Article inspection device and calibration method therefor

Abstract

To realize a calibration method for an article inspection device with which it is possible to reduce a variation in X-ray detection characteristics between individual articles, and provide an article inspection device capable of executing uniform article inspection with high inspection accuracy.SOLUTION: Provided is an article inspection device of X-ray inspection type, wherein a calibration control unit includes: first imaging processing means for generating image data Ds1 of a first X-ray image based on the output Lx of a plurality of sensor elements in a pre-carry-in state where an article P does not exist in an inspection region Zx; second imaging processing means for generating image data Ds2 of a second X-ray image based on the output signal Lx in a post-carry-in state where a calibration member is brought into the inspection region Zx; and calibration data creation processing means 46, 47 for generating calibration data Dhs for calibrating, on the basis of the image data Ds1, Ds2 and a preset model-specific correction target value Ct, the output signal Lx of the plurality of sensor elements in the pre-carry-in state and the output signal Lx of the plurality of sensor elements in the post-carry-in state to a reference level corresponding to the correction target value in the respective states.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an article inspection apparatus and a calibration method thereof, and more particularly to an article inspection apparatus equipped with an X-ray detector and a calibration method for the article inspection apparatus that calibrates the output of the X-ray detector. [Background technology]

[0002] In an object inspection device that uses X-rays (electromagnetic waves that pass through an object), generally, X-rays are irradiated onto the object to be inspected within the inspection area, and the X-rays that pass through the inspection area are incident on an X-ray detector and detected, causing the X-ray detector to output an image detection signal, and image data of the X-ray transmission image is generated based on this output signal.

[0003] Such object inspection devices often use X-ray detectors with a flat array of sensor elements. In this case, even if the thickness of the conveyor belt or the object being inspected is constant, the material penetration distance of X-rays incident on each sensor element of the X-ray detector increases the further the sensor element is from the center of the arrangement of the multiple sensor elements, so the amount of X-rays received by each sensor element tends to decrease at both ends of the X-ray detector.

[0004] Therefore, conventionally, before inspecting an object, a calibration process is performed in which the sensitivity of the output level of the X-ray detector is corrected so that the image density level of the inspection image based on the output of each element of the X-ray detector becomes a constant value, thereby suppressing variations in the inspection image relative to the reference image density and suppressing blurring and unnecessary gradations in the X-ray transmission image.

[0005] Known examples of this type of object inspection device include a first average value calculation means that calculates the average value of all density data of a line scan image of a predetermined number of lines output from the X-ray detector, a second average value calculation means that calculates the average value of density data corresponding to individual elements from the density data of a predetermined number of lines output from the X-ray detector for each element, a difference calculation means that calculates differential data for each element by subtracting the average value calculated by the second average value calculation means from the average value calculated by the first average value calculation means, and a correction means that corrects the density data from the X-ray detector by subtracting the differential data of elements corresponding to the density data of each element for each line of the X-ray detector including the inspected object (see, for example, Patent Document 1).

[0006] In recent years, X-ray detectors that use a direct conversion method to convert incident X-rays into an output signal proportional to the dose have become increasingly popular.Instead of using a scintillator like the conventional indirect conversion method, these direct conversion X-ray detectors allow X-rays to be incident on and absorbed by an X-ray sensor element made of a compound semiconductor, generating an electric charge proportional to the energy of the X-ray, thereby enabling the generation of an output signal with high detection accuracy.

[0007] An example of an object inspection device equipped with this type of X-ray detector is one that generates calibration data based on the output signals of multiple sensor elements of the X-ray detector when a calibration member is interposed between the conveying belt surface and the X-ray irradiation unit, and the output signals of multiple sensor elements of the X-ray detector when no calibration member is interposed, so that calibration conditions suitable for the object to be inspected can be set in a timely manner from the image data of the corresponding X-ray images, thereby improving the calibration accuracy of the output signals of the X-ray detector (see, for example, Patent Document 2).

[0008] Alternatively, there is known an X-ray detector that includes an incident condition changing means for changing the X-ray incident conditions for the multiple sensor elements of the X-ray detector to two or more types, and creates the necessary calibration data for each incident condition so that the density of the image generated based on the output signals from the multiple sensor elements of the X-ray detector is uniform (see, for example, Patent Document 3). [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-91016 [Patent Document 2] Japanese Patent Publication No. 2022-109774 [Patent Document 3] JP 2019-12011 A Summary of the Invention [Problem to be solved by the invention]

[0010] In conventional item inspection devices such as those described above, the scanning line length and the number of parallel scan lines required for the X-ray detector differ depending on the model, so the X-ray detector is constructed by joining the required number of sensor element modules in the scanning line direction or by arranging multiple sensor element modules in parallel in multiple rows so that the scanning lines are parallel.

[0011] Therefore, particularly when using a direct conversion type X-ray detector in which each sensor element module is configured to the required detection accuracy and variation range without being affected by a scintillator as in the indirect conversion type, sudden changes in the output characteristics of adjacent sensor elements between adjacent sensor element modules can be prevented, but there are cases in which the overall detection characteristics of multiple sensor element modules vary due to differences in the characteristics of the X-ray source and X-ray detector installed in each item inspection device.

[0012] For example, if the standard (correction target value) during calibration is set to the average brightness value of the X-ray transmission image based on the output of multiple sensor elements of the X-ray detector in the article inspection device being calibrated, the calibration standard (correction target value) may end up being different for each inspection device due to variations in the characteristics of the X-ray source and X-ray detector.In such cases, on an article production line where the same type of inspected object is inspected using multiple article inspection devices of the same model, there is concern that the required article inspection accuracy, for example, foreign object detection accuracy, may decrease.

[0013] Furthermore, even if X-ray images of the same food are acquired between a calibrated product inspection device A and another product inspection device B of the same model, the brightness values ​​of the food in the X-ray images may differ between product inspection device A and product inspection device B. In such cases, even if product inspection device A is calibrated and the settings for food foreign body inspection are configured, and then the calibrated parameters are transferred to product inspection device B, it may not be possible to reliably perform the inspection with the same level of accuracy.

[0014] Therefore, the present invention aims to realize a calibration method for an item inspection device that can reduce the variation in X-ray detection characteristics between individual item inspection devices, and to provide an item inspection device that can perform uniform item inspection with high inspection accuracy. [Means for solving the problem]

[0015] (1) To achieve the above object, the article inspection apparatus of the present invention is an X-ray inspection type article inspection apparatus including: an imaging unit that irradiates an inspection area where an object to be inspected is carried with X-rays, and detects the X-rays that pass through the inspection area with an X-ray detector having a plurality of sensor elements adjacent in a predetermined arrangement direction to generate image data of an X-ray image; an inspection control unit that inspects the object to be inspected using the image data generated by the imaging unit and a preset judgment criterion; and a calibration control unit that uses a calibration member that has a uniform thickness in the X-ray irradiation direction for the inspection area over the entire area in the arrangement direction of the plurality of sensor elements, and calibrates output signals of the plurality of sensor elements for the X-rays that have passed through the calibration member, a first imaging processing means for generating image data of a first X-ray image based on output signals from the plurality of sensor elements in a pre-carry-in state where the calibration member is not within the inspection area; a second imaging processing means for generating image data of a second X-ray image based on output signals from the plurality of sensor elements in a post-carry-in state where the calibration member has been carried into the inspection area; and a calibration data creation processing means for generating calibration data for calibrating the output signals from the plurality of sensor elements in the pre-carry-in state and the post-carry-in state to reference levels corresponding to the correction target values ​​in each state, based on the image data of the first X-ray image and the image data of the second X-ray image and a correction target value that is set in advance uniquely for the model.

[0016] This configuration not only reduces variations in the output characteristics of the multiple sensor elements of the X-ray detector within each inspection system, but also enables standardization of calibration results between other inspection systems of the same model. Furthermore, using calibration materials with X-ray transmission characteristics similar to those of the object being inspected enables consistent, high-precision inspection across multiple inspection systems of the same model. Furthermore, the calibration process for the output of the multiple sensor elements of the X-ray detector, performed based on a model-specific correction target value, allows for the incorporation of a variation suppression function based on the model-specific correction target value, enabling the sharing of image processing parameters between inspection systems of the same model. Additionally, by understanding the magnitude of variation in the calibration data itself, the degree of variation and changes in the output of the multiple sensor elements of the X-ray detector of each inspection system can be effectively evaluated based on the variation and changes in the calibration data itself.

[0017] Of course, the model-specific correction target value can be input from outside the apparatus and stored in a memory for saving settings. Furthermore, the calibration member is not limited to an actual calibration member having a predetermined X-ray transmittance. Instead, a virtual calibration member can be used instead. A corresponding dose reduction unit can be provided to uniformly reduce the incident X-ray dose of the X-ray detector across the entire area in the predetermined scanning direction without introducing the calibration member into the inspection area, similar to the reduction in the incident X-ray dose of the X-ray detector when the actual calibration member is introduced. Furthermore, if the predetermined scanning direction is, for example, the line scanning direction (main scanning direction), the thickness of the calibration member in the X-ray transmission direction can change stepwise or continuously in the article transport direction (sub-scanning direction) perpendicular to the line scanning direction. In this case, the second image processing unit can generate multiple types of second X-ray images with different brightness (image density) stepwise or continuously.

[0018] (2) In a preferred embodiment of the present invention, the apparatus may further include a mode switching means capable of switching between an inspection mode in which the object to be inspected is inspected and a calibration mode in which the output signals of the plurality of sensor elements are calibrated using the calibration member.

[0019] In this case, when the mode switching means switches from the inspection mode to the calibration mode, the output signals of the plurality of sensor elements can be calibrated using the calibration members, so that calibration work can be performed accurately when necessary.

[0020] (3) In a preferred embodiment of the present invention, the apparatus may further include a calibration processing unit that performs a first calibration process based on image data of the first X-ray image to align image densities of each sensor element of the image data based on output signals of the plurality of sensor elements in the pre-load state to a first correction target value, and that performs a second calibration process based on image data of the first X-ray image and image data of the second X-ray image to align image densities of each sensor element of the image data based on output signals of the plurality of sensor elements in the post-load state to a second correction target value; and a memory in which at least the second correction target value of the first correction target value and the second correction target value is set and stored in advance as a correction target value specific to the model prior to the first calibration process and the second calibration process.

[0021] In this case, for example, the first calibration process enables calibration to equalize the output characteristics of multiple sensor elements of the X-ray detector within the device, and the output characteristics of multiple sensor elements can also be equalized between other object detection devices of the same model based on the image data of the second X-ray image. Furthermore, since the calibration process is performed in stages, it is possible to more effectively suppress variations in density of the image data.

[0022] (4) In a preferred embodiment of the present invention, the imaging unit has an X-ray tube as a part of an X-ray irradiation source that generates the X-rays, and the calibration processing unit executes a third calibration process to adjust the image density of each sensor element of the image data based on output signals from the plurality of sensor elements to a third correction target value, and an output adjustment unit is provided that adjusts a tube voltage of the X-ray tube when the calibration data creation processing unit executes the third calibration process, and adjusts the irradiation intensity of the X-rays so that the image density of each sensor element of the image data based on output signals from the plurality of sensor elements in the pre-load state approaches the third correction target value.

[0023] In this case, it becomes possible to create effective calibration data that effectively suppresses variations by adjusting the X-ray irradiation intensity. Also, by using the output adjustment means as the aforementioned equivalent dose reduction means instead of the calibration member, it becomes possible to perform a simple calibration process without using the calibration member, or to increase or decrease the reduction amount of the X-ray dose incident on the X-ray detector by the calibration member so as to reduce the number of types of calibration member.

[0024] (5) In a preferred embodiment of the present invention, the imaging unit can adjust the tube voltage of the X-ray tube or the X-ray irradiation intensity in multiple stages, and the calibration control unit can have a multi-stage setting means for setting the first correction target value to a multi-stage correction target value corresponding to each stage of the tube voltage or X-ray irradiation intensity among the multiple stages.

[0025] In this case, by adjusting the X-ray irradiation intensity in multiple stages, it becomes possible to create effective calibration data in which the variations are more effectively suppressed.

[0026] (6) In a preferred embodiment of the present invention, the calibration control unit can be configured to operate the first imaging processing means, the second imaging processing means, and the calibration data creation processing means to generate the calibration data again when the X-ray irradiation source of the imaging unit or the X-ray detector is replaced.

[0027] In this case, when the X-ray radiation source or the X-ray detector of the imaging unit is replaced, the calibration data creation processing means can be operated to reliably generate calibration data that is compatible with the new device configuration.

[0028] (7) To achieve the above object, the present invention provides a method for calibrating an article inspection apparatus using an X-ray inspection system, the method comprising: an imaging unit that irradiates an inspection area into which an object to be inspected is carried with X-rays; and an inspection control unit that detects the X-rays that pass through the inspection area using an X-ray detector having a plurality of sensor elements adjacent in a predetermined arrangement direction to generate image data of an X-ray image; and an inspection control unit that inspects the object to be inspected using the image data generated by the imaging unit and a preset criterion; the method uses a calibration member that has a uniform thickness in the X-ray irradiation direction for the inspection area over the entire area in the arrangement direction of the plurality of sensor elements, and calibrates output signals of the plurality of sensor elements in response to X-rays that have passed through the calibration member; a first imaging step of generating image data of a first X-ray image based on output signals of the plurality of sensor elements in a pre-loading state in which the calibration member is not within the inspection area; a second imaging step of generating image data of a second X-ray image based on output signals of the plurality of sensor elements in a post-loading state in which the calibration member has been loaded into the inspection area; and a calibration data creation step of generating calibration data for calibrating the output signals of the plurality of sensor elements in the pre-loading state and the post-loading state to reference levels corresponding to the correction target values ​​in each state, based on the image data of the first X-ray image and the image data of the second X-ray image and a correction target value that is set in advance uniquely for the model.

[0029] Therefore, by setting the correction target value during calibration to a model-specific value, the present invention reduces the variation in X-ray detection characteristics between individual item inspection devices, thereby achieving an item inspection method that enables uniform item inspection. Furthermore, not only is variation in the output characteristics of the multiple sensor elements of the X-ray detector suppressed within each item inspection device, but calibration results can also be standardized between other item inspection devices of the same model. Furthermore, by using calibration materials that closely resemble the X-ray transmission characteristics of the object being inspected, item inspection with uniform, high inspection accuracy can be achieved across multiple item inspection devices of the same model that inspect the same object. Furthermore, the calibration process for the output of the multiple sensor elements of the X-ray detector, performed based on the model-specific correction target value, suppresses variation based on the model-specific correction target value, enabling the sharing of image processing parameters between item inspection devices of the same model. Additionally, by understanding the magnitude of variation in the calibration data itself, the degree of variation and changes in the output of the multiple sensor elements of the X-ray detector of each item inspection device can be effectively evaluated based on the variation and changes in the calibration data itself.

[0030] As with the apparatus, correction target values ​​specific to the model of the article inspection apparatus can be input from outside the apparatus and stored in advance in a memory for saving settings, etc. Also, an equivalent dose reduction means may be provided that sets a virtual calibration member and uniformly reduces the amount of X-rays incident on the X-ray detector over the entire area of ​​a predetermined scanning direction without having to carry an actual calibration member into the inspection area. Furthermore, when the predetermined scanning direction is, for example, a line scanning direction, the thickness of the calibration member in the X-ray transmission direction may change stepwise or continuously in the article transport direction perpendicular to the line scanning direction. In this case, the second imaging process may generate multiple types of second X-ray images with stepwise or continuously different brightness (image density). [Effects of the Invention]

[0031] According to the present invention, a calibration method for an item inspection device can be realized that can reduce the variation in characteristics between individual item inspection devices, and an item inspection device can be provided that can perform uniform item inspection with high inspection accuracy. [Brief explanation of the drawings]

[0032] [Figure 1] 1 is a schematic configuration diagram of an X-ray inspection type article inspection device according to an embodiment of the present invention. [Figure 2] This is a schematic cross-sectional view showing the general configuration of an X-ray detector in an item inspection device according to one embodiment of the present invention, cut at a predetermined conveying position on a conveying belt, and shows the arrangement of multiple sensor element modules and the arrangement of multiple sensor elements within each sensor element module. [Figure 3] This is a characteristic diagram showing the characteristics of how the amount of X-ray incident on each sensor element of an X-ray detector in an object inspection device according to one embodiment of the present invention changes depending on the X-ray transmittance in the X-ray irradiation path, where the vertical axis shows the sensor element output level corresponding to the image density (brightness), and the horizontal axis shows the amount of X-ray incident on the sensor element or the X-ray transmittance which changes depending on the X-ray irradiation path. [Figure 4] 1 is a graph showing the outputs of multiple sensor elements in a pre-load state in which there is no calibration element in the inspection area of ​​an object inspection device according to one embodiment of the present invention, the outputs of multiple sensor elements in a first post-load state in which a calibration element with low X-ray transmittance has been loaded into the inspection area, and the outputs of multiple sensor elements in a second post-load state in which a calibration element with medium to high X-ray transmittance has been loaded into the inspection area, all associated with element numbers indicating the arrangement positions of the sensor elements in the X-ray detector. The vertical axis represents the sensor element output level corresponding to image density, and the horizontal axis represents the arrangement positions of multiple sensor element modules and the respective multiple sensor elements. [Figure 5] 10 is a flowchart illustrating a procedure for setting a reference characteristic in an article inspection device according to an embodiment of the present invention. [Figure 6] 1 is a flowchart illustrating an example of an inspection process procedure in an article inspection device according to an embodiment of the present invention. [Figure 7] 10 is a flowchart illustrating a procedure of a calibration process in an article inspection device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0033] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0034] Figures 1 to 4 show an article inspection apparatus according to one embodiment of the present invention, and Figures 5 to 7 show the general steps of a calibration method for an article inspection apparatus according to one embodiment, which calibrates the output of the X-ray detector of the apparatus.

[0035] First, the configuration of this embodiment will be described.

[0036] In Figure 1, an item inspection device 1 according to one embodiment of the present invention comprises a conveying unit 10 that conveys an item P to be inspected, an inspection unit 20 that performs X-ray inspection of the item P while being conveyed, a control unit 40 that controls the conveying unit 10 and the inspection unit 20, and a display / operation unit 50 that is signal-connected to the control unit 40 and is positioned near the inspection unit 20 to allow manual operation input.

[0037] 1 and 2, this X-ray inspection type article inspection device 1 irradiates X-rays on article P conveyed by conveying section 10 in inspection section 20, generates image data corresponding to the distribution of the transmitted X-ray dose, and inspects the quality state of article P based on the image data. Note that the quality state referred to here refers to whether article P meets the quality and physical quantities required for a product, such as the presence or absence of contaminants, the presence or absence of missing parts, the pass or fail of the shape, size, storage state, etc. of the contents, and the distribution of density, thickness, volume, or mass.

[0038] The conveying unit 10 is a conveyor that winds a loop-shaped conveying belt 11 around multiple conveying rollers 12 and 13 and can sequentially convey items P through an inspection area Zx on the conveying belt 11 in a predetermined conveying direction (hereinafter also referred to as conveying direction d1) indicated by arrow d1 in Figure 1, and is supported by a housing not shown.

[0039] The inspection unit 20 is configured to include an X-ray generator 21 (X-ray irradiation source) with an X-ray tube 22 built in, and a line sensor-shaped X-ray detector 23 arranged at a predetermined position in the conveying direction so that X-rays emitted from the X-ray generator 21 and transmitted through the inspection area Zx are incident thereon, making it possible to image the item P in the inspection area Zx using X-rays.

[0040] The X-ray generator 21 includes a power supply required to drive the X-ray tube 22 as an X-ray source, cooling means for the X-ray tube 22, and X-ray shielding means, and is configured to cause the X-ray tube 22 to generate X-rays with a wavelength and intensity according to its tube current and tube voltage.

[0041] The X-ray generator 21 is also configured to irradiate the articles P in the inspection zone Zx on the conveyor belt 11 with, for example, fan-beam-shaped X-rays Xfb that spread (and may be deflected) in a line scanning direction d2 perpendicular to the conveying direction d1 of the conveyor unit 10 through an X-ray window portion of an enclosure not shown in detail. However, the X-ray generator 21 is not limited to irradiating the articles P on the conveyor belt 11 with X-rays that pass through the conveyor belt 11, and may be installed in a manner that passes through the articles P on the conveyor belt 11 in an approximately horizontal direction.

[0042] The X-rays generated by the X-ray generator 21 include a plurality of different wavelength regions, and the energy of the X-rays needs to be set within a range suitable for inspecting the article P. Here, the tube voltage and tube current applied to the X-ray tube 22 can be variably controlled by the control unit 40. Furthermore, the X-rays emitted from the X-ray generator 21 or detected by the X-ray detector 23 are assumed to have a certain radiation quality (energy, wavelength) specified according to the quality of the article W, but it is also possible to generate so-called dual-energy or multi-energy X-ray images by using a plurality of different types of X-ray radiation qualities.

[0043] As shown in FIG. 2, the X-ray detection unit 23 is configured to include a plurality of sensor elements e1 to en, each of which receives X-rays and converts them into an electrical signal, and these plurality of sensor elements e1 to en are arranged in a line in adjacent rows in the line scanning direction d2 so that X-rays that have passed through the object P can be incident on each of the corresponding X-ray irradiation paths.

[0044] More specifically, the X-ray detection unit 23 has a configuration in which a required number (k in the figure) of sensor element modules M1 to Mk (k is an arbitrary natural number (positive integer)) are joined together and arranged in a row in the scanning line direction, and each of the sensor element modules M1 to Mk has a plurality of sensor elements e1 to en, as exemplified by any one sensor element module Mi in the figure (hereinafter also referred to as each sensor element module Mi). Note that the sensor element modules M1 to Mk may be arranged in parallel in a plurality of rows so that a plurality of scanning lines are arranged in parallel. Note that in the following description, a plurality of sets of detection elements e1 to en in the sensor element modules M1 to Mk are also referred to as N sensor elements e, based on element numbers 1 to N (= k × n) corresponding to their positions in the scanning line direction.

[0045] As shown in the enlarged partial view at the bottom of Figure 2, each sensor element module Mi has a direct conversion type semiconductor element 32 in its X-ray conversion section formed, for example, from a flat layer of cadmium telluride (CdTe), and is provided with opposing electrodes on both sides, one of which serves as a bias electrode 31 and the other of which serves as an arrangement layer 33 of multiple pixel electrodes corresponding to multiple sensor elements e1 to en. When X-rays that have passed through the article P are incident on the semiconductor element 32, each sensor element module Mi generates electrons and holes according to the energy of the incident X-rays (photons), generating a voltage detection signal between the opposing electrodes 31, 33 according to the incident X-ray dose, and the signal can be output to a readout circuit 34 composed of an ASIC or the like.

[0046] The control unit 40 has a transport control function that controls the transport speed Vc and transport interval of the items W by the transport belt 11 in the transport unit 10, and an inspection control function that controls the X-ray irradiation intensity and irradiation period in the inspection unit 20, and controls the X-ray detection cycle and detection period of each item P in the X-ray detector 23 according to the transport speed Vc of the items W. Note that the configuration of the transport control function unit is similar to that of a known unit, so detailed illustration thereof will be omitted.

[0047] The control unit 40 is configured as hardware including, for example, a microcomputer having a CPU, ROM, RAM, and I / O interface (not shown), an auxiliary storage device that readably stores control programs for performing image processing functions, inspection and judgment functions, and their control functions in cooperation with the ROM, and a timer circuit, etc., and the CPU executes predetermined arithmetic processing while exchanging data with the RAM, etc., and also executes the control programs in accordance with software such as control programs and setting information stored in the ROM, etc.

[0048] (Inspection Control Unit) Specifically, the control unit 40 includes an image data generating unit 41, an image data storage unit 42, an image processing unit 43, and a quality determining unit 44 as an inspection control unit having a control processing function for X-ray inspection.

[0049] The image data generating unit 41 sequentially receives X-ray detection signals (image detection signals) at predetermined intervals from the X-ray detector 23, and generates image data of an X-ray image corresponding to the distribution of the amount of transmission of X-rays that have passed through the inspection region Zx. The image data generating unit 41 performs A / D conversion on image detection signals Lx, which are brightness detection signals from multiple sets of detecting elements e1 to en in multiple sensor element modules M1 to Mk of the X-ray detector 23, and performs an operation (hereinafter referred to as line scanning) of writing, for each predetermined unit transport time corresponding to the size of the detecting elements in the X-ray detector 23, data Lx of the cumulative amount of transmission within that unit time for all detecting element regions having a number N=n×k (N is an integer greater than 1, for example, 256×8) of detecting elements as digital data Di of density levels representing gradations from 0 to 1023, for example. In the following description, any one module will be referred to as the sensor element module Mi with module number i, and each of the multiple sensor elements e1 to en in each sensor element module Mi will be referred to as each sensor element e(i, j), and the multiple sets of detection elements e1 to en in the multiple sensor element modules M1 to Mk will also be referred to as N sensor elements e based on their overall element numbers 1 to N.

[0050] The image data storage unit 42 accumulates and stores the image data Di of the X-ray image from the image data generation unit 41 during a predetermined inspection period corresponding to the length of the item P in the conveying direction, and generates and stores image data of the X-ray transmission image of the item P. This image data storage unit 42 has a data processing program and memory area that perform the function of generating image data Dpx of the X-ray photographed image corresponding to the dose distribution of the X-rays that have passed through the item P, based on the detection data Di of the line scan image that is written sequentially while the line scan by the X-ray detector 23 is repeated a predetermined number of times corresponding to the inspection period of the item P, and outputting this to the image processing unit 43 as imaging data of the item P.

[0051] The image processing unit 43 is configured to perform predetermined filter processing (including pre-processing) capable of extracting image features by importing image data of the X-ray transmission image of the article P from the image data storage unit 42, and image analysis processing such as feature measurement for determining the feature amount of the extracted image features. In this image processing unit 43, a predetermined inspection algorithm combining image processing filters, etc. is set and stored in advance in order to perform a predetermined article inspection based on the image data Dpx imported from the image data generation unit 41.

[0052] The image processing filter included in the inspection algorithm of the image processing unit 43 is a processing program for extracting image features required for a predetermined item inspection based on the image data Dpx of the item P, and when the inspection algorithm includes a filter for foreign object detection, it has a feature extraction filter that performs edge detection processing to emphasize the outline of a foreign object in the item P, and emphasizes the edge of the foreign object by performing differential processing or the like on the area near the pixel of interest. Also, the feature measurement of image features executed by the image processing unit 43 is a process of measuring feature quantities required for the judgment processing in the pass / fail judgment unit 44 by performing calculations on, for example, shading features, color features, shape features, etc. on an image that has been subjected to the necessary pre-processing and image processing on the image data Dpx of the item P imported from the image data generation unit 41.

[0053] The pass / fail judgment unit 44 is configured to perform judgment processing to judge whether or not the item P has a predetermined quality state based on the feature data extracted by the image processing unit 43, such as judgment processing to determine whether or not there is any foreign matter mixed in, whether or not there is a missing part, and whether or not the shape, size, or storage state of the contents are pass / fail. Based on the feature amounts extracted and measured by the image processing unit 43, the pass / fail judgment unit 44 detects characteristic shapes, foreign matter, etc. detected in the item P, and compares the feature amounts such as the area, contour length, and density sum of the detected object with predetermined judgment reference values ​​(hereinafter referred to as predetermined judgment reference) to perform judgment processing to determine whether or not the item P contains a local characteristic shape, etc. corresponding to a foreign matter or a defective part that satisfies the judgment conditions.

[0054] In this manner, the object inspection device 1 of this embodiment irradiates the inspection area Zx into which the object to be inspected P is brought with X-rays Xfb from the X-ray irradiation unit 21 of the inspection section 20, detects the X-rays that pass through the inspection area Zx with an X-ray detector 23 having N sensor elements e, generates image data Di of the X-ray image based on the image detection signal Lx, and performs inspection control in the control unit 40 to inspect the object P using the image data Di and the predetermined judgment criteria that have been set in advance.

[0055] (Calibration control unit) The control unit 40 also functions as a calibration control unit that can calibrate the detected luminance outputs Lx of the N sensor elements e for X-rays that have passed through a calibration member Sp, which typically has a uniform thickness Tf in the X-ray irradiation direction d3 relative to the inspection area Zx, over the entire area in the line scanning direction d2, which is the arrangement direction of the N sensor elements e of the X-ray detector 23. However, instead of using the calibration member Sp, the control unit 40 can also perform the function of the calibration control unit in a pseudo-post-loading state in which the X-ray irradiation intensity from the X-ray generator 21 is reduced to an incident X-ray dose equivalent to the X-ray dose that enters the X-ray detector 23 after passing through the calibration member Sp. Furthermore, the control unit 40 can perform a calibration process to align the detected luminance outputs Lx of the N sensor elements e with a preset zero-point correction reference value even when the X-rays from the X-ray generator 21 pass only through the conveyor belt 11 and enter the X-ray detector 23 without using the calibration member Sp. The zero point correction standard referred to here is, as described in, for example, JP 2006-300887 A (see, for example, paragraphs 0053 to 0065), a method of approximating the phenomenon in which the number of X-ray photons decreases as they pass through a material using the Beer-Lambert law, so that the X-ray irradiation intensity for each pixel (the amount of X-ray incident on each sensor element) becomes a flat amount of light received on the conveyor belt surface.

[0056] This control unit 40 is configured to switch from the inspection mode to the calibration mode when a mode switching request is input from the display operation unit 50, or when, instead of that input, the insertion of a calibration member Sp is detected by the detection signal of the object detection sensor 24 and a predetermined calibration member detection process in the image processing unit 43, assuming that a request to switch to the calibration mode has been made.

[0057] Specifically, the control unit 40 includes an operating condition setting unit 45, a calibration data creation processing unit 46, and a calibration data storage unit 47 as a calibration control unit that calibrates the image data Di generated by the image data generation unit 41 based on the image detection signal Lx from the X-ray detector 23, and further has a mode switching unit 48 that can switch between a normal inspection mode and a calibration mode.

[0058] The operating condition setting unit 45 has a setting means 45a for setting various operating conditions and an operating condition adjustment means 45b that can adjust specific operating conditions, and the setting means 45a displays an operation input screen for setting conditions on the display operation unit 50, and is also capable of performing the function of a calibration condition setting means for specifying the first imaging processing condition, the second imaging processing condition, the third imaging processing condition, and the calibration data creation processing condition.

[0059] Here, the first calibration processing condition is a condition for executing a first imaging process to generate image data Ds1 of a first X-ray image based on the outputs Lx of the N sensor elements e for a predetermined number of scans when the object P is not within the inspection region Zx (pre-load state), and the second imaging processing condition is a condition for executing a second imaging process to generate image data Ds2 of a second X-ray image based on the outputs Lx of the N sensor elements e for a predetermined number of scans when the calibration member Sp is loaded into the inspection region Zx (post-load state). The third imaging processing condition is a condition for executing a third imaging process to change the X-ray irradiation intensity to a value different from the first and second calibration processing conditions by adjusting the tube voltage of the X-ray tube 22 by the operating condition adjustment means 45b when the pre-load state or post-load state is described above, and generate image data D3 of a third X-ray image based on the outputs Lx of the N sensor elements e for a predetermined number of scans.

[0060] The inspection unit 20 and the control unit 40 cooperate in accordance with the setting conditions of the setting means 45a and the switching mode by the mode switching unit 48, and perform the first to third imaging processes described above in the calibration mode.

[0061] The calibration data creation processing conditions set by the setting means 45a are processing conditions for generating calibration data Dhs(i;j) for calibrating the measurement values ​​Lx1(i,j), Lx2(i,j), and Lx3(i,j) (see Figure 3) of the output of each sensor element e(i,j) among the measurement values ​​Lx1, Lx2, and Lx3 of the output of each sensor element e(i,j) per line scan under the first to third imaging processing conditions of the output Lx of N sensor elements e shown in Figure 4 to within the reference levels Rc1, Rc2, and Rc3 within the allowable error ranges (details will be described later) corresponding to the correction target values ​​Ct0, Ctf, and Cts, respectively, based on the image data Ds1 of the first X-ray image, the image data Ds2 of the second X-ray image, or the image data Ds3 of the third X-ray image and the correction target values ​​Ct, for example the first correction target value Ct0 and the second and third correction target values ​​Ctf, Cts, which have been set in advance by the setting means 45a specifically for the model of the item inspection device 1. Here, specific reference characteristic data such as the correction target value Ct set uniquely for the model of the article inspection device 1 is stored in a setting information memory (not shown) built into the setting means 45a of the operating condition setting unit 45 by an input request operation on the display operation unit 50 and data input (input from a memory medium or data communication input) via a communication interface (not shown) in advance. The specific reference characteristic data to be input may be stored in the calibration data storage unit 47 so as to be readable from the setting means 45a. The correction target value Ct is at least the latter of the reference sensor element output (voltage) value and the image density (brightness) value of the X-ray image corresponding to that value, and will be described here as the latter image density (brightness) value.

[0062] The calibration data creation processing unit 46 operates in accordance with the setting conditions of the setting means 45a, and stores in the calibration data storage unit 47 calibration data Dhs(i,j) for calibrating the measurement value Lx1(i,j) of the output of each sensor element e(i,j) in the pre-loading state and the measurement values ​​Lx2(i,j) and Lx3(i,j) of the output of each sensor element (i,j) in the first and second post-loading states in which the incident X-ray doses are different from each other, within the ranges of reference levels Rc1(i,j), Rc2(i,j) and Rc3(i,j) corresponding to the correction target values ​​Ct0(i,j), Ctf(i,j) and Cts(i,j) in the respective measurement states, based on the image data Ds1 of the first X-ray image in the pre-loading state and the image data Ds2 and Ds3 of the second and third X-ray images in the post-loading state and the correction target value Ct previously set uniquely for the model, as shown in Figures 3 and 4. Furthermore, the calibration data creation processing unit 46 functions as a calibration data creation processing means that creates calibration data Dhs(i,j) for the N sensor elements e as calibration data Dhs. Here, the correction target values ​​Ct0(i,j), Ctf(i,j), and Cts(i,j) for each sensor element e(i,j) are the same as Ct0, Ctf, and Cts that have been stored in advance as correction target values ​​common to the N sensor elements e, and the corresponding reference characteristic values ​​Lxa1(i,j), Lxa2(i,j), and Lxa3(i,j), which will be described later, are also similarly common to the N sensor elements e.

[0063] The calibration data Dhs(i,j) created by the calibration data creation processing unit 46 is data for calibrating the output characteristics of the N sensor elements e(i,j) of the X-ray detector 23 to the reference characteristic value Lxa(i,j) by correcting the measured value Lx(i,j) of the X output of each sensor element e(i,j) of the direct conversion type X-ray detector 23 based on the difference between the measured value Lx(i,j) of the detected luminance output and the corresponding reference characteristic value Lxa(i,j), for example, the difference ΔLx from the upper measured value Lx(i,j) shown in Figure 3 (or the difference (-ΔLx) from the lower measured value Lx(i,j) may also be used). The calibration process will be described in detail below.

[0064] The reference characteristic value Lxa(i,j) is set by a calculation formula of a reference characteristic curve based on correction target values ​​Ct0, Ctf, and Cts common to the N sensor elements e, or is prepared as reference characteristic data for all gradation densities including the correction target values ​​Ct0, Ctf, and Cts as corresponding reference characteristic values ​​Lxa1(i,j), Lxa2(i,j), and Lxa3(i,j), and is stored in advance in a setting information memory (not shown) built into the setting means 45a of the operating condition setting unit 45. In other words, the correction target value Ct may be input and stored as reference characteristic data for setting the reference characteristic value Lxa(i,j) together with other data, calculation formulas, etc.

[0065] The calibration process of the output of the entire X-ray detector 23 by correcting the sensitivity of each sensor element e(i, j) includes both sensitivity correction for correcting the variation in the detected brightness Lx(i, j) of the multiple sensor elements e1 to en resulting from differences in the position of each sensor element e(i, j) in the line scanning direction and the amount of incident X-rays, and sensitivity correction for correcting the variation in the detected brightness Lx(i, j) of the multiple sensor elements e1 to en resulting from variations in the detectability of each sensor element e(i, j).

[0066] The former sensitivity correction is a process of setting corresponding sensitivity correction coefficients Ct0(i,j), Ctf(i,j), and Cts(i,j) for each sensor element e(i,j) of each sensor element module Mi so that the measurement value Lx1 in the pre-loading state, the measurement value Lx2 in the first post-loading state, and the measurement value Lx3 in the second post-loading state by N sensor elements are aligned with the correction target values ​​Ct0, Ctf, and Cts of the respective detection sensitivities, as shown in Figure 4, for example.

[0067] In the latter sensitivity correction, for example, as shown in Figure 3, the change characteristic in which the measured value Lx(i,j) of the detected brightness of each sensor element e(i,j) of each sensor element module Mi increases with the amount of X-rays incident on that sensor element e(i,j) is preset as a preferable characteristic curve whose shape can be finely adjusted using shape parameters. Then, the measured value Lx1(i,j) in the pre-loading state, the measured value Lx2(i,j) in the first post-loading state in which a first calibration member Sp1 with low transmittance Trm1 is loaded into the inspection region Zx or a corresponding state of reduced incident X-rays, and the measured value Lx3(i,j) in the second post-loading state in which a second calibration member Sp2 with medium- to high transmittance Trm2 is loaded into the inspection region Zx or a corresponding state of reduced incident X-rays are measured, and the characteristic curve Lx in the same figure corresponding to each sensor element e(i,j) is determined. Note that the characteristic curve Lx may not be a curved line but may be a curve that bends at multiple points.

[0068] Next, measurement values ​​Lx1(i,j), Lx2(i,j), and Lx3(i,j) are acquired for each of the N sensor elements e(i,j), and their characteristic curves are determined, and correction coefficients C1, C2, and C3 are calculated to match them with the correction target values ​​Ct0, Ctf, and Cts.

[0069] Then, for each of the sensor element output intervals Ez1, Ez2, Ez3, and Ez4, the correction target values ​​Cxv or corresponding correction coefficient values ​​Cxk corresponding to all gradation densities of the output Lx(i,j) of each sensor element e(i,j) are calculated by interpolation based on the rate of change in each interval of the reference characteristic curve Lxa(i,j) including the correction target values ​​Ct0, Ctf, and Cts and the calculated correction coefficients C1, C2, and C3. The correction target values ​​Cxv or corresponding correction coefficient values ​​Cxk correspond to all gradation densities of the output Lx(i,j) of each sensor element e(i,j). The correction target values ​​Cxv or corresponding correction coefficient values ​​Cxk correspond to all gradation densities of the output Lx(i,j) of each sensor element e(i,j). The correction target values ​​Cxv, Ctf, and Cts are calculated by interpolation based on the rate of change in each interval of the reference characteristic curve Lxa(i,j) including the correction target values ​​Ct0, Ctf, and Cts and the calculated correction coefficients C1, C2, and C3.

[0070] In this case, the correction target values ​​Cxv or the corresponding correction coefficient values ​​Cx, which are a group of target values ​​corresponding to the reference characteristic values ​​of each image density (brightness) on the reference characteristic curve Lxa(i,j), enable correction to effectively bring the X-ray detection sensitivity of each sensor element e(i,j) of the X-ray detector 23 closer to the reference characteristic for the sensor element output Lx of all gradation densities of the inspection X-ray image. Note that the interpolation calculation of the correction target values ​​Cxv or the correction coefficient values ​​Cxk corresponding to all gradation densities of the output Lx(i,j) of each sensor element e(i,j) using the minimum value 0 and the correction target values ​​Ct0, Ctf, and Cts may include linear interpolation so that the characteristic curve bends at multiple points within the allowable error range of the detection sensitivity, thereby simplifying the calculation process. In this case, the correction target value Cxv or the correction coefficient value Cxk becomes calibration data Dhs(i,j) that calibrates the measurement value Lx(i,j) corresponding to all gradation densities of the output Lx(i,j) of each sensor element e(i,j) to within the reference levels Rc1, Rc2, and Rc3 within the allowable error ranges corresponding to the correction target values ​​Ct0, Ctf, and Cts on the high transmittance side of the sensor element output sections Ez1, Ez2, and Ez3, respectively. The allowable error range referred to here is, at most, approximately the error (error from the reference characteristic value) when linearly interpolating the sensor element output Lx(i,j) between the characteristic reference values ​​Lxa1 and Lxa2 in FIG. 3, and can be an error sufficiently small compared to the difference ΔLx in the same figure.

[0071] When the sensor element output Lx(i,j) is interpolated between the characteristic reference values ​​Lxa1, Lxa2 in accordance with the output change rate corresponding to the output of the reference characteristic curve Lxa(i,j), it goes without saying that the reference levels Rc1, Rc2, and Rc3 become values ​​substantially equal to the correction target values ​​Ct0, Ctf, and Cts. In other words, by sensitivity-correcting the measured value Lx(i,j) of a certain gradation density (luminance), which can be regarded as a point on the characteristic curve of each sensor element e(i,j), to the output value of the reference characteristic of the corresponding gradation density on the reference characteristic curve Lxa, it is possible to calibrate each of the N sensor elements e(i,j) so that it can output at the standard detection sensitivity for the incident X-ray dose. In addition, a correction coefficient or correction formula (Lxa(i,j) / Lx(i,j)) can be set to correct the measurement value Lx(i,j) of each gradation density of each sensor element e(i,j) to the reference characteristic value Lxa(i,j) of the same gradation density, and the calibration process can be performed.

[0072] Variations of this calibration process include a first calibration process that does not use the calibration member Sp, a second calibration process that uses the calibration member Sp, and a third calibration process that adjusts the X-ray irradiation intensity of the X-ray generator 21 regardless of whether the calibration member Sp is used or not, to cause a change in irradiation intensity corresponding to a predetermined X-ray transmission thickness (for example, the X-ray transmission thickness of the calibration member Sp) or change the X-ray transmission thickness.

[0073] The calibration data memory unit 47 is configured to store the calibration data Dhs including the correction target value Cxv or the correction coefficient value Cxk created by the calibration data creation processing unit 46 in association with the position (i, j) of each sensor element e(i, j) in each of the multiple sensor element modules M1 to Mk.

[0074] The mode switching unit 48, in cooperation with the display operation unit 50, or in cooperation with the item detection sensor 24 of the inspection unit 20 and the calibration member detection processing unit 43a of the image processing unit 43, determines at a predetermined interval whether there is a request to switch to the calibration mode, and depending on the result, it is able to switch between a normal inspection mode in which the item P is inspected and a calibration mode in which the output signals Lx(i,j) of multiple sensor elements e1 to en are calibrated using the calibration member Sp.

[0075] The image processing unit 43 also has a calibration processing unit 43b that executes a predetermined calibration process while acquiring calibration data Dhs(i,j) associated with the position (i,j) of each sensor element e(i,j) from the calibration data storage unit 47. The image processing unit 43 then performs the function of a calibration processing means that cooperates with the calibration data creation processing unit 46 via the calibration data storage unit 47, by using this calibration processing unit 43b to correct a specific output value on the characteristic curve of the output of each sensor element e(i,j) using the corresponding gradation density value from the correction target value Cxv or correction coefficient value Cxk, and executes a process of calibrating the output signals Lx of the N sensor elements e to output data that matches the reference characteristic curve Lxa.

[0076] Here, the correction target values ​​Cxv or the corresponding correction coefficient values ​​Cxk for the entire gradation density range of the output of each sensor element e(i, j), including the correction target values ​​Ct0, Ctf, and Cts, or further the reference characteristic curve Lxa and reference characteristic data relating to its time-varying characteristics, are set in advance as correction target values ​​specific to the model and stored in the memory 47 prior to the calibration process by the image processing unit 43.

[0077] The operating condition setting unit 45 of the control unit 40 adjusts the tube voltage of the X-ray tube 22 using the operating condition adjustment means 45b when the image processing unit 43 (calibration processing means) performs the first calibration process, so as to bring the image density A1(i, j) for each pixel of the image data D1 based on the output signals Lx1 of the N sensor elements e in the pre-load state closer to the first correction target value Ct0.

[0078] Furthermore, the inspection unit 20, which is an imaging unit, can adjust the tube voltage or X-ray irradiation intensity of the X-ray tube 22 in multiple stages, and the setting means 45a of the operating condition setting unit 45, which is part of the calibration control unit of the control unit 40, is a multi-stage setting means that sets the first correction target value Ct0 to a multi-stage correction target value corresponding to the tube voltage or X-ray irradiation intensity of each stage among the multiple stages.

[0079] In addition, when the X-ray irradiation source 21 of the inspection unit 20 or the X-ray detector 23 is replaced, the operating condition setting unit 45 is configured to set operating conditions that cause the aforementioned first imaging process, second imaging process, and calibration data creation process to be executed, respectively, and the generation of the calibration data Dhs to be executed again.

[0080] Note that signal G1 from the inspection unit 20 to the control unit 40 shown in FIG. 1 indicates detection signal inputs such as an object detection signal from the inspection unit 20, an encoder pulse for the transport motor, and the tube voltage of the X-ray tube 22, and signal G2 from the control unit 40 to the inspection unit 20 indicates a control signal output for controlling a speed control signal for the transport motor of the transport unit 10 and the tube voltage of the X-ray tube 22, etc.

[0081] The material and thickness of the calibration member Sp can be selected and set according to the type of item P to be inspected. For example, the control unit 40 may select the material and thickness of the calibration member Sp suitable for calibration based on the type information of the item P selected as the inspection object, and display and output it on the display operation unit 50.

[0082] Furthermore, as the material for the calibration member Sp, if the item P is, for example, meat, polyacetal resin, polyethylene terephthalate, acrylic resin, or a container of such a resin filled with water or the like can be used, and if the item P contains a lot of fat, polyethylene or a container of such a resin filled with oil or the like can be used. Furthermore, if the item P contains bone or salt, the calibration member Sp can be made of aluminum, polyvinyl chloride, or the like.

[0083] Furthermore, in this embodiment, the calibration member Sp has a uniform thickness Tf in the X-ray irradiation direction d3 relative to the inspection region Zx throughout the line scanning direction d2, which is the sensor element arrangement direction of the X-ray detector 23, and has a thick portion (large plate thickness) corresponding to low transmittance Trm1 on one side in the item conveying direction d1 and a thin portion (small plate thickness) corresponding to medium to high transmittance Trm2 on the other side in the item conveying direction d1. However, the uniform thickness Tf throughout the line scanning direction d2 may vary in multiple steps or continuously depending on the position in the item conveying direction. Alternatively, if the X-ray detector 23 has sensor elements arranged in multiple rows adjacent to each other in the item conveying direction, a calibration member Sp may be used whose X-ray transmittance varies depending on the incident position and incident angle of each row (each scan line).

[0084] Furthermore, calibration data may be generated by acquiring an image detection signal of the calibration member Sp without driving the transport unit 10. Furthermore, if the calibration member Sp cannot cover the plurality of sensor element modules M1 to Mk of the X-ray detection unit 23, image detection signals of the calibration member Sp that covers a predetermined number of the sensor element modules M1 to Mk may be collected multiple times to generate calibration data.

[0085] The specific calculation procedure for calibration data regarding points other than those mentioned above can be roughly the same as the procedure exemplified in Patent Document 2.

[0086] In other words, when calculating the calibration data, the calibration data creation processing unit 46 drives the conveying device 21 and uses image data in which the fluctuations in the image data due to the thickness of the belt and seams have been averaged, thereby preventing a decrease in accuracy.

[0087] When creating this calibration data, the following may be considered: element Ha for compensating for changes in the incident intensity of X-rays depending on the distance from the X-ray generator 21 to each sensor element ei, particularly the material penetration distance; element Hb for compensating for sensitivity differences due to differences in the incident direction and position of the X-rays on the item P to be inspected; element Hc for compensating for sensitivity differences to X-rays for each sensor element module Mi; element Hd for compensating for sensitivity differences for each sensor element within each sensor element module Mi; element He for compensating for noise generated at the boundary between adjacent sensor element modules; and element Hf for compensating for sensitivity differences due to other X-ray transparent members (for example, the belt support member of the conveying section 10 and the exit port cover material of the X-ray generator 21).

[0088] Furthermore, of these compensation elements, Ha and Hb depend on the material and thickness of the substance through which the X-rays pass, and so the calibration error can be reduced by using, as the calibration member Sp, a member having a transmittance and thickness close to that of the inspected item P. For example, it is conceivable to use, as the calibration member Sp, a portion of the contents of the inspected item P that is made of a uniform material and has a uniform thickness.

[0089] Suppose that the density of the image obtained from the output Lx1(i,j) of each sensor element e(i,j) of each sensor element module Mi under a first incident condition (for example, transmittance 100%) is A1(i,j), the ideal density of the image is R1(i,j), and the compensation term according to the incident condition is H(i,j), then: A1(i,j)=R1(i,j)+H(i,j) ···(1) H(i,j)=Ha(i,j)+Hb(i,j)+Hc(i,j)+Hd(i,j)+He(i,j)+Hf(i,j) ···(2) This becomes:

[0090] Furthermore, if the density of the image obtained from the output Lx2(i,j) under the second incident condition (for example, a transmittance of 30%) is A2(i,j) and the ideal density of that image is R2(i,j), then: A2(i,j)=R2(i,j)+H(i,j) ···(3) In the above formula (3), the density obtained by image processing is expressed as the sum of the ideal density R2(i,j) and the compensation term H(i,j), but it can also be expressed as the product of the ideal density and the compensation term.

[0091] In this way, from a plurality of different transmittances and the image densities obtained for each transmittance, it is possible to grasp the characteristics of each sensor element e(i,j) using a characteristic diagram of transmittance (incident X-ray amount) versus image density (sensor element output), such as the example shown in Figure 3. This characteristic diagram shows a characteristic curve that shows that the image density for two different incident conditions, namely, 100% and 30% X-ray transmittance of the calibration member Sp, sufficiently accurately represents the output characteristics of the sensor element e(i,j) in the low transmittance region. This characteristic curve can be approximated by, for example, a quadratic function or an exponential function.

[0092] Therefore, assuming that H(i,j) is a value close to 0 at the output Lx2(i,j) of each sensor element e(i,j) for a predetermined low transmittance, for example a transmittance of 30%, that is, that the density of the image obtained from the output of sensor element e(i,j) is a density close to the ideal value, a characteristic curve Lxa of transmittance versus image density close to the ideal can be calculated from multiple density values ​​including densities A1(i,j) and A2(i,j) at transmittances of 100% and 30%, respectively, and preferably density A3(i,j) at an additional transmittance (for example, 50% or 60%).

[0093] However, in this embodiment, instead of obtaining a predetermined number of samples of the output Lx1(i,j) of each sensor element e(i,j) at a predetermined high and low transmittance in a preliminary experiment and calculating the average value, a model-specific correction target value Ct is set that can be used commonly between the item inspection device 1 and other item inspection devices of the same model, and calibration data Dhs(i,j) including the correction target value Cxv or correction coefficient value Cxk for each sensor element e(i,j) calculated based on this is stored in the correction data memory unit 47.

[0094] Therefore, by correcting the image data obtained from the output of each sensor element e(i,j) during inspection using this calibration data Dhs(i,j), it is possible to accurately obtain image data of the item P being inspected.

[0095] Here, the transmittance of interest is set to multiple levels, for example, three levels, but it may be set to two or more levels, and the reference characteristics may be set by focusing on two or more types of transmittance that do not include 100%.

[0096] As described above, it is possible to change the incidence conditions not only by changing the incidence conditions using the calibration member Sp but also by changing the tube current or tube voltage of the X-ray tube of the X-ray generator 21. By changing the tube current or tube voltage, it is possible to adjust the incidence conditions of X-rays for each sensor element e(i, j), but it is also possible to combine changing the transmittance using the calibration member Sp (for example, 100% and 30%), changing the tube voltage, or changing the tube current. Furthermore, the calibration data can also be obtained for each energy of X-rays incident on the X-ray detector 23.

[0097] Next, the operation will be described.

[0098] (Inspection mode) In the inspection mode, the control unit 40 acquires an X-ray transmission image of the article P being conveyed by the conveying unit 10, and determines whether the article P is good or bad by determining whether or not there is a foreign matter.

[0099] (Input and storage of reference characteristic data) Prior to control of the inspection mode, the control unit 40 executes a process for setting reference characteristics as shown in the outline of the procedure in FIG.

[0100] First, the correction target values ​​Ct, for example, the correction target values ​​Ct0, Ctf, and Cts, which are set uniquely for the model of the article inspection device 1, are input as data and stored in the setting information memory built into the setting means 45a (step S1).

[0101] Next, using the inputted reference characteristic data and a characteristic calculation formula or the like previously stored in the setting means 45a or the calibration data creation processing unit 46, a reference characteristic curve Lxa is determined in which the correction target values ​​Ct0, Ctf, and Cts correspond to the sensor element outputs Lx1, Lx2, and Lx3 of a specific incident X-ray dose, and a reference characteristic value Lxa(i,j) corresponding to the incident X-ray dose for each sensor element e(i,j) is calculated (step S2). Alternatively, reference characteristic data instead of the reference characteristic value Lxa(i,j) may be acquired by data input.

[0102] Next, the correction target value Cxv or the corresponding correction coefficient value Cxk corresponding to the full gradation density of the output Lx(i,j) corresponding to the incident X-ray dose to each sensor element e(i,j) of the X-ray detector 23 is calculated using the procedure described above (step S3).

[0103] Then, the setting information relating to the calculated or acquired reference characteristics is stored in the setting information memory built into the setting means 45a, or is stored in the calibration data storage unit 47 so as to be readable from the setting means 45a (step S4).

[0104] Next, when the start of inspection is requested on the display operation unit 50 without the calibration mode being selected, the control unit 40 executes the processing of the inspection mode as shown in FIG.

[0105] First, when the inspection starts, X-rays that have passed through the inspection area Lx are incident on the X-ray detector 23, and an X-ray image of the belt surface of the conveyor unit 10 is acquired (step S11), and the article detection sensor 24 determines whether or not the introduction of the article P or calibration member Sp has been detected (step S12). If the determination result is NO (NO in step S12), an X-ray image of the belt surface of the conveyor unit 10 is acquired again at a predetermined line scanning period, and if the determination result is YES (YES in step S12), then the X-rays that have passed through the article P or calibration member Sp are incident on the X-ray detector 23, and image data of the X-ray transmission image of the article P or calibration member Sp is acquired (step S13).

[0106] Next, the calibration member detection processing unit 43a of the image processing unit 43 determines whether or not a calibration member Sp having a uniform thickness Tf in the X-ray irradiation direction d3 has been detected across the entire area of ​​the line scanning direction d2 of the X-ray detector 23 from the degree of variation in the outputs of the multiple sensor elements e1 to en in each of the multiple sensor element modules M1 to Mk, for example, the degree of variation in the output of a specific sensor element e(i, j) in each sensor element module Mi (step S14).

[0107] If it is determined that the calibration member Sp has not been detected (NO in step S14), then, based on the image data Di of the X-ray transmission image of the detected item P, a judgment process is performed in the quality judgment unit 44, for example, whether or not there is any foreign matter in the item P, and the quality of the item P is judged (step S15), and then it is checked whether the inspection quantity has reached a predetermined set value (step S16).

[0108] At this time, if the number of inspections has not reached the preset value (NO in step S16), the above-mentioned steps S11 to S15 are repeated from the start of the inspection, and when the number of inspections has reached the preset value (YES in step S16), the current processing ends.

[0109] On the other hand, if it is determined in the step S14 for determining the delivered item that the calibration member Sp has been detected (YES in step S14), then the mode switching unit 48 is activated, and when the item P is in the pre-delivery state where it is not within the inspection area Zx, a first imaging process is executed to generate image data Ds1 of a first X-ray image based on the image detection signal Lx1 from the image generation unit 41, and when the item P is in the post-delivery state where it has been delivered into the inspection area Zx, a second imaging process is executed to generate image data Ds2 of a second X-ray image based on the image detection signal Lx2 or Lx3 from the image generation unit 41, or a second imaging process is executed to generate image data Ds3 of a third X-ray image. Then, the calibration data creation processing unit 46 operates according to the setting conditions of the setting means 45a, and based on the image data Ds1, Ds2, Ds3 of the first to third X-ray images and the correction target value Ct previously set uniquely for each model, calibration data Dhs(i,j) is calculated for calibrating the output signal Lx1(i,j) of each sensor element e(i,j) in each sensor element module Mi in the pre-loaded state and the output signal Lx2(i,j) or Lx3(i,j) of each sensor element (i,j) in the post-loaded state to reference levels Rc1(i,j), Rc2(i,j) and Rc32(i,j) corresponding to the correction target values ​​Ct0(i,j), Ctf(i,j), Cts(i,j) in each state (step S17).

[0110] Next, the calibration processing unit 43b of the image processing unit 43 sequentially applies the calibration data Dhs(i,j) to each corresponding pixel data A(i,j) of the image data Di of the X-ray transmission image Dpx of the object P imported from the image data memory unit 42, and the image data Di of the X-ray transmission image Dpx is calibrated (step S18).

[0111] Next, it is determined whether the calibration member Sp is reflected in the correct position of the X-ray image of the calibration member Sp depending on whether the image data Di after the calibration process of the X-ray transmission image Dpx is within the normal image density range for each pixel data A(i, j) (step S19). If it is normal (YES in step S19), the current process ends, and if it is not normal (NO in step S19), an error stop state occurs (step S20) and the current process ends.

[0112] (Calibration mode) When the calibration mode is selected on the display operation unit 50, the control unit 40 executes the calibration mode process shown in FIG.

[0113] First, an X-ray image of the belt surface of the conveying unit 10 is acquired (step S21), and then a predetermined wizard screen for specifying the type of calibration member Sp is displayed on the display operation unit 50, and the specified calibration member Sp is transported into the inspection area Zx of the conveying unit 10.

[0114] Next, when a specified calibration member Sp, for example, a PET calibration member Sp having a thickness Tf of 20 mm, is carried into the inspection region Zx of the transport unit 10 (step S22), the calibration member Sp is detected by the object detection sensor 24, and then image data Ds2 or Ds3 of the X-ray image of the calibration member Sp is acquired (step S23). That is, a first imaging process is executed to generate image data Ds1 of the first X-ray image based on the image detection signal Lx1 from the X-ray detector 23, and at the same time, when the calibration member Sp is in the post-carry-in state after being carried into the inspection region Zx, a second or third imaging process is executed to generate image data Ds2 or Ds3 of the second or third X-ray image based on the image detection signal Lx2 or Lx3 from the X-ray detector 23.

[0115] Next, it is confirmed whether the calibration member Sp is reflected in the correct position in the X-ray image of the calibration member Sp (step S24), and the confirmation result is input by selection operation on the next wizard screen, and it is determined whether there is a problem with the image (step S25).

[0116] If it is determined that there is a problem with the image (NO in step S25), the process returns to step S22 for carrying in the calibration member Sp, and if it is determined that there is no problem with the image (YES in step S25), the process proceeds to the next step, where it is determined whether an image has been acquired for the entire area of ​​the calibration member Sp (step S26).

[0117] At this time, if it is determined that acquisition of all X-ray images over the entire area of ​​the calibration member Sp (here, portions of multiple calibration members with different X-ray transmittances) has been completed (YES in step S26), the calculation process of the calibration data as described above is executed (step S27), and if acquisition of all X-ray images over the entire area of ​​the calibration member Sp has not been completed (NO in step S26), the process returns to step S22 for carrying in the calibration member Sp. Note that the X-ray images of the calibration member Sp here can be acquired as data that varies over the conveyance period, corresponding to a configuration in which multiple calibration member portions with different thicknesses, such as a 20 mm thick one made of PET or a 60 mm thick one, are divided and arranged in the article conveyance direction and integrated.

[0118] Next, calibration data Dhs is calculated based on the image data of the acquired X-ray image (step S27), and the next wizard screen displays an image of a component for image confirmation on the display operation unit 50, prompting the user to confirm whether the calibration component Sp is displayed normally in the correct position on the confirmation component image, and to input the result to confirm that there are no problems with the calibrated image (steps S28, S29).

[0119] At this time, if it is determined that the image is a normal calibrated image without any problems (YES in step S29), the calibration process will be terminated, but if it is not determined that the image is a normal calibrated image (NO in step S29), the calibration process will be executed again from the initial step S21.

[0120] (One embodiment of the calibration method) In the article inspection apparatus 1 of this embodiment, the article inspection apparatus 1 is an X-ray inspection type that is equipped with an imaging unit 20 that irradiates an inspection area Zx into which an article P to be inspected is transported with fan-beam-shaped X-rays Xfb, and detects the X-rays that pass through the inspection area Zx with an X-ray detector 23 having multiple sensor element modules M1 to Mk adjacent in a predetermined arrangement direction to generate image data Di of the X-ray image, and an inspection control function unit 40A of a control unit 40 that inspects the article P using the image data Di generated by the imaging unit 20 and preset judgment criteria.In this case, the output signal Lx of the multiple sensor elements e1 to en of each sensor element module Mi in response to the X-rays that have passed through the calibration member Sp is calibrated using a calibration member Sp that has a uniform thickness Tf in the X-ray irradiation direction d3 relative to the inspection area Zx over the entire area of ​​the line scanning direction d2 of the X-ray detector 23, which is the arrangement direction area of ​​the multiple sensor element modules M1 to Mk, and here has different large and small thicknesses (X-ray transmission thicknesses corresponding to Trm1 and Trm2) on one side and the other side of the article conveying direction d1.

[0121] The calibration method for this article inspection device 1 includes a first imaging step (step S21) of generating image data Ds1 of a first X-ray image based on output signals Lx1 of N sensor elements e for a predetermined number of scans in a pre-load state in which the article P is not within the inspection region Zx, and a second imaging step (steps S22, S23) of generating image data Ds2 of a second X-ray image based on output signals Lx2 of N sensor elements e for a predetermined number of scans in a post-load state in which the calibration member Sp has been loaded into the inspection region Zx. Alternatively, instead of the first or second imaging step, the operating condition adjustment means 45b adjusts the tube voltage of the X-ray tube 22 to change the X-ray irradiation intensity to a value different from the first and second calibration processing conditions, thereby generating a virtual calibration member Sp having an X-ray transmittance different from that of the actual calibration member Sp. and a calibration data creation process (step S27) for generating calibration data Dhs for calibrating the output signals Lx1 of the N sensor elements e in the pre-load state and the output signals Lx2 or Lx3 of the N sensor elements e in the post-load state to reference levels Rc1, Rc2, Rc3 corresponding to the correction target values ​​Ct0, Ctf, Cts in the respective states, based on the image data Ds1, Ds2, D3 of the first to third X-ray images and a correction target value Ct previously set uniquely for the model.

[0122] Next, the operation will be described.

[0123] The article inspection apparatus 1 of this embodiment configured as described above not only suppresses variations in the output characteristics of the plurality of sensor elements e1 to en in each sensor element module Mi of the X-ray detector 23 within each article inspection apparatus 1, but also makes it possible to standardize calibration results among other article detection apparatuses of the same model. Furthermore, by using a calibration member Sp whose X-ray transmission characteristics are similar to those of the article P, article inspection with uniform high inspection accuracy becomes possible among multiple article detection apparatuses of the same model that inspect the same article P.

[0124] Furthermore, the calibration process of the outputs of the multiple sensor elements e1 to en in each sensor element module Mi of the X-ray detector 23, which is carried out based on the correction target values ​​Ct0, Ctf, and Cts set uniquely for each model, makes it possible to incorporate a function for suppressing variation based on the correction target values ​​Ct0, Ctf, and Cts unique to each model, and makes it possible to share image processing parameters between item inspection devices of the same model.

[0125] In addition, by understanding the magnitude of variation in the calibration data itself, it becomes possible to effectively evaluate the degree of variation and changes in the output of the N sensor elements e in the X-ray detector 23 of each item inspection device based on the variation and changes in the calibration data itself.

[0126] In addition, in this embodiment, the operating condition setting unit 45, calibration data creation processing unit 46, and calibration data storage unit 47, which are the calibration control units of the control unit 40, further have a mode switching unit 48 that can switch between an inspection mode in which the item P is inspected and a calibration mode in which the output signals Lx of the N sensor elements e are calibrated using the calibration members Sp. Therefore, when the mode switching unit 48 switches from the inspection mode to the calibration mode, the output signals Lx of the N sensor elements e of the X-ray detector 23 can be calibrated using the calibration members Sp, making it possible to perform calibration work accurately when necessary.

[0127] Furthermore, in this embodiment, the operating condition setting unit 45, calibration data creation processing unit 46, and calibration data storage unit 47, which are the calibration control units, perform a first calibration process based on the image data Ds1 of the first X-ray image to align the image density A1(i,j) based on the output of each sensor element in the pre-load state to a first correction target value Ct0, and also perform second and third calibration processes based on the image data Ds2 or Ds3 of the second or third X-ray image to align the image densities A2(i,j) and A3(i,j) based on the output of each sensor element in the post-load state to second and third correction target values ​​Ctf and Cts, and the first correction target value Ct0 and the second and third correction target values ​​Ctf and Cts are set in advance as correction target values ​​Ct specific to the model prior to the first to third calibration processes and are stored in the memory of the calibration data storage unit 47. Therefore, for example, the first calibration process enables calibration to equalize the output characteristics of the multiple sensor element modules M1-Mk and the multiple sensor elements e1-en of the X-ray detector 23 within the device itself, and it becomes possible to equalize the output characteristics of the multiple sensor elements e of the X-ray detector even between other object detection devices of the same model. Furthermore, since the calibration process is performed in stages, it is possible to more effectively suppress variations in density of image data.

[0128] In this embodiment, the inspection unit 20, which serves as the imaging unit, includes an X-ray tube 22 as part of the X-ray irradiator 21 that generates X-rays. The calibration control units of the control unit 40, which include an operating condition setting unit 45, a calibration data creation processing unit 46, and a calibration data storage unit 47 (hereinafter simply referred to as the calibration control units 45-47), include an operating condition adjustment unit 45b (output adjustment unit) that adjusts the tube voltage of the X-ray tube 22 when the calibration data creation processing unit 46 performs the first calibration process, thereby adjusting the X-ray irradiation intensity so that the image density A1(i,j) for each pixel of the image data based on the output signals of the multiple sensor elements e1-en in the pre-load state approaches the first correction target value Ct0. Therefore, by adjusting the X-ray irradiation intensity, effective calibration data can be created that effectively suppresses variations in image density A1(i,j). Furthermore, by utilizing the function of the equivalent X-ray reduction unit, a simple calibration process without using a calibration member can be performed.

[0129] Furthermore, in this embodiment, the inspection unit 20 can adjust the tube voltage or X-ray irradiation intensity of the X-ray tube 22 in multiple stages, and the calibration control units 45 to 47 have multi-stage setting means for setting the first correction target value to a multi-stage correction target value corresponding to each stage of the tube voltage or X-ray irradiation intensity among the multiple stages. Therefore, by adjusting the X-ray irradiation intensity in multiple stages, it is possible to create effective calibration data in which variation is further effectively suppressed.

[0130] Furthermore, in this embodiment, when the X-ray tube 22 of the X-ray generator 21 is replaced or the X-ray detector 23 is replaced, the calibration control units 45 to 47 appropriately perform the functions of the first to third imaging processes and the calibration data creation process to generate calibration data again. Therefore, when the X-ray tube 22 of the X-ray generator 21 is replaced or the X-ray detector 23 is replaced, the calibration data creation processing means is operated, making it possible to reliably generate calibration data that is compatible with the new device configuration.

[0131] Additionally, the calibration method of this embodiment is a calibration method for the article inspection device 1, which uses a calibration member Sp having a uniform thickness in the X-ray irradiation direction relative to the inspection area over the entire area in the arrangement direction of the plurality of sensor element modules M1 to Mk, to calibrate output signals Lx of N sensor elements e for X-rays that have passed through the calibration member Sp, and includes a first imaging step (step S21) of generating image data Ds1 of a first X-ray image based on output signals Lx1 of the N sensor elements e in a pre-load state where the article P is not within the inspection area Zx, and a second imaging step (step S22) of generating image data Ds1 of a first X-ray image based on output signals Lx2, Lx3 of the N sensor elements e in a post-load state where the calibration member Sp has been loaded into the inspection area Zx or in an X-ray reduced state where a virtual calibration member Sp has been loaded and set. 3, and a calibration data creation process (step S27) for generating calibration data Dhs for calibrating the output signals Lx1 of the N sensor elements e in the pre-loading state and the output signals Lx2, Lx3 of the N sensor elements e in the post-loading state to normal reference levels Rc1, Rc2, Rc3 corresponding to the correction target values ​​Ct0, Ctf, Cts in the respective states, based on the image data Ds1 of the first X-ray image and the image data Ds2, Ds3 of the second and third X-ray images and correction target values ​​Ct0, Ctf, Cts set in advance specific to the model.

[0132] Therefore, in the calibration method of this embodiment, by setting the correction target values ​​Ct0, Ctf, and Cts during calibration to model-specific values, variation in X-ray detection characteristics between individual item inspection devices is reduced, resulting in an item inspection method that enables uniform item inspection. Furthermore, not only is variation in the output characteristics of each sensor element module Mi and each of the multiple sensor elements e1-en of the X-ray detector 23 suppressed within each item inspection device, but calibration results are also standardized among other item inspection devices of the same model. Furthermore, by using a calibration member Sp that has X-ray transmission characteristics similar to those of the item P, item inspection with uniform high inspection accuracy is possible among multiple item inspection devices of the same model that inspect the same item P. Furthermore, the calibration process of the outputs of the N sensor elements e of the X-ray detector 23, which is performed based on the model-specific correction target values ​​Ct0, Ctf, and Cts, suppresses variation based on the model-specific correction target values ​​Ct0, Ctf, and Cts, enabling the sharing of image processing parameters among item inspection devices of the same model. In addition, by understanding the magnitude of variation in the calibration data Dhs itself, it becomes possible to effectively evaluate the degree of variation and changes in the output of the N sensor elements e in the X-ray detector 23 of each item inspection device based on the variation and changes in the calibration data itself.

[0133] Thus, according to this embodiment, a calibration method for an item inspection device is realized that can reduce the variation in characteristics between individual item inspection devices 1 and item inspection devices of the same model, and an item inspection device that can perform uniform item inspection with high inspection accuracy can be provided.

[0134] As a means for changing the incidence conditions from the X-ray generator 21 to the X-ray detector 23, at least one of the following can be used: a method for changing the incidence conditions of X-rays on the X-ray detector 23 by moving the calibration member Sp back and forth in a direction intersecting the X-ray passage path or changing its posture; and a method for changing the incidence conditions of X-rays for the sensor elements e1 to en of each sensor element module Mi of the X-ray detector 23 by changing the tube voltage or tube current supplied to the X-ray tube 22 of the X-ray generator 21.

[0135] Therefore, the calibration member Sp may be, for example, a flat plate made of a uniform material with an X-ray transmittance of less than 100% (for example, 50%) from the top to the bottom and approximately constant throughout the entire area in the specified scanning direction; a plate made of multiple layers of a uniform material with a constant thickness; a plate made of a constant material with a thickness that varies between the front and rear ends and whose X-ray transmittance varies as the thickness varies, for example, from 50 to 20%; or a plate made of different materials stacked together (for example, one with an X-ray transmittance of 50 to 20% and the other with an X-ray transmittance of 70 to 90%) to form a plate with a constant thickness.

[0136] Furthermore, for a single calibration member Sp, two incidence conditions can be set: a position where X-rays pass from the top surface to the bottom surface (X-ray transmittance 50%), and a position where X-rays pass through without intersecting (X-ray transmittance 100%).In addition, three incidence conditions can be set: a position where X-rays pass through from the top row to the bottom surface (X-ray transmittance 25%), a position where X-rays pass through from the middle row to the bottom surface (X-ray transmittance 50%), and a position where X-rays pass through without intersecting (X-ray transmittance 100%).

[0137] In the above-described embodiment, as a direct conversion type X-ray detector, when X-rays that have passed through the article P are incident on the semiconductor element 32 of each sensor element module Mi, electrons and holes (charges) are generated according to the energy of the incident X-rays (photons), thereby generating a detection signal of a voltage corresponding to the incident dose of X-rays between the opposing electrodes 31, 33, and outputting the signal to a readout circuit 34 composed of an ASIC or the like. However, since the X-rays from the X-ray generator 21 have large photon energy (short wavelength) and are not constant, it goes without saying that, rather than being limited to charge-voltage conversion by charge accumulation, a pulse signal of a peak value corresponding to the energy of the photon can be output each time an X-ray photon is incident on each sensor element, and the peak value can be measured by a photon-counting type readout circuit 34. In this case, the predetermined energy range from the X-ray generator 21 is divided into a plurality of energy (wavelength) ranges by the peak value of the pulse signal, and the number of pulse signals incident per unit time for each pixel electrode in the pixel electrode arrangement layer 33 in the energy range effective for inspection is counted, and the counted value can be made to correspond to the image density or brightness for each pixel of the X-ray image.

[0138] As described above, the present invention realizes a method for calibrating an article inspection device that can reduce variations in X-ray detection characteristics between individual article inspection devices, and can provide an article inspection device that can perform uniform article inspection with high inspection accuracy. The present invention is useful for article inspection devices equipped with X-ray detectors and for general calibration methods for article inspection devices that calibrate the output of such X-ray detectors. [Explanation of symbols]

[0139] 1. Item inspection equipment 10 Conveying section 11 Conveyor belt 12,13 Conveyor roller 20 Inspection unit (imaging unit) 21 X-ray generator (X-ray irradiation source) 22 X-ray tube 23 X-ray detector 24 Item detection sensor 31 Bias electrode (one of the opposing electrodes) 32 Semiconductor elements 33 Alignment layer (alignment layer of multiple pixel electrodes; the other electrode of the opposing electrode) 34 Readout circuit 40 Control Unit 41 Image data generation unit (inspection control unit) 42 Image data storage unit (inspection control unit) 43 Image processing unit (inspection control unit) 43a Calibration member detection processing unit 43b Calibration processing section 44 Good / bad judgment unit (inspection control unit) 45 Operating condition setting section (calibration control section) 45a Setting means (multi-stage setting means) 45b Operating condition adjustment means (output adjustment means) 46 Calibration data creation processing unit (calibration control unit; calibration data creation processing means) 47 Calibration data storage unit (calibration control unit; calibration data creation processing means, memory) 48 Mode switching unit (mode switching means) 50 Display operation section A1(i,j),A2(i,j),A3(i,j) Image density Ct correction target value (model-specific correction target value, image density correction target value) Ct0 First correction target value (First correction target value) Ctf Second correction target value (Second correction target value) Cts correction target value (third correction target value) d1 Conveying direction d2 Line scan direction d3 X-ray irradiation direction Dhs calibration data Ds1, Ds2 image data e1~en Sensor element (multiple sensor elements) Ez1, Ez2, Ez3 sensor element output section G1 signal (detection signal input) G2 signal (control signal output) H(i,j) compensation term Lx X-ray detection signal (image detection signal, output signal, sensor element output) Lx(i,j) Measured value (characteristic curve of each sensor element, measured output value) Lxa(i,j) Reference characteristic curve (reference characteristic value of each sensor element) Lx1 measurement value (measurement value before loading, output, sensor element output, image detection signal) Lx2 measurement value (measurement value, output, sensor element output, image detection signal after the first loading state) Lx3 measurement value (measurement value, output, sensor element output, image detection signal after the second loading state) Lxa1(i,j),Lxa2(i,j),Lxa3(i,j) Reference characteristic values M1~Mk Sensor element module (multiple sensor element modules) Mi Sensor Element Module (any one Sensor Element Module) P Article (object to be inspected) Rc1, Rc2, Rc3 reference level Sp calibration material Trm,Trm1,Trm2 X-ray transmittance ΔLx difference (difference between the sensor element output and the reference characteristic value)

Claims

1. an imaging unit (20) that irradiates an inspection area (Zx) into which an object (P) to be inspected (P) is carried with X-rays (Xfb), and detects the X-rays that pass through the inspection area with an X-ray detector (23) having a plurality of sensor elements adjacent in a predetermined arrangement direction, thereby generating image data of an X-ray image; an inspection control unit (41 to 44) that inspects the object to be inspected using image data generated by the imaging unit and a preset judgment criterion; and a calibration control unit (45 to 47) that uses a calibration member (Sp) having a uniform thickness (Tf) in an X-ray irradiation direction (d3) relative to the inspection area over the entire area of ​​the arrangement direction (d2) of the plurality of sensor elements to calibrate output signals (Lx) of the plurality of sensor elements in response to X-rays that have passed through the calibration member, The calibration control unit a first image processing means (S21) for generating image data (Ds1) of a first X-ray image based on output signals from the plurality of sensor elements in a pre-carry-in state in which the object to be inspected is not within the inspection area; a second image processing means (S22 and S23) for generating image data (Ds2) of a second X-ray image based on output signals from the plurality of sensor elements in a state after the calibration member has been carried into the inspection area; and a calibration data creation processing means (46 and 47) for generating calibration data (Dhs) for calibrating the output signals (Lx) of the plurality of sensor elements in the pre-loading state and the output signals of the plurality of sensor elements in the post-loading state to reference levels corresponding to the correction target values ​​in each state, based on the image data of the first X-ray image and the image data of the second X-ray image and a correction target value (Ct) previously set specific to the model.

2. 2. The object inspection device according to claim 1, further comprising a mode switching means (48) capable of switching between an inspection mode in which the object to be inspected is inspected and a calibration mode in which the output signals of the plurality of sensor elements are calibrated using the calibration member.

3. a calibration processing unit (43b) that executes a first calibration process based on image data of the first X-ray image to adjust image densities (A1(i, j)) of each sensor element of the image data based on output signals from the plurality of sensor elements in the pre-carry-in state to a first correction target value (Ct0(i, j)), and executes a second calibration process based on the image data of the first X-ray image and the image data of the second X-ray image to adjust image densities (A2(i, j)) of each sensor element of the image data based on output signals from the plurality of sensor elements in the post-carry-in state to a second correction target value (Ctf(i, j)); 3. The object inspection device of claim 1, further comprising: a memory (47) in which at least the second correction target value of the first correction target value and the second correction target value is set and stored as a correction target value specific to the model in advance of the first calibration process and the second calibration process.

4. the imaging unit has an X-ray tube (22) as a part of an X-ray irradiation source (21) that generates the X-rays, and the calibration processing unit executes a third calibration process that aligns an image density (A3(i, j)) of each sensor element of the image data based on output signals of the plurality of sensor elements to a third correction target value (Cts(i, j)); The object inspection device described in claim 3, characterized in that an output adjustment means (45b) is provided that adjusts the tube voltage of the X-ray tube when the calibration data creation processing means performs the third calibration process, and adjusts the irradiation intensity of the X-rays so that the image density of each sensor element of the image data based on the output signals of the multiple sensor elements in the pre-load state approaches the third correction target value.

5. The object inspection device described in claim 4, characterized in that the imaging unit is capable of adjusting the tube voltage of the X-ray tube or the irradiation intensity of the X-rays in multiple stages, and the calibration control unit has a multi-stage setting means (45a) that sets the first correction target value to a multi-stage correction target value corresponding to each stage of the tube voltage or X-ray irradiation intensity among the multiple stages.

6. The object inspection device described in claim 1 or 2, characterized in that when the X-ray irradiation source (21) of the imaging unit or the X-ray detector is replaced, the calibration control unit operates the first imaging processing means, the second imaging processing means, and the calibration data creation processing means to generate the calibration data again.

7. an imaging unit (20) that irradiates an inspection area (Zx) into which an object (P) to be inspected is carried with X-rays (Xfb), and detects the X-rays that pass through the inspection area with an X-ray detector (23) having a plurality of sensor elements adjacent in a predetermined arrangement direction to generate image data of an X-ray image, and an inspection control unit (41 to 44) that inspects the object to be inspected using the image data generated by the imaging unit and a preset judgment criterion, the calibration method for an X-ray inspection apparatus comprising: a calibration member (Sp) having a uniform thickness (Tf) in the X-ray irradiation direction for the inspection area over the entire area in the arrangement direction of the plurality of sensor elements, and calibrating output signals (Lx) of the plurality of sensor elements in response to the X-rays that have passed through the calibration member, a first imaging step (S21) of generating image data (Ds1) of a first X-ray image based on output signals of the plurality of sensor elements in a pre-load state in which the object to be inspected is not within the inspection area; a second imaging step (S22 and S23) of generating image data (Ds2) of a second X-ray image based on output signals of the plurality of sensor elements in a state after the calibration member has been carried into the inspection area; and a calibration data creation step (S27) of generating calibration data (Dhs) for calibrating the output signals of the plurality of sensor elements in the pre-loading state and the post-loading state to reference levels corresponding to the correction target values ​​in each state, based on the image data of the first X-ray image and the image data of the second X-ray image and a correction target value (Ct) previously set specific to the model.

Images