area has been extracted to the external device by the output circuit.
an infrared camera that generates image data by capturing a temperature image indicating the temperature of the welded object using infrared rays emitted from the welded object after welding;
An inspection system comprising the inspection device according to any one of claims 1 to 8.
Further comprising an excitation source that heats the welded object by emitting excitation energy toward the welded object after welding, or a cooling device that cools the welded object after welding,
The inspection system according to claim 9, wherein the infrared camera captures the plurality of temperature images according to temperature changes due to heating or cooling of the workpiece.
an input circuit of the computer receives image data generated by the infrared camera;
a step in which an arithmetic circuit of the computer analyzes the image data,
The infrared camera captures a plurality of temperature images each indicating the temperature of the welded object using infrared rays emitted from the welded object after welding, and generates image data indicating the plurality of temperature images,
The plurality of temperature images are taken in time series according to temperature changes of the welding object, and reflect changes in heat conduction due to the internal structure of the welding object,
The arithmetic circuit is
calculating an analysis image regarding the internal structure of the welded object based on the image data;
A welding area formed on the workpiece by welding in the analysis image based on a contrast according to the emissivity difference from each part of the surface of the workpiece in at least one temperature image of the plurality of temperature images. Extract the inspection area corresponding to
An inspection method that outputs the extracted inspection area.
A program for causing the arithmetic circuit to execute the inspection method according to claim 11.