X-ray inspection equipment

The X-ray inspection apparatus generates pseudo-defective images by adjusting pixel values based on X-ray attenuation rates to accurately depict virtual foreign objects, addressing reproduction challenges and reducing operator workload.

JP2026115582APending Publication Date: 2026-07-09ISHIDA CO LTD

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
ISHIDA CO LTD
Filing Date
2024-12-27
Publication Date
2026-07-09

AI Technical Summary

Technical Problem

Existing X-ray inspection systems face challenges in accurately reproducing a test piece with a virtual foreign object in a simulated defective image, leading to potential mixing of test pieces with actual products and increased operator workload.

Method used

An X-ray inspection apparatus that generates a pseudo-defective image by changing pixel values based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object, using factors like transmission distance, density, and mass absorption coefficient to accurately depict the foreign object in the image.

Benefits of technology

Enables accurate reproduction of virtual foreign objects, reducing the risk of mixing test pieces with products and minimizing operator workload by automating inspection accuracy checks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026115582000001_ABST
    Figure 2026115582000001_ABST
Patent Text Reader

Abstract

The present invention provides an X-ray inspection device that can accurately reproduce a test piece using a virtual foreign object in a simulated defective image. [Solution] The X-ray inspection apparatus 1 comprises a conveyor 5 for transporting articles, an X-ray irradiation unit 6 for irradiating the articles transported by the conveyor 5 with X-rays, an X-ray detection unit 7 for detecting X-rays, and a controller 10 that generates an inspection image including the articles from the X-ray detection results of the X-ray detection unit 7, inspects whether or not foreign objects are contained in the articles based on the inspection image, generates a pseudo-defective image containing a virtual foreign object by changing the pixel values ​​of some of the pixels constituting the inspection image, and checks the accuracy of the inspection based on the pseudo-defective image. The controller 10 changes the pixel values ​​based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an X-ray inspection apparatus.

Background Art

[0002] An X-ray inspection apparatus is known that includes a transport unit that transports an article, an irradiation unit that irradiates the article transported by the transport unit with X-rays, a sensor that detects the X-rays transmitted through the article, and a control unit that generates an inspection image (X-ray transmission image) from the X-rays detected by the sensor and performs inspection of the article based on the inspection image (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a production site where an X-ray inspection apparatus as described above is used, during normal operation on a production line, a test piece is inserted into the production line at random to check whether the target foreign object can be normally detected by the X-ray inspection apparatus, and the reliability of the inspection may be inspected. Preparing an article sample with a test piece attached and flowing the article sample through the production line to inspect the reliability of the inspection places a large workload on the operator. In addition, since it is an inspection performed during the production of the article, there is also a risk that the test piece will be mixed into the article as a foreign object. Therefore, in recent years, consideration has been given to generating a pseudo-defective image including a virtual foreign object and inspecting the reliability of the inspection based on the pseudo-defective image. However, when using, for example, a pre-photographed image of a foreign object or an image of a circle uniformly painted black as a virtual foreign object, the test piece may not be accurately reproduced as a virtual foreign object in the pseudo-defective image.

[0005] The present invention aims to provide an X-ray inspection apparatus that can accurately reproduce a test piece with a virtual foreign object in a simulated defective image. [Means for solving the problem]

[0006] (1) An X-ray inspection apparatus according to one aspect of the present invention comprises: a transport unit for transporting articles; an irradiation unit for irradiating articles transported by the transport unit with X-rays; a sensor unit for detecting X-rays; an image generation unit for generating an inspection image including the article from the X-ray detection result in the sensor unit; an inspection unit for inspecting whether or not the article contains foreign matter based on the inspection image; and an inspection unit for generating a pseudo-defective image containing a virtual foreign matter by changing the pixel values ​​of some of the pixels constituting the inspection image, and checking the accuracy of the inspection based on the pseudo-defective image, wherein the inspection unit changes the pixel values ​​based on the X-ray attenuation rate of the substance corresponding to the virtual foreign matter.

[0007] In an X-ray inspection apparatus according to one aspect of the present invention, the inspection unit generates a pseudo-defective image containing a virtual foreign object by changing the pixel values ​​of some of the pixels constituting the inspection image. The inspection unit changes the pixel values ​​based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object. As a result, the virtual foreign object is drawn in the pseudo-defective image according to the X-ray attenuation rate of the substance, and compared to cases where, for example, an image of a foreign object taken in advance is used as the virtual foreign object, or an image of a circle uniformly filled with black is used, the test piece can be reproduced with greater accuracy using the virtual foreign object in the pseudo-defective image.

[0008] (2) In (1) above, the inspection unit may change the pixel value based on at least one of the following: the X-ray transmission distance in the virtual foreign object, the density of the material, and the mass absorption coefficient of the material. In this case, at least one of the following is used: the transmission distance, the density of the material, and the mass absorption coefficient of the material, depending on the type and shape of the material designated as the virtual foreign object. This makes it possible to calculate the X-ray attenuation rate of the material in the pseudo-defective image with greater accuracy.

[0009] (3) In (2) above, the transmission distance may be the length of the portion of a hypothetical straight line connecting the irradiation unit and the sensor unit that passes through a hypothetical foreign object. In this case, the transmission distance can be calculated geometrically.

[0010] (4) In any one of (1) to (3) above, the sensor unit has a plurality of detection elements arranged along an intersection direction that is horizontal to the transport direction of the transport unit, and the inspection unit may change the pixel value based on the X-ray attenuation rate calculated in correspondence with each of the plurality of detection elements. In this case, since a virtual foreign object can be drawn with a resolution corresponding to the inspection image, it becomes easier to accurately reproduce the test piece with the virtual foreign object.

[0011] (5) In any one of (1) to (4) above, the inspection unit may store information on multiple virtual foreign objects that differ from each other in at least one of their size, shape, or density, and change the pixel value based on the X-ray attenuation rate of the material corresponding to the virtual foreign object selected from the multiple virtual foreign objects. In this case, an inspection can be performed by generating a pseudo-defect image corresponding to one of the multiple virtual foreign objects stored in advance.

[0012] (6) In any one of the above (1) to (5), if the inspection unit obtains an inspection result that no foreign matter is present based on the inspection of the simulated defective image, it may determine that there is a problem with the reliability of the inspection. In this case, it is possible to detect an anomaly in which a foreign matter should be determined to be present, but is instead determined to be absent.

[0013] (7) In any one of the above (1) to (6), the inspection unit may generate a false defect image and automatically perform an inspection when a predetermined time has elapsed or when a predetermined number of items have been inspected. In this case, the accuracy of the inspection will be automatically checked according to a predetermined rule. This will reduce the burden on the worker.

[0014] (8) In any one of (1) to (7) above, the inspection unit may extract the outline of the article included in the inspection image and generate a pseudo-defect image by changing the pixel values ​​of some of the pixels inside the extracted outline. In this case, the pseudo-defect image generated by the inspection unit will be similar to the inspection image obtained when a foreign object is mixed into an article being produced on the production line. As a result, the reliability of the inspection is checked based on an inspection image that is actually obtained when a foreign object is mixed into an article being produced on the production line, and thus the accuracy of the inspection can be improved.

[0015] (9) In any one of the above (1) to (8), if the inspection unit obtains an inspection result that no foreign matter is present by an inspection based on an inspection image that does not contain a hypothetical foreign matter during the execution of the inspection, it may control the sorting unit to sort the items in a direction different from the direction in which normal items that do not contain foreign matter are sorted by the sorting unit when the inspection is not being performed. If the inspection unit obtains an inspection result that no foreign matter is present by an inspection based on a pseudo-defective image during the execution of the inspection, it may stop the transport of items by the transport unit. In this case, normal items that have been determined not to contain foreign matter during the execution of the inspection can be sorted separately from normal items when the inspection is not being performed. Also, if it is determined that the inspection has not been performed correctly, the transport by the transport unit will be stopped, so that items containing foreign matter after the inspection are not processed as normal items. [Effects of the Invention]

[0016] According to one aspect of the present invention, a test piece can be accurately reproduced with a virtual foreign object in a simulated defective image. [Brief explanation of the drawing]

[0017] [Figure 1] This is a diagram showing the configuration of an X-ray inspection apparatus according to an embodiment. [Figure 2] Figure 1 is a diagram showing the internal configuration of the shield box. [Figure 3] Figure 1 is a block diagram showing the functional configuration of an X-ray inspection device. [Figure 4] It is a schematic diagram for explaining an outline of a method for generating a suspected defective image. [Figure 5] (a) is a schematic diagram illustrating a plurality of detection elements arranged along the crossing direction. (b) is a schematic diagram showing an enlargement of the plurality of detection elements in (a). [Figure 6] It is a schematic diagram showing an example of calculation of the X-ray transmission distance. [Figure 7] It is a schematic diagram for explaining a change in pixel value based on the X-ray attenuation rate of a substance corresponding to a virtual foreign object. [Figure 8] It is a flowchart exemplifying the operation of the X-ray inspection apparatus. [Figure 9] It is a flowchart exemplifying the process of intermediate inspection of the X-ray inspection apparatus. [Figure 10] (a) and (b) are diagrams for explaining problems of the conventional example. [Figure 11] It is a diagram for explaining problems of the conventional example.

Embodiments for Carrying Out the Invention

[0018] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or corresponding elements are denoted by the same reference numerals, and redundant descriptions are omitted.

[0019] As shown in FIGS. 1 to 3, the X-ray inspection apparatus 1 includes an apparatus main body 2, support legs 3, a shield box 4, a transport conveyor 5 (transport unit), an X-ray irradiation unit 6 (irradiation unit), an X-ray detection unit 7 (sensor unit), a display 8, a controller 10 (image generation unit, inspection unit, inspection unit), and a storage unit 10A (inspection unit). The X-ray inspection apparatus 1 acquires an inspection image IM of the article G while transporting the article G, and performs a foreign object inclusion inspection of the article G based on the inspection image IM.

[0020] Before inspection, items G are brought into the X-ray inspection device 1 by the input conveyor 9A, and after inspection, items G are discharged from the X-ray inspection device 1 by the output conveyor 9B. In normal (in-production) foreign object inspection, which is not an intermediate inspection (check) as described later, items G determined to be defective by the X-ray inspection device 1 are sorted out of the production line (outside the system) by a sorting device (sorting unit) 15 located downstream of the output conveyor 9B, while items G determined to be good by the X-ray inspection device 1 pass through the sorting device 15 as is.

[0021] The shield box 4 has an entrance 4a and an exit 4b. Before inspection, the items G are brought into the shield box 4 from the entrance conveyor 9A via the entrance 4a, and after inspection, the items G are brought out from inside the shield box 4 via the exit 4b to the exit conveyor 9B. The detection sensor 13 detects the items G being transported by the entrance conveyor 9A.

[0022] The conveyor belt 5 is located inside the shield box 4 and transports the items G along the transport direction D1 from the entrance 4a to the exit 4b. The X-ray irradiation unit 6 is located inside the shield box 4 and irradiates the items G being transported by the conveyor belt 5 with X-rays.

[0023] The X-ray detection unit 7 is located inside the shield box 4 and detects X-rays irradiated from the X-ray irradiation unit 6 that have passed through the article G and the conveyor belt 5. The X-ray detection unit 7 has a plurality of detection elements 7p arranged along an intersecting direction D2 that is horizontal to the conveying direction D1 of the conveyor belt 5. The X-ray detection unit 7 is configured as a line sensor having, for example, a plurality of photodiodes arranged in one dimension along a horizontal direction perpendicular to the conveying direction D1, and a scintillator positioned on the X-ray incident side for each photodiode. The electrical signal detected by the X-ray detection unit 7 is acquired by the controller 10.

[0024] The display 8 is located on the main body 2 of the device. The display 8 has a touch panel display screen and a speaker. The display 8 functions as an operation input unit that accepts input of various conditions via the display screen. The display 8 also functions as a display unit that displays inspection results of the X-ray inspection device 1 via the display screen.

[0025] The controller 10 is located within the main body 2 of the device and controls the operation of each part of the X-ray inspection apparatus 1. The controller 10 includes a processor such as a CPU, memory such as ROM and RAM, and storage such as an SSD. The controller 10 may be configured as hardware such as electronic circuits. The storage unit 10A is composed of either or a combination of HDD and flash memory. The storage unit 10A may be provided within the main body 2 of the device, or it may be provided so as to be able to communicate with the controller 10 via a network.

[0026] The controller 10 generates an inspection image including the item G from the X-ray detection result of the X-ray detection unit 7. The inspection image is an image having pixel values ​​such that the grayscale corresponds to the brightness of the X-rays that have passed through the item G.

[0027] The controller 10 generates a pseudo-defective image containing a virtual foreign object by changing the pixel values ​​of some of the pixels that make up the inspection image, and checks the accuracy of the inspection based on the pseudo-defective image. In a production site where the X-ray inspection device 1 is used, during the normal operation of the X-ray inspection device 1 to inspect item G during its production, intermediate inspections are performed to check the accuracy of the inspection, for example, at predetermined intervals, to confirm whether the target foreign object has been properly detected by the X-ray inspection device 1. The controller 10 generates a pseudo-defective image containing a virtual foreign object and automatically checks the accuracy of the inspection based on the pseudo-defective image, for example, at predetermined intervals during the normal operation of the X-ray inspection device 1.

[0028] Figure 10(a) shows an acrylic member AC with actual foreign objects F1 and F2 attached to its upper surface. The acrylic member AC has a portion AC1 with a thickness of TH1 and a portion AC2 with a thickness of TH2, which is greater than the thickness of TH1. The foreign objects F1 and F2 are, for example, stainless steel spheres, and the diameter of foreign object F2 is larger than the diameter of foreign object F1. The two foreign objects F1 are attached to portion AC1 and portion AC2, respectively. The one foreign object F2 is attached to the boundary between portion AC1 and portion AC2. Figure 10(b) shows the detection results of transmitted X-rays when X-rays are irradiated onto the acrylic member AC with foreign objects F1 and F2 attached from above as shown in Figure 10(a). As shown in Figure 10(b), the brightness of the transmitted X-rays differs between portion AC1 and portion AC2, with the brightness XAC2 corresponding to portion AC2 being lower than the brightness XAC1 corresponding to portion AC1. Furthermore, even if the foreign objects F1 have the same diameter, the brightness waveforms differ between brightness XF1 and brightness XF3. Here, there is a problem that if we use an image of a circle uniformly filled with black, as in Figure 10(b), we cannot reproduce the fact that the brightness waveforms of transmitted X-rays differ even for the same foreign object. For example, as in Figure 11, if we paste an image of a circle uniformly filled with black as a virtual foreign object at two locations with different thicknesses of an object, we cannot adequately reproduce the fact that the X-ray brightness would actually be different at those two locations.

[0029] Therefore, the controller 10 of this embodiment changes the pixel values ​​of some of the pixels that make up the inspection image IM generated when inspecting the item G, based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object. As an example, the controller 10 changes the pixel values ​​based on the X-ray attenuation rate calculated in relation to each of the multiple detection elements 7a, 7b, etc. of the X-ray detection unit 7.

[0030] Figure 4 is a schematic diagram illustrating the outline of the method for generating a pseudo-defective image. In the example in Figure 4, the virtual foreign object IF is a stainless steel sphere with a predetermined diameter. The controller 10 assumes a state in which the virtual foreign object IF is virtually irradiated with X-rays and estimates the brightness of the virtually transmitted X-rays. For example, the controller 10 estimates the brightness of the virtually transmitted X-rays by assuming that a real foreign object of the same shape and material as the virtual foreign object IF exists at the location of the virtual foreign object IF. As shown in Figures 5(a) and 5(a), when the virtual foreign object IF is virtually irradiated with X-rays, it can be assumed that the X-rays that have passed through the foreign object IF reach multiple detection elements 7p of the X-ray detection unit 7. It can be assumed that such a virtual X-ray transmission path XP exists for each of the multiple detection elements 7p.

[0031] Figure 4 shows a portion of the inspection image IM, which includes item G, generated from the detection results of X-rays transmitted through item G by the X-ray detection unit 7. The inspection image IM includes pixels Px1 and Px2 corresponding to multiple detection elements 7a and 7b of the X-ray detection unit 7. In Figure 4, the pixel values ​​A and B of pixels Px1 and Px2 are visually represented by the intensity of the shading within the area enclosed by the thick line for ease of explanation. In the example in Figure 4, the pixel values ​​A and B of pixels Px1 and Px2 are pixel values ​​that are not affected by the virtual foreign object IF. If, for example, a virtual foreign object IF is further placed on the path of X-rays transmitted through item G in the inspection image IM including item G in Figure 4, then transmission paths XPa and XPb can be assumed as transmission paths XP as shown in Figures 5(a) and 5(a). Here, transmission paths XPa and XPb refer to the paths of X-rays that pass through the virtual foreign object IF, pass through item G, and reach pixels Px1 and Px2, respectively. X-ray paths such as transmission paths XPa and XPb can be assumed to correspond to each of the detection elements 7p within a range corresponding to the position where the virtual foreign object IF is located. The controller 10 calculates the attenuation rate of the X-rays that have virtually passed through the virtual foreign object IF for each of the virtual transmission paths XPa and XPb corresponding to each of the multiple detection elements 7a, 7b, etc. of the X-ray detection unit 7, and changes the pixel value based on the attenuation rate of the X-rays.

[0032] The controller 10 may change the pixel value based on the X-ray transmission distance in the virtual foreign object IF, the density of the material corresponding to the virtual foreign object IF, and the mass absorption coefficient of the material corresponding to the virtual foreign object IF. The transmission distance may be the length of the portion of a virtual straight line connecting the X-ray irradiation unit 6 and the X-ray detection unit 7 that passes through the virtual foreign object IF. For example, as shown in Figure 6, the transmission distance may be the length L of the transmission path XP. If the virtual foreign object IF is spherical, the length L may be calculated geometrically based on the radius R of the virtual foreign object IF and the shift amount SFT. The shift amount SFT is the distance of separation of the transmission path XP along the intersection direction D2 with respect to the center M of the virtual foreign object IF. The position of the center M of the virtual foreign object IF may be a coordinate value determined according to the setting for where the virtual foreign object IF is placed in the inspection image IM. The position of the transmission path XP may be a coordinate value determined according to the position of the detection element 7p that is the target of the pixel value change (estimation). The inclination of the transmission path XP with respect to the X-ray detection unit 7, as shown in Figure 5, may or may not be considered. If this inclination is not considered, the arrow for the shift amount SFT and the arrow for the transmission path XP may be considered orthogonal in Figure 6.

[0033] The density of the material corresponding to the hypothetical foreign object IF, and the mass absorption coefficient of the material corresponding to the hypothetical foreign object IF, are examples of material properties for calculating the X-ray attenuation rate. Only one of these may be used to calculate the X-ray attenuation rate, or other material properties may be used. The material properties for calculating the X-ray attenuation rate may be stored in the memory unit 10A as parameters, for example. The material corresponding to the hypothetical foreign object IF is, for example, stainless steel. The material corresponding to the hypothetical foreign object IF is not particularly limited and may be selected according to the purpose of the intermediate inspection.

[0034] As shown in Figure 7, the controller 10 calculates the X-ray attenuation rate for each of the multiple detection elements 7p by multiplying, for example, the X-ray transmission distances L1 and L2 in the virtual foreign object IF, the density of the material corresponding to the virtual foreign object IF, and the mass absorption coefficient of the material corresponding to the virtual foreign object IF. The controller 10 obtains the pixel values ​​α and β by multiplying the calculated X-ray attenuation rate by, for example, the pixel values ​​A and B of pixels Px1 and Px2 in Figure 4. The pixel values ​​α and β are the pixel values ​​of pixels Pxv1 and Pxv2 that constitute the pseudo-defective image VIM. Pixels Pxv1 and Pxv2 are pixels in the pseudo-defective image VIM corresponding to pixels Px1 and Px2. The pixel values ​​α and β are pixel values ​​that simulate the further attenuation of X-rays transmitted through the item G by the virtual foreign object IF. The controller 10 generates a pseudo-defective image VIM consisting of pixels Pxv1 and Pxv2 by changing the pixel values ​​of pixels Px1 and Px2 to pixel values ​​α and β.

[0035] Incidentally, the controller 10 may extract the outline of the item G contained in the inspection image IM and generate a pseudo-defective image VIM by changing the pixel values ​​of some pixels inside the extracted outline of the item G (dark areas with the outline of the item G as edges, corresponding to the inspected object). The controller 10 acquires the inspection image IM of the item G being transported from the upstream side and extracts the outline of the item G. The outline of the item G can be extracted by known methods such as binarization, sharpening, and pattern matching.

[0036] Furthermore, the virtual foreign object IF is not limited to cases where it is attached to or mixed with item G, but may also be attached to areas of the container on which item G is placed where item G is not present. In this case, instead of the pixel values ​​A and B of pixels Px1 and Px2, pixel values ​​corresponding to the brightness of X-rays that did not pass through item G but passed through the container may be used. The controller 10 may generate a pseudo-defective image VIM by changing the pixel values ​​of some pixels outside the outline of the extracted item G (bright areas with the outline of item G as the edge, corresponding to the inspected object), some pixels on the boundary line of the extracted item G (weak edge areas), some pixels outside the outline of the container on which the extracted item G is placed (bright areas with the outline of the container as the edge, corresponding to areas other than the inspected object), and some pixels on the boundary line of the container on which the extracted item G is placed (strong edge areas).

[0037] Regarding where to place the virtual foreign object IF within the inspection image IM from the above placement examples, considering locations that affect inspection performance, it may be a selection method based on the operator's input, randomly selected by the controller 10, or a placement pattern may be pre-set in the controller 10. Alternatively, the virtual foreign object IF may be placed in the inspection image IM by directly specifying the coordinates within the inspection image IM through the operator's input. The operator's input is not particularly limited and may include touching or clicking the desired position on the inspection image IM displayed on the display 8, or inputting coordinate values. By being able to place the virtual foreign object IF in this way, the reproducibility of the position of the virtual foreign object IF is increased compared to cases where test pieces are attached manually. As a result, sensitivity differences (variation) can be reduced even in the inspection of items G where sensitivity differences occur, such as difficulty in detecting foreign objects depending on the position of the test piece, and it becomes easier to quantify the sensitivity evaluation at specific locations.

[0038] The memory unit 10A may store multiple virtual foreign object IFs, each differing from the others in at least one of their size, shape, or density. The controller 10 may change the pixel value based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object IF selected from the multiple virtual foreign object IFs. For example, the controller 10 can select the type of virtual foreign object IF to be used for intermediate inspection from the multiple virtual foreign object IFs stored in the memory unit 10A, in response to an operator's input or according to a preset intermediate inspection pattern. When following a preset intermediate inspection pattern, the selection of the type of virtual foreign object IF is performed automatically, making it easier to quantify the foreign object detection accuracy for a specific inspection image IM for each type of virtual foreign object IF, compared to cases where test pieces are attached manually.

[0039] The controller 10 checks the accuracy of the inspection based on the pseudo-defect image VIM generated as described above. Checking the accuracy of the inspection here means, for example, confirming whether a foreign object identified by one or more elements of the foreign object, such as size, thickness, and material, has been detected. The elements to be checked (in other words, the required performance) are configured to be configurable via the display 8. The controller 10 determines whether the predetermined required performance is met based on the pseudo-defect image VI.

[0040] The controller 10 generates a pseudo-defect image VIM and automatically performs an inspection when a preset time (e.g., 1 hour) has elapsed since the start of production of item G (inspection has started), or when a preset number of items G (e.g., 1000 items) have been inspected. In other words, the controller 10 performs a normal inspection to check whether or not an item G contains foreign matter based on the inspection image IM until a preset time (e.g., 1 hour) has elapsed since the start of production of item G, or until a preset number of items G (e.g., 1000 items) have passed through the X-ray inspection device 1. The controller 10 generates a pseudo-defect image VIM for the first time when a preset time has elapsed since the start of production of item G, or when a preset number of items G have passed through the X-ray inspection device 1.

[0041] If the controller 10 obtains an inspection result indicating that no foreign matter is present based on the inspection based on the pseudo-defective image VIM, it determines that there is a problem with the reliability of the inspection. If the controller 10 obtains an inspection result indicating that a foreign matter is present based on the inspection based on the pseudo-defective image VIM, it determines that there is no problem with the reliability of the inspection. In this way, the controller 10 checks the reliability of the inspection based on the pseudo-defective image VIM.

[0042] The controller 10 controls the sorting of the sorting device 15, which is located downstream of the X-ray inspection device 1. If the controller 10 determines that the item G is normal (does not contain foreign matter) in the normal inspection described above, it does not activate the sorting device 15 and transports the item G, which is being transported by the discharge conveyor 9B, to the downstream side. If the controller 10 determines that the item G is abnormal (contains foreign matter) in the normal inspection described above, it activates the sorting device 15 and sorts the item G outside the production line (in a direction different from the direction in which it is transported in the transport section). Examples of sorting devices 15 include arm-type sorting devices using arms, drop-up belt-type sorting devices, pusher-type sorting devices using pusher devices, drop-flap-type sorting devices, air-jet-type sorting devices, and fin-type sorting devices.

[0043] If the controller 10 obtains an inspection result indicating that no foreign matter is present in the inspection accuracy check based on the pseudo-defective image VIM described above, in other words, if it determines that there is a problem with the accuracy of the inspection, it stops the transport of the item G by the input conveyor 9A, the transport conveyor 5, and the output conveyor 9B. In other words, in this case, the controller 10 determines that the current inspection by the X-ray inspection device 1 does not meet the predetermined required performance and stops the operation of the production line. The controller 10 may also notify the worker, for example, by displaying on the display 8 that there was a problem with the inspection accuracy check result.

[0044] Next, the operation of the X-ray inspection device 1 will be described. As shown in Figure 8, when production of goods G begins (step S1), the controller 10 resets the counter that counts the number of items inspected (step S2). The controller 10 counts the goods G that are being transported to the X-ray inspection device 1 (step S3). The controller 10 counts the goods G based on the detection results of the detection sensor 13 that detects the goods G flowing on the input conveyor 9A. The X-ray inspection device 1 acquires an inspection image (X-ray transmission image) IM of the transported goods G (step S4). The controller 10 confirms the number of items inspected when the inspection image IM was acquired (step S5).

[0045] When the controller 10 confirms that the number of inspections i is less than a predetermined number N (for example, N=1000), it performs a normal inspection to check whether or not the item G contains foreign matter based on the inspection image IM (step S6). If the controller 10 determines that the item G contains foreign matter based on the results of the normal inspection based on the inspection image IM (step S6: YES), it activates the sorting device 15 (step S7) to discharge the item G from the system. If the controller 10 determines that the item G does not contain foreign matter based on the results of the normal inspection based on the inspection image IM (step S6: NO), it transports the item G to the downstream side of the discharge conveyor 9B without activating the sorting device 15.

[0046] Subsequently, the controller 10 determines whether or not the production of item G has finished (step S8). If it determines that the production of item G has finished (step S8: YES), it terminates the series of processes. This terminates the series of processes in the X-ray inspection device 1. The termination of the production of item G is determined, for example, by whether or not the number of items G to be produced in a day has been reached. If the controller 10 determines that the production of item G has not finished (step S8: NO), it returns to step S3 and increases the inspection count i by one. After that, the controller 10 executes the steps from step S4 onward.

[0047] When the controller 10 confirms that the number of inspections i is equal to or greater than a predetermined number N (for example, N=1000), it performs an intermediate inspection (step S9). As an intermediate inspection, as shown in Figure 9, the controller 10 manually or automatically selects the type of virtual foreign object IF (step S10) and manually or automatically sets the arrangement of the virtual foreign object IF (step S11), as described above. The controller 10 calculates the X-ray attenuation rate of the material corresponding to the virtual foreign object IF (step S12) and generates a pseudo-defective image VIM by changing the pixel value based on the X-ray attenuation rate (step S13).

[0048] The controller 10 checks the reliability of the inspection if it is not based on an inspection image that does not contain a virtual foreign object, i.e., an inspection using a pseudo-defective image VIM (step S14: NO). If the controller 10 determines that no foreign object is present in the inspection using the pseudo-defective image VIM (step S15: YES), it determines that there is a problem with the reliability of the inspection (step S16) and stops the transport of the item G by the input conveyor 9A, transport conveyor 5 and output conveyor 9B (step S17). This completes the intermediate inspection process.

[0049] On the other hand, if the controller 10 determines that a foreign object is present during inspection using the simulated defective image VIM (step S15: NO), it determines that there is no problem with the accuracy of the inspection (step S18), and activates the sorting device 15 to discharge the corresponding item G from the system (step S19).

[0050] On the other hand, the controller 10 can perform inspections based on inspection images that do not contain virtual foreign objects, i.e., inspections that do not use pseudo-defective images (VIM) (Step S14: YES). If the controller 10 determines that no foreign objects are present in an inspection that does not use pseudo-defective images (VIM) (Step S20: YES), it activates the sorting device 15 to sort the items as normal items under inspection in a direction different from that of Step S19 and Step S22 (Step S21). On the other hand, if the controller 10 determines that foreign objects are present in an inspection that does not use pseudo-defective images (VIM) (Step S20: NO), it activates the sorting device 15 to discharge the corresponding item G from the system (Step S22). This completes the intermediate inspection process. After that, the controller 10 returns to the process shown in Figure 8 and determines whether the production of item G has been completed in Step S8 described above.

[0051] As explained above, in the X-ray inspection device 1, the controller 10 generates a pseudo-defective image VIM that includes a virtual foreign object IF by changing the pixel values ​​of some of the pixels that make up the inspection image. The controller 10 changes the pixel values ​​based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object IF. As a result, in the pseudo-defective image VIM, the virtual foreign object IF is drawn according to the X-ray attenuation rate of the substance. Therefore, compared to cases where, for example, an image of a foreign object taken in advance is used as the virtual foreign object IF, or an image of a circle uniformly filled with black is used, the test piece can be reproduced with greater accuracy using the virtual foreign object IF in the pseudo-defective image VIM.

[0052] The controller 10 modifies the pixel value based on at least one of the following: the X-ray transmission distance L in the virtual foreign object IF, the density of the material, and the mass absorption coefficient of the material. This allows for more accurate calculation of the X-ray attenuation rate of the material in the pseudo-defective image VIM, as it uses at least one of the following: the transmission distance, the density of the material, and the mass absorption coefficient of the material, depending on the type and shape of the material designated as the virtual foreign object IF.

[0053] The transmission distance L is the length of the portion of a hypothetical straight line connecting the X-ray irradiation unit 6 and the X-ray detection unit 7 that passes through a hypothetical foreign object IF. The transmission distance L can be calculated geometrically.

[0054] The X-ray detection unit 7 has multiple detection elements 7p arranged along an intersection direction D2 that crosses horizontally with the transport direction D1 of the transport conveyor 5. The controller 10 changes the pixel value based on the X-ray attenuation rate calculated for each of the multiple detection elements 7p. This makes it possible to draw a virtual foreign object IF at a resolution corresponding to the inspection image, making it easier to accurately reproduce the test piece with the virtual foreign object IF.

[0055] The memory unit 10A stores information on multiple virtual foreign object IFs, each differing from the others in at least one of their size, shape, or density. The controller 10 changes the pixel value based on the X-ray attenuation rate of the material corresponding to the virtual foreign object IF selected from the multiple virtual foreign object IFs. This allows for the generation and inspection of a pseudo-defective image VIM corresponding to one of the multiple pre-stored virtual foreign object IFs.

[0056] If the controller 10 obtains an inspection result indicating that no foreign matter is present based on the inspection using the pseudo-defective image VIM, it determines that there is a problem with the reliability of the inspection. This makes it possible to detect anomalies where the controller incorrectly determines that no foreign matter is present when it should actually be determined that foreign matter is present.

[0057] The controller 10 generates a pseudo-defect image VIM when a preset time has elapsed or when a preset number of items G have been inspected, and automatically performs an inspection. This allows the accuracy of the inspection to be checked automatically according to preset rules. This reduces the burden on the worker.

[0058] The controller 10 extracts the outline of item G included in the inspection image and generates a pseudo-defective image VIM by changing the pixel values ​​of some of the pixels inside the extracted outline. As a result, the pseudo-defective image VIM generated by the controller 10 is similar to the inspection image IM obtained when foreign matter is mixed into item G being produced on the production line. This allows the reliability of the inspection to be checked based on the inspection image IM that is actually obtained when foreign matter is mixed into item G being produced on the production line, thereby improving the accuracy of the inspection.

[0059] If the controller 10 obtains an inspection result indicating that no foreign matter is present based on an inspection image that does not include a virtual foreign matter IF during the execution of the intermediate inspection, it controls the sorting device 15 to sort the items in a direction different from the direction in which normal items G that do not contain foreign matter are sorted by the sorting device 15 when the intermediate inspection is not being performed. If the controller 10 obtains an inspection result indicating that no foreign matter is present based on an inspection image VIM during the execution of the intermediate inspection, it stops the transport of items G by the input conveyor 9A, transport conveyor 5, and output conveyor 9B. This makes it possible to distinguish and sort normal items that have been determined to be free of foreign matter during the execution of the intermediate inspection from normal items when the intermediate inspection is not being performed (during normal operation). In addition, if it is determined that the inspection was not performed correctly, the transport by the input conveyor 9A, transport conveyor 5, and output conveyor 9B is stopped, so that items G containing foreign matter after the intermediate inspection are not processed as normal items.

[0060] While embodiments of the present invention have been described above, the present invention is not necessarily limited to the embodiments described above, and various modifications are possible without departing from the spirit of the invention.

[0061] In the above embodiment, the controller 10 used the X-ray transmission distance in the virtual foreign object IF, the density of the material corresponding to the virtual foreign object IF, and the mass absorption coefficient of the material corresponding to the virtual foreign object IF. However, the controller 10 is not limited to this example, and the pixel value may be changed based on at least one of these.

[0062] In the above embodiment, the transmission distance L was the length of the portion of a hypothetical straight line connecting the X-ray irradiation unit 6 and the X-ray detection unit 7 that passes through a hypothetical foreign object IF, but the example is not limited to this. The transmission distance may be given by a predetermined parameter corresponding to the hypothetical foreign object.

[0063] In the above embodiment and the modified X-ray inspection apparatus 1, an X-ray detection unit 7 consisting of a single line sensor was described as an example, but it may also be configured as a multi-energy sensor consisting of a first line sensor and a second line sensor that can detect different energy bands. Furthermore, the X-ray detection unit 7 may be capable of detecting X-rays using a photon counting method. The X-ray detection unit 7 may be a direct conversion type detection unit or an indirect conversion type detection unit. These sensors are arranged, for example, at least in a direction perpendicular to the transport direction and vertical direction (width direction) of the transport conveyor 5. The elements may be arranged not only in the width direction but also in the transport direction.

[0064] In the above embodiment, the storage unit 10A stores information on multiple virtual foreign object IFs, but it is not limited to this example, and may store information on a single virtual foreign object IF.

[0065] In the above embodiment, during the execution of the intermediate inspection, an inspection was performed based on an inspection image that does not include a virtual foreign object IF. However, this inspection may be omitted, and steps S20 to S22 in Figure 9 may be omitted.

[0066] In describing the operation of the X-ray inspection apparatus 1 in the above embodiment, an example was given in which the controller 10 determines whether or not to perform the inspection probability test based on the number of inspections, as shown in Figure 4, but the explanation is not limited to this. For example, the controller 10 may determine whether or not to perform the inspection probability test based on the elapsed time (e.g., 1 hour) since the start of the inspection.

[0067] In the above embodiment, the controller 10 generates a pseudo-defect image VIM when a predetermined time has elapsed since the start of production of item G, or when a predetermined number of items G have been inspected, and automatically performs a check of the accuracy of the inspection. However, the check of the accuracy of the inspection may also be performed in response to instructions from an operator. [Explanation of symbols]

[0068] 1...X-ray inspection device, 5...Conveyor belt (transport section), 6...X-ray irradiation section (irradiation section), 7...X-ray detection section (sensor section), 7a, 7b, 7p...Detection elements, 10A...Storage section (inspection section), 15...Sorting device (sorting section), D1...Transportation direction, D2...Crossing direction, G...Item, VI, VIM...Simulated defective image.

Claims

1. A conveying unit for transporting goods, An irradiation unit that irradiates the article being transported by the transport unit with X-rays, The sensor unit for detecting the aforementioned X-rays, An image generation unit that generates an inspection image including the article from the X-ray detection result in the sensor unit, An inspection unit that inspects whether or not the article contains foreign matter based on the inspection image, The system includes an inspection unit that generates a pseudo-defective image containing a virtual foreign object by changing the pixel values ​​of some of the pixels that make up the inspection image, and checks the accuracy of the inspection based on the pseudo-defective image, The inspection unit is an X-ray inspection device that changes the pixel value based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object.

2. The X-ray inspection apparatus according to claim 1, wherein the inspection unit changes the pixel value based on at least one of the X-ray transmission distance in the virtual foreign object, the density of the substance, and the mass absorption coefficient of the substance.

3. The X-ray inspection apparatus according to claim 2, wherein the transmission distance is the length of the portion through which a hypothetical straight line connecting the irradiation unit and the sensor unit passes the hypothetical foreign object.

4. The sensor unit has a plurality of detection elements arranged along an intersecting direction that is horizontal to the transport direction of the transport unit. The X-ray inspection apparatus according to claim 1 or 2, wherein the inspection unit changes the pixel value based on the X-ray attenuation rate calculated in accordance with each of the plurality of detection elements.

5. The X-ray inspection apparatus according to claim 1 or 2, wherein the inspection unit stores information on a plurality of virtual foreign objects, each having at least one of the following characteristics: size, shape, and density, and changes the pixel value based on the X-ray attenuation rate of the substance corresponding to the virtual foreign object selected from the plurality of virtual foreign objects.

6. The X-ray inspection apparatus according to claim 1 or 2, wherein the inspection unit determines that there is a problem with the reliability of the inspection if it obtains an inspection result that the foreign matter is not present based on the inspection based on the pseudo-defective image.

7. The X-ray inspection apparatus according to claim 1 or 2, wherein the inspection unit generates the pseudo-defect image when a preset time has elapsed or when a preset number of the articles have been inspected, and automatically performs the inspection.

8. The X-ray inspection apparatus according to claim 1 or 2, wherein the inspection unit extracts the outline of the article included in the inspection image and generates the pseudo-defective image by changing the pixel values ​​of some of the pixels inside the extracted outline.

9. If the inspection unit obtains an inspection result indicating that no foreign matter is present based on the inspection image which does not contain the virtual foreign matter during the execution of the inspection, the sorting unit controls the sorting unit to sort the normal articles which do not contain foreign matter in a direction different from the direction in which the sorting unit sorts them during the non-execution of the inspection. The X-ray inspection apparatus according to claim 1 or 2, wherein if the inspection based on the suspected defective image during the execution of the inspection results in finding that no foreign matter is present, the transport of the article by the transport unit is stopped.