Image correction method for x-ray internal inspection system and program

Abstract

To effectively remove only a characteristic streak-like noise in an X-ray transmission image appearing near a detection limit, and create a good X-ray transmission image that does not largely affect image information other than the streak-like noise.SOLUTION: An image correction method for an X-ray internal inspection system includes: picking up X-ray transmission images of a plurality of dummy analytes that are formed of the same component as a main component of an analyte and different in a thickness from each other to acquire dummy X-ray transmission images DD1-DD4; generating pieces of dummy data DC1-DC4 that are one-dimensional averaged data for the dummy X-ray transmission images DD1-DD4; calculating correction data DCA that is interpolation one-dimensional averaged data corresponding to an analyte with an arbitrary thickness d to be inspected through interpolation processing using the dummy data DC1-DC4; irradiating the analyte with an X-ray to acquire a pre-correction image that is a pre-correction X-ray transmission image; and executing correction using the correction data DCA on the respective pixels in the scanning direction of the pre-correction image to remove a streak-like noise DL from the pre-correction image and thereby create a post-correction image.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] The present invention relates to an image correction method and a program for an X-ray internal inspection system that can effectively remove only characteristic streak-like noise in an X-ray transmission image observed near the detection limit and generate a good X-ray transmission image that does not significantly affect image information other than the streak-like noise.

Background Art

[0002] In recent years, the aging of buildings such as bridges has become a problem, and maintenance is urgently needed. As one of the methods for non-destructively inspecting the interior of a building for this maintenance, there is X-ray transmission image inspection. Examples of this X-ray transmission image inspection include deterioration diagnosis of reinforced concrete and internal defect inspection of prestressed concrete. In this X-ray transmission image inspection, for example, an X-ray detector using an X-ray line sensor is used. An X-ray detector using an X-ray line sensor is often used in non-destructive inspections for security and industrial use.

[0003] For example, in Patent Document 1, an X-ray imaging system is proposed that includes an X-ray source that irradiates X-rays, a detector that detects X-rays transmitted through an imaging target subject, a semiconductor detector array that moves inside the detector, a semiconductor detector array drive unit that moves the semiconductor detector array in the vertical direction, and a signal processing circuit that processes a measurement signal measured by the detector and forms an image. This X-ray imaging system uses a semiconductor detector that is highly sensitive to high-energy X-rays.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] By the way, in an X-ray internal inspection system that uses an X-ray detector with an X-ray line sensor, the X-ray line sensor is scanned and the direction of the X-ray detector is controlled to the center of the X-ray source. In non-destructive X-ray inspection using this X-ray detector, in order to obtain a good X-ray transmission image, dark noise (dark data) when no X-rays are irradiated and air signals (air data), which are signals when X-rays are irradiated in the absence of an object, are acquired in advance, and signal correction is performed using each of these data. This makes it possible to cancel the signal variations inherent to the X-ray detector and X-ray source.

[0006] However, when inspecting the inside of thick concrete structures, such as bridges, the detectable signal level is close to the detection limit, so the effect of scattered X-rays from the concrete specimen itself cannot be ignored, and streak-like noise unique to the structure of the X-ray sensor's element arrangement and connections appears in the X-ray transmission image.

[0007] Here, as described in Patent Document 2, when trying to remove the above-mentioned streak-like noise from an X-ray transmission image using two-dimensional data as correction data, there are cases where unnecessary noise remains or information that should not be removed is removed.

[0008] The present invention has been made in consideration of the above, and has an object to provide an image correction method and program for an X-ray internal inspection system using an X-ray line sensor, which can effectively remove only the characteristic streak-like noise in an X-ray transmission image seen near the detection limit, and generate a good X-ray transmission image that does not significantly affect image information other than the streak-like noise. [Means for solving the problem]

[0009] In order to solve the above-described problems and achieve the object, an image correction method for an X-ray internal inspection system according to the present invention includes an X-ray source that irradiates an object with X-rays, and an X-ray line sensor in which a plurality of sensor elements for detecting X-rays are arranged. The object is placed between the X-ray source and the X-ray line sensor, and the X-ray line sensor moves in a scan direction that is perpendicular to the element arrangement direction while irradiating X-rays to perform an internal inspection of the object. In the internal inspection system, it is an image correction method for removing streak-like noise in the obtained X-ray transmission image, including: an imaging step of imaging X-ray transmission images of a plurality of dummy objects having substantially the same components as the main components of the object and different thicknesses to obtain a plurality of dummy X-ray transmission images; a dummy data generation step of generating dummy data, which is one-dimensional averaged data in the element arrangement direction obtained by averaging the pixel values of each pixel in the scan direction in each dummy X-ray transmission image; a correction reference data calculation step of calculating correction reference data that is interpolation one-dimensional averaged data corresponding to an object of an arbitrary thickness to be inspected by interpolation processing between the dummy data and functions as a calibration curve; a pre-correction image acquisition step of irradiating the object with X-rays to obtain a pre-correction image that is a pre-correction X-ray transmission image; a correction data creation step of specifying the thickness of the object and creating correction data corresponding to the thickness of the object using the correction reference data; and a post-correction image generation step of generating a post-correction image in which the streak-like noise is removed by performing correction on each pixel in the scan direction of the pre-correction image using the correction data.

[0010] Further, in the image correction method for an X-ray internal inspection system according to the present invention, in the above invention, the correction reference data is obtained in advance as a three-dimensional function having the thickness of the object as a parameter based on the dummy data, and the correction data creation step is characterized by creating correction data corresponding to an object of an arbitrary thickness to be inspected using the three-dimensional function.

[0011] Also, an image correction method for an X-ray internal inspection system according to the present invention includes an X-ray source that irradiates an object with X-rays, and an X-ray line sensor in which a plurality of sensor elements for detecting X-rays are arranged. The object is placed between the X-ray source and the X-ray line sensor, and the internal inspection of the object is performed by moving the X-ray line sensor in a scan direction that is perpendicular to the element arrangement direction while irradiating the object with X-rays. In the internal inspection system, it is an image correction method for removing streak-like noise in the obtained X-ray transmission image, including: a dummy image acquisition step of acquiring a dummy X-ray transmission image by imaging an X-ray transmission image of a dummy object composed of substantially the same components as the main components of the object; a dummy data generation step of generating dummy data, which is one-dimensional averaged data in the element arrangement direction obtained by averaging the pixel values of each pixel in the scan direction in the dummy X-ray transmission image; a pre-correction image acquisition step of irradiating the object with X-rays to obtain a pre-correction image that is a pre-correction X-ray transmission image; and a post-correction image generation step of generating a post-correction image in which the streak-like noise is removed by performing correction on each pixel in the scan direction of the pre-correction image using the dummy data.

[0012] Further, the program according to the present invention includes an X-ray source that irradiates a subject with X-rays, and an X-ray line sensor in which a plurality of sensor elements for detecting X-rays are arranged. The subject is placed between the X-ray source and the X-ray line sensor, and while irradiating the subject with X-rays, the X-ray line sensor moves in a scan direction that is perpendicular to the element arrangement direction, thereby performing an internal inspection of the subject. The control unit of the internal inspection system performs imaging of X-ray transmission images of a plurality of dummy subjects having the same components as the main components of the subject and different thicknesses, and acquires a plurality of dummy X-ray transmission images. A dummy image acquisition procedure, a dummy data generation procedure for generating dummy data that is one-dimensional averaged data in the element arrangement direction obtained by averaging the pixel values of each pixel in the scan direction in each dummy X-ray transmission image, and using the dummy data, a correction reference data calculation procedure for calculating correction reference data that is interpolated one-dimensional averaged data corresponding to a subject of an arbitrary thickness to be inspected and functions as a calibration curve by interpolation processing between the dummy data, a pre-correction image acquisition procedure for irradiating the subject with X-rays and acquiring a pre-correction image that is a pre-correction X-ray transmission image, a correction data creation procedure for specifying the thickness of the subject and creating correction data corresponding to the thickness of the subject using the correction reference data, and for each pixel in the scan direction of the pre-correction image, a post-correction image generation procedure for generating a post-correction image in which streak-like noise in the scan direction is removed by performing correction using the correction data.

Effect of the Invention

[0013] According to the present invention, it is possible to effectively remove only the characteristic streak-like noise in the X-ray transmission image seen near the detection limit, and to generate a good X-ray transmission image that does not significantly affect the image information other than the streak-like noise.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Embodiments for Carrying Out the Invention

[0015] Hereinafter, embodiments for carrying out this invention will be described with reference to the accompanying drawings.

[0016] <Overview of the X-ray Internal Inspection System> FIG. 1 is a perspective view showing a schematic configuration of an X-ray internal inspection system 100 according to an embodiment of the present invention. FIG. 2 is a side view of the X-Z plane of the X-ray internal inspection system 100 shown in FIG. 1. As shown in FIGS. 1 and 2, the X-ray internal inspection system 100 includes an X-ray source 5 that irradiates X-rays and an X-ray detector 1, and the X-ray source 5 and the X-ray detector 1 are arranged so as to sandwich a subject 6. The X-ray detector 1 detects the X-rays 7 that have passed through and attenuated the subject 6. The X-ray detector 1 includes a collimator 2 and an X-ray line sensor 3. The X-ray line sensor 3 has X-ray detection elements arranged in a one-dimensional manner in the horizontal direction (±Y direction), and detects the X-rays 7 input through the collimator 2. The collimator 2 is provided at the end of the X-ray line sensor 3 facing the X-ray source 5, and prevents various scattered rays from entering the X-ray line sensor 3. The collimator 2 forms a parallel gap that extends horizontally and is sandwiched between flat plates, as will be described later. Note that the collimator 2 may be a through hole formed horizontally for each element of the X-ray line sensor 3. Also, the gap of the collimator 2 does not have to be parallel.

[0017] The X-ray detector 1 is arranged inside an X-ray detection box 1a. The X-ray detector 1 is simultaneously moved when the X-ray detection box 1a moves inside the housing 4. This movement (scan) moves the X-ray detector 1 in the vertical direction (±Z direction) and rotates the X-ray detector 1 in the X-Z plane about an axis, that is, about an axis, with the pixel pitch direction (horizontal direction: element arrangement direction) of the X-ray line sensor 3 as the axis, so as to be inclined with respect to the horizontal plane (X-Y plane), and aligns the irradiation direction of the irradiated X-rays with the X-ray introduction direction of the collimator 2.

[0018] FIG. 3 is a perspective view showing the configuration inside the housing 4 with the outermost peripheral outer wall surface removed. The movement inside the housing 4 of the X-ray detector 1 described above is performed by a drive control mechanism. The drive control mechanism has drive parts 11a and 11b with a guide structure at the support parts at the four corners of the housing 4. Here, the drive part 11a can move the collimator 2 side up and down via the X-ray detection box 1a. Also, the drive part 11b can move the X-ray line sensor 3 side up and down via the X-ray detection box 1a. And these drive parts 11a and 11b can independently control the movement amount. Note that the drive parts 11a and 11b can move up and down on the guide lines 12 provided on the support parts of the housing 4 at the four corners of the X-ray detection box 1a.

[0019] The X-ray detector 1 sets the inclination at each height position as shown in FIG. 2 by controlling the movement amounts of the drive part 11a and the drive part 11b provided on the support of the housing 4, and is set so that the X-ray detector 1 is arranged on a straight line from the generation point of the X-ray source 5 at any height position.

[0020] FIG. 4 is a plan view of the configuration of the X-ray detector 1 and a diagram schematically showing the components connected thereto. Also, FIG. 5 is a cross-sectional view taken along line A-A of the X-ray detector 1 shown in FIG. 4. As shown in FIGS. 4 and 5, the end of the X-ray line sensor 3 facing the X-ray source 5 is arranged at the X-direction end of the parallel gap of the collimator 2. The X-ray line sensor 3 has a scintillator 31 and a photodetector 32. The scintillator 31 emits scintillation light by the incident X-ray. The scintillation light is detected by the photodetector 32 provided on the side part (±Z direction) of the scintillator 31. The scintillator 31 is provided with a separator corresponding to each photodetection element in which the photodetectors 32 are arranged in an array to prevent X-ray scattering and crosstalk. Note that the photodetector 32 attached to the side part of the scintillator 31 is not limited to the ±Z direction of the scintillator 31 and is arbitrary, for example, it may be in the +X direction.

[0021] The photodetector 32 is provided on the substrate 33, and the scintillator 31 is disposed above (in the Z direction) the photodetector 32. A reflective film is formed on the surfaces other than the photodetector 32 side so that scintillation light does not leak outside within the region divided by the separator. A light waveguide film is provided on the surface of the photodetector 32 side to guide the scintillation light to the photodetector 32.

[0022] The signal processing circuit 10 is connected to the other end side (in the X direction) of the substrate 33, and wiring for connecting the photodetector 32 and the signal processing circuit 10 is formed on the substrate 33. The signal processing circuit 10 amplifies the measurement signals from the respective elements of the photodetector 32, converts the amplified analog signals into digital signals, and generates an X-ray transmission image based on the converted digital signals. The X-ray line sensor 3 of the present embodiment has four X-ray line sensors connected in the element array direction. The number of element arrays of the X-ray line sensor 3 is, for example, 512. In the present embodiment, it is divided into four parts with 128 elements each, and four signal processing circuits 10 perform parallel processing on the divided measurement signals.

[0023] The X-ray transmission image output from the signal processing circuit 10 is output to the control device 20. The control device 20 performs image correction processing for removing streak-like noise on the acquired X-ray transmission image, and drives and controls the above-described drive units 11a and 11b. The control device 20 associates the drive control (scan control) of the drive units 11a and 11b with the position of the acquired X-ray transmission image. Here, the streak-like noise is specific noise that occurs in the scan direction due to the structure of the element array and connection parts of the X-ray line sensor 3 in the X-ray transmission image when the concrete thickness increases and the X-ray signal approaches the detection limit, and the influence of scattered X-rays from the concrete, which is the subject 6 itself, cannot be ignored.

[0024] FIG. 6 is a diagram showing the relationship between the X-ray 7 and the X-ray inlet of the collimator 2 accompanying the scan of the X-ray detector 1. As described above, the X-ray detector 1 is tilted in the X-Z plane as the X-ray detector 1 is scanned, and the X-ray inlet of the collimator 2 is directed toward the X-ray source 5 side. The one-dimensional data in the element array direction detected by the X-ray line sensor 3 becomes two-dimensional data by scanning in the scanning direction of the X-ray line sensor 3, and an X-ray transmission image is obtained.

[0025] <Control device> FIG. 7 is a block diagram showing the configuration of the control device 20. As shown in FIG. 7, the control device 20 includes an input unit 21, a display unit 22, a storage unit 23, and a control unit 24. The input unit 21 is an input device such as a mouse or a keyboard for inputting various setting information and the like. The display unit 22 is a display device such as a liquid crystal display for displaying various information. Note that the display unit 22 may be an operation display unit provided with a touch panel that also serves as the input unit 21. The storage unit 23 is a storage device for storing various information.

[0026] The control unit 24 is a control unit that controls the entire control device 20, and includes an image acquisition unit 25, a signal correction unit 26, an image correction unit 27, and a correction data generation unit 28. The control unit 24 stores programs corresponding to these functional units in a storage device such as a non-volatile memory or a magnetic disk device, loads these programs into the memory, and executes them by the CPU to execute the corresponding processes.

[0027] The image acquisition unit 25 acquires an X-ray transmission image (pre-correction image) input from the signal processing circuit 10. The signal correction unit 26 corrects the individual signal variations of the X-ray detector 1 and the X-ray source 5 based on the previously acquired air data DA1 and dark data DA2. The dark data DA2 is dark noise data when no X-ray is irradiated, and the air data DA1 is signal data when the X-ray is irradiated with no subject 6 present. The signal correction is a calibration process of the X-ray detector 1, and is a tone assignment process of equally dividing, by a predetermined number of tones, the value (signal difference) obtained by subtracting the dark data DA2 from the air data DA1 from the data obtained by subtracting the dark data DA2 from the data of each element. That is, the signal correction is a process of normalizing the signal difference of each element to the maximum tone value. Note that the air data DA1 mainly indicates the influence of the output variation of the scintillator 31. Also, the dark data DA2 mainly indicates the influence of the thermal noise of the photodetector 32, the electric circuit on the substrate 33, and the signal processing circuit 10.

[0028] The correction data generation unit 28 captures X-ray transmission images of a plurality of dummy subjects having the same components as the main components of the subject 6 but different thicknesses, obtains a plurality of dummy X-ray transmission images, and generates dummy data which is one-dimensional averaged data in the element array direction obtained by averaging each pixel in the scan direction in each dummy X-ray transmission image. Then, the correction data generation unit 28 calculates correction data DCA which is interpolated one-dimensional averaged data corresponding to a subject of an arbitrary thickness to be inspected by interpolation processing using the dummy data, and stores it in the storage unit 23.

[0029] Note that the dummy subject does not necessarily have to be the same components as the main components of the subject 6, and may be substantially the same components or those having the same or substantially the same X-ray attenuation characteristics. Substantially the same components means that, for example, if the difference is about the same as that of concrete of a different product number, it can be regarded as the same components as the main components of the subject 6 because the degree of X-ray attenuation is the same. Note that the degree of X-ray attenuation is greatly contributed by the density and mass of the subject 6. Therefore, when the density of the dummy subject is expressed in units of [g / cm 3 , if it matches the density of the subject 6 up to two decimal places, it can be regarded as obtaining substantially the same X-ray attenuation characteristics.

[0030] The image correction unit 27 performs a process of removing streak-like noise in the pre-correction image whose signal has been corrected. The image correction unit 27 corrects each pixel in the scan direction of the pre-correction image using the correction data DCA to generate a post-correction image in which the streak-like noise has been removed. The correction data DCA is correction data in which only the positions where the streak-like noise occurs are emphasized, and even if image correction is performed using this correction data DCA, there is almost no influence on the image area where the streak-like noise does not occur.

[0031] <Overall Image Correction Process> FIG. 8 is an overall flowchart showing the overall image correction process procedure by the control unit 24. As shown in FIG. 8, the control unit 24 first performs a correction data generation process of generating the correction data DCA for a subject with an arbitrary thickness d to be inspected (step S11). Thereafter, the control unit 24 performs an image correction process of correcting each pixel in the scan direction of the pre-correction image using the correction data DCA to generate a post-correction image in which the streak-like noise has been removed (step S12), and ends this process.

[0032] <Correction Data Generation Process> FIGS. 9 and 10 are explanatory diagrams for explaining the correction data generation process of generating the correction data DCA. First, as shown in FIG. 9(a), the air data DA1 in the element array direction is acquired based on the air signal S1 when irradiating X-rays in the state where the subject 6 is absent. Further, as shown in FIG. 9(b), the dark data DA2 in the element array direction is acquired based on the dark signal S2 when the X-rays are not irradiated. This air data DA1 and dark data DA2 are used for signal correction for performing tone assignment of each element in the element array direction.

[0033] Thereafter, as shown in FIG. 9(c), a plurality of dummy specimens 36 made of the same component (concrete) as the main component of the specimen 6 and having different thicknesses and not including structures 6a such as reinforcing bars are imaged to obtain dummy X-ray transmission images, and as shown in FIG. 9(d), a plurality of dummy X-ray transmission images DD1 to DD4 are acquired. The dummy X-ray transmission images DD1 to DD4 are generated based on the dummy X-ray transmission signal SD3.

[0034] Thereafter, as shown in FIG. 10(e), dummy data DC1 to DC4, which are one-dimensional averaged data in the element array direction obtained by averaging each pixel in the scan direction in each of the dummy X-ray transmission images DD1 to DD4, are generated. Note that the region for generating the dummy data DC1 to DC4, which are one-dimensional averaged data, is part or all of each of the dummy X-ray transmission images DD1 to DD4.

[0035] Then, the correction data generation unit 28 calculates correction data DCA, which is interpolated one-dimensional averaged data corresponding to the specimen 6 having an arbitrary thickness d to be inspected, by interpolation processing using the dummy data DC1 to DC4. This correction data DCA is obtained by interpolating or extrapolating the correction value for the specimen having the thickness d for each channel in the element array direction, for example, channel CH10, or by obtaining a function F10 and performing this interpolation processing for all channels.

[0036] <Correction Data Generation Processing Procedure> FIG. 11 is a flowchart showing the correction data generation processing procedure by the control unit 24. As shown in FIG. 11, first, air data DA1 is acquired (step S101). Thereafter, dark data DA2 is acquired (step S102). Thereafter, dummy X-ray transmission images DD1 to DD4 of a plurality of dummy specimens 36 are acquired (step S103), and signal correction and brightness / contrast adjustment using the air data DA1 and the dark data DA2 are performed on the dummy X-ray transmission images DD1 to DD4 (step S104). Note that the brightness / contrast adjustment is performed because the brightness of the obtained pre-correction image is often low, and thus it is necessary to increase the brightness of the entire image.

[0037] Thereafter, the correction data generation unit 28 receives a selection to set the entire area of each of the dummy X-ray transmission images DD1 to DD4 as the noise cancellation area (step S105), and generates dummy data DC1 to DC4, which are one-dimensional averaged data for each of the dummy X-ray transmission images DD1 to DD4 (step S106). Thereafter, the correction data generation unit 28 calculates correction data DCA, which is one-dimensional averaged data for the subject 6 with a thickness d, by interpolation processing based on the dummy data DC1 to DC4 (step S107), and returns to step S11.

[0038] <Image correction processing> FIGS. 12 and 13 are explanatory diagrams for explaining image correction processing including signal correction. First, as shown in FIG. 12(a), air data DA1 in the element array direction is acquired based on an air signal S1 when irradiating X-rays in a state where the subject 6 is absent. Further, as shown in FIG. 12(b), dark data DA2 in the element array direction is acquired based on a dark signal S2 when not irradiating X-rays. This air data DA1 and dark data DA2 are used for signal correction for performing tone assignment of each element in the element array direction. Note that the air data DA1 and dark data DA2 may be the air data DA1 and dark data DA2 during the correction data generation process.

[0039] Thereafter, as shown in FIG. 12(c), an X-ray transmission image is captured based on an X-ray transmission signal S3 of the subject 6 with a thickness d of the inspection target including the structure 6a, and as shown in FIG. 12(d), a pre-correction image D10, which is an X-ray transmission image, is acquired. In this pre-correction image D10, streak-like noise DL extending in the scan direction is generated, and an image of the structure 6a is included.

[0040] Thereafter, as shown in FIG. 13(e), the image correction unit 27 obtains correction data DCA for the subject 6 with a thickness d obtained by correction data generation processing with respect to the pre-correction image D11 whose signal has been corrected. As shown in FIG. 13(f), based on this correction data DCA, the pixel values of each pixel in the scan direction of the pre-correction image D11 are added, subtracted, or multiplied by a coefficient to remove streak noise. Specifically, since the correction data DCA at the position of the streak noise DL has a pixel value larger than the average value (value = 0), a process of adding or multiplying a negative pixel value with the sign reversed to all the pixels of the streak noise DL in the scan direction is performed. As a result, as shown in FIG. 13(g), the brightness of the streak noise DL becomes the average brightness, and the corrected image D12 is generated in which the emphasized streak noise DL is suppressed and removed. Note that the correction data DCA is interpolated based on the dummy data DC1 to DC4, and at positions other than the position of the streak noise DL, since it is close to the average value, the corrected image D12 has substantially the same brightness as the pre-correction image DC11 in the region other than the streak noise DL of the pre-correction image DC11.

[0041] Note that the correction data DCA may be calculated by performing an interpolation process based on the dummy data DC1 to DC4 immediately before the image correction for the pre-correction image D11 to calculate the correction data DCA for the subject 6.

[0042] Further, the correction data DCA may be obtained in advance as a three-dimensional function having the thickness d of the subject 6 as a parameter based on the dummy data DC1 to DC4, and immediately before the image correction for the pre-correction image D11, the correction data corresponding to the subject 6 with an arbitrary thickness d to be inspected may be calculated using the three-dimensional function.

[0043] <Image Correction Processing Procedure> FIG. 14 is a flowchart showing the image correction processing procedure by the control unit 24. As shown in FIG. 14, first, air data DA1 is acquired (step S201). Then, dark data DA2 is acquired (step S202). Then, an X-ray transmission image of the subject 6 to be inspected is acquired (step S203), and signal correction and brightness / contrast adjustment using the air data DA1 and the dark data DA2 are performed (step S204). Note that the brightness / contrast adjustment is performed because the brightness of the obtained pre-correction image is often low, and it is necessary to increase the brightness of the entire image. This brightness / contrast adjustment is determined based on the average value of the brightness of the pre-correction image D10.

[0044] After that, the image correction unit 27 acquires correction data DCA (step S205), performs image processing on the pre-correction image D11 using the correction data DCA, and generates a corrected image D12 from which streak-like noise DL has been removed (step S206). Then, the image correction unit 27 performs brightness / contrast adjustment on this corrected image D12 as necessary (step S207), and ends this process.

[0045] Note that the signal correction using the air data DA1 and the dark data DA2 does not need to be performed every time, but considering temperature changes, it is preferably performed every time.

[0046] FIG. 15 is a diagram showing an example of conventional image correction for removing streak-like noise DL. Conventionally, for example, as shown in FIG. 15(a), a dummy image D0 of the subject 6 without the structure 6a is acquired, and the dummy image D0 as two-dimensional correction data is subtracted from the pre-correction image D10 to generate a corrected image D12' shown in FIG. 15(b). This conventional image correction can surely remove the streak-like noise DL, but the information of the subject 6 made of concrete is also deleted, and only the image of the structure 6a remains.

[0047] In contrast, in the present embodiment, it is possible to remove streak noise DL without affecting the information of the specimen 6 which is concrete, and to obtain a good X-ray transmission image.

[0048] In the above embodiment, a plurality of dummy specimens with different thicknesses are used to obtain correction reference data which is interpolation one-dimensional averaged data corresponding to a specimen with an arbitrary thickness and functions as a calibration curve, create correction data according to the thickness of the specimen, and generate a corrected image in which streak noise is removed by correcting the pre-correction image of the specimen using this correction data. However, when there is a dummy specimen having the same thickness as the specimen, it is not always necessary to create correction data according to the thickness of the specimen, and a corrected image can be obtained by using one dummy data as it is as the correction data.

[0049] Also, each configuration illustrated in the above embodiment and modification example is schematic in function, and it is not necessarily physically configured as illustrated. That is, the form of dispersion and integration of each device and component is not limited to that illustrated, and all or part of them can be functionally or physically dispersed and integrated in arbitrary units according to various usage situations and the like.

Description of Reference Numerals

[0050] 1 X-ray detector 1a X-ray detection box 2 Collimator 3 X-ray line sensor 4 Housing 5 X-ray source 6 Specimen 6a Structure 7 X-ray 10 Signal processing circuit 11a, 11b Driving unit 12 Guideline 20 Control device 21 Input unit 22 Display unit 23 Storage unit 24 Control unit 25 Image acquisition unit 26 Signal correction unit 27 Image correction unit 28 Correction data generation unit 31 Scintillator 32 Photodetector 33 Substrate 36 Dummy specimen 100 X-ray internal inspection system D0 Dummy image D10, D11 Pre-correction images D12 Post-correction image DC1~DC4 Dummy data DCA Correction data DD1~DD4 Dummy X-ray transmission images DL Streak noise S1 Air signal S2 Dark signal S3 X-ray transmission signal SD3 Dummy X-ray transmission signal

Claims

1. An X-ray internal inspection system comprising an X-ray source that irradiates a subject with X-rays and an X-ray line sensor in which a plurality of sensor elements for detecting X-rays are arranged, wherein the subject is placed between the X-ray source and the X-ray line sensor, and the X-ray line sensor moves in a scan direction that is perpendicular to the element arrangement direction while irradiating the subject with X-rays to perform an internal inspection of the subject. An image correction method for removing streak-like noise in the obtained X-ray transmission image, comprising: a dummy image acquisition step of imaging X-ray transmission images of a plurality of dummy subjects having substantially the same components as the main components of the subject and different thicknesses to obtain a plurality of dummy X-ray transmission images; a dummy data generation step of generating dummy data which is one-dimensional averaged data in the element arrangement direction obtained by averaging the pixel values of each pixel in the scan direction in each dummy X-ray transmission image; a correction reference data calculation step of calculating correction reference data which is interpolation one-dimensional averaged data corresponding to a subject of an arbitrary thickness to be inspected by interpolation processing between the dummy data and functions as a calibration curve using the dummy data; a pre-correction image acquisition step of irradiating the subject with X-rays to obtain a pre-correction image which is a pre-correction X-ray transmission image; a correction data creation step of specifying the thickness of the subject and creating correction data corresponding to the thickness of the subject using the correction reference data; a post-correction image generation step of generating a post-correction image in which the streak-like noise is removed by performing correction on each pixel in the scan direction of the pre-correction image using the correction data; An image correction method for an X-ray internal inspection system, characterized by including the above steps.

2. The correction reference data is obtained in advance as a three-dimensional function using the thickness of the subject as a parameter based on the dummy data, The correction data creation step creates correction data corresponding to a subject of an arbitrary thickness to be inspected using the three-dimensional function. The image correction method for an X-ray internal inspection system according to Claim 1, characterized by this.

3. An internal inspection system comprising an X-ray source that irradiates a subject with X-rays and an X-ray line sensor in which a plurality of sensor elements for detecting X-rays are arranged, wherein the subject is placed between the X-ray source and the X-ray line sensor, and the X-ray line sensor moves in a scan direction that is perpendicular to the element arrangement direction while irradiating the subject with X-rays to perform an internal inspection of the subject. An image correction method for removing streak-like noise in the obtained X-ray transmission image, comprising: A dummy image acquisition step of imaging an X-ray transmission image of a dummy subject composed of substantially the same components as the main components of the subject to obtain a dummy X-ray transmission image; A dummy data generation step of generating dummy data which is one-dimensional averaged data in the element arrangement direction obtained by averaging the pixel values of each pixel in the scan direction in the dummy X-ray transmission image; A pre-correction image acquisition step of irradiating the subject with X-rays to obtain a pre-correction image which is a pre-correction X-ray transmission image; A post-correction image generation step of performing correction using the dummy data on each pixel in the scan direction of the pre-correction image to generate a post-correction image in which the streak-like noise is removed; An image correction method for an X-ray internal inspection system, characterized by including the above steps.

4. In a control unit of an internal inspection system comprising an X-ray source that irradiates a subject with X-rays and an X-ray line sensor in which a plurality of sensor elements for detecting X-rays are arranged, wherein the subject is placed between the X-ray source and the X-ray line sensor, and the X-ray line sensor moves in a scan direction that is perpendicular to the element arrangement direction while irradiating the subject with X-rays to perform an internal inspection of the subject, A dummy image acquisition procedure of imaging X-ray transmission images of a plurality of dummy subjects composed of the same components as the main components of the subject and having different thicknesses to obtain a plurality of dummy X-ray transmission images; A dummy data generation procedure of generating dummy data which is one-dimensional averaged data in the element arrangement direction obtained by averaging the pixel values of each pixel in the scan direction in each dummy X-ray transmission image; A correction reference data calculation procedure of calculating correction reference data that functions as a calibration curve and is interpolation one-dimensional averaged data corresponding to a subject of an arbitrary thickness to be inspected by interpolation processing between the dummy data using the dummy data; A pre-correction image acquisition procedure of irradiating the subject with X-rays to obtain a pre-correction image which is a pre-correction X-ray transmission image; A correction data creation procedure for specifying the thickness of the subject and creating correction data corresponding to the thickness of the subject using the reference correction data; A program for executing a corrected image generation procedure for generating a corrected image in which streak-like noise in the scan direction is removed by performing correction using the correction data on each pixel in the scan direction of the pre-correction image.

Images