JP2006105794A - X-ray inspection apparatus, X-ray inspection method, and X-ray inspection program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem wherein it is difficult to detect of a relatively large product with accuracy, such as a cast product that is used as a part of a car with high precision. <P>SOLUTION: In detecting the flaw of the cast product, the cast product is irradiated with X rays, to detect the transmitted X rays through the cast product and the transmitted X rays, in a case that the cast product has no flaw, are calculated on the basis of the detected transmitted X rays; while the detected transmitted X rays are compared with the transmitted X rays in the case the cast product has no flaw to detect the flaw of the cast product. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an X-ray inspection apparatus, an X-ray inspection method, and an X-ray inspection program.

Conventionally, in this type of X-ray inspection apparatus, an object is irradiated with X-rays to detect transmitted X-rays, and the detected transmitted X-rays are compared with transmitted X-rays in an ideal object. Thus, a defect or the like in the object is detected (for example, see Patent Document 1).
Japanese Patent Laid-Open No. 2004-12407

In the conventional X-ray inspection apparatus described above, it has been difficult to detect defects with high accuracy in relatively large products such as cast products used as automobile parts. That is, in the conventional X-ray inspection apparatus, data indicating transmission X-rays in an ideal target (normal product) is prepared in advance, but the transmission X-rays in a product different from the actual inspection target are used as a reference. This makes it difficult to verify with high accuracy. More specifically, since the size of the defect that causes a defect in the cast product is considerably smaller than the size of the cast product, the size of the defect can be a dimensional error range of the cast product. Therefore, it cannot be reliably determined whether the difference between the ideal object and the actual inspection object is caused by a dimensional error or a defect, and accurate quality determination becomes difficult.
The present invention has been made in view of the above problems, and an object thereof is to provide an X-ray inspection apparatus, an X-ray inspection method, and an X-ray inspection program capable of reliably detecting defects in a cast product.

  In order to achieve the above object, in the invention according to claim 1, the transmitted X-ray when the cast product is defect-free is calculated based on the transmitted X-ray transmitted through the cast product, and the calculated transmitted X-ray and actual The defect of the cast product is detected by comparing with the detected transmission X-ray. That is, the comparison object to be compared with the transmitted X-ray transmitted through the cast product is created based on the actual transmitted X-ray of the cast product. Therefore, even if a dimensional error has occurred in the cast product to be inspected as compared with a standard cast product, this dimensional error has no effect on defect detection. As a result, defects can be detected with high accuracy.

  Here, the X-ray irradiating means only needs to be able to irradiate the cast product with X-rays, and the cast product may be arranged in the X-ray irradiation direction so that the X-rays pass through the site to be inspected. . The transmitted X-ray detection means only needs to be able to detect X-rays that have passed through the cast product, and various configurations can be employed. For example, a configuration (X-ray image intensifier) in which visible light is received by a CCD after X-rays are received by a scintillator and visualized can be employed. Of course, the CCD may be two-dimensionally arranged, or scanning may be performed by a one-dimensionally arranged CCD.

  The defect-free X-ray calculating means only needs to be able to calculate the transmitted X-ray when the cast product is defect-free based on the transmitted X-ray actually transmitted through the cast product, and various configurations can be adopted. . That is, if there is a defect in the cast product, the amount of X-ray absorption at that portion decreases, and a spatial change occurs in the transmitted X-ray. The spatial change of the transmitted X-ray depends on the thickness of the cast product. It is steeper than change. Therefore, by selectively correcting a portion where the transmitted X-ray is spatially changed due to a defect and making the change equivalent to the surrounding spatial change, the transmitted X-ray when the cast product is defect-free is obtained. Can be calculated.

  Such correction can be performed by various methods. As in claim 2, a portion where the transmitted X-ray dose is changed due to the defect is detected, and the change caused by the defect is canceled. Can be realized. More specifically, a low-pass filter may be applied to the spatial change of the transmitted X-ray, an envelope may be formed from the spatial change of the transmitted X-ray, or the space of the transmitted X-ray The expansion / contraction process may be performed with respect to the change. Further, any combination of the above may be used.

  Here, the expansion processing is focused on the portion of the profile showing the spatial change of the transmitted X-ray, where the profile is changed due to the defect, and extracting the maximum profile value from the vicinity. This process replaces the profile value with the maximum profile value, and removes minute dents in the profile. Shrinkage processing is processing that extracts the minimum profile value from the vicinity of the profile value of interest and replaces the profile value to be processed with the minimum profile value, and returns the expanded profile to the original size. Can do. As described above, the spatial change of the transmitted X-ray due to the defect is steeper than the change due to the thickness of the cast product, so that both can be clearly distinguished, and only the defect existing in the cast product is present. It is also possible to detect defects present on the surface of the cast product.

  The defect detection means only needs to be able to detect a defect by comparing the detected transmitted X-ray with the calculated transmitted X-ray in the case of no defect. That is, it can be determined that there is a defect if there is a predetermined change in comparison with the transmitted X-ray when there is no defect. For example, if the difference between the two is taken, it is possible to extract a portion of the detected transmitted X-ray that has a larger amount of transmission compared to the case where there is no defect. It is possible to determine that there is a defect.

  Furthermore, it is possible to determine the quality of the cast product by detecting defects in the cast product. Therefore, as in claim 3, the distance between the detected defect and a predetermined reference position in the cast product is calculated, and the cast product is a non-defective product when the distance is equal to or greater than the predetermined reference distance. It is possible to adopt a configuration that determines that That is, assuming that the casting product is destroyed due to the presence of defects, the casting product can be made good if there is no defect that causes destruction.

  Whether or not such a defect exists can be evaluated by the distance between the defect (predetermined reference position) that can cause cracks in the cast product and the defect, and the distance and the distance can be determined in advance. Thus, it is possible to determine whether or not the cast product is a non-defective product. Such a predetermined reference position may be a position that can cause destruction due to the presence of a defect in the vicinity, and the reference position can be set at various positions in the cast product such as a portion where stress is concentrated. it can.

  As such a reference position, various other positions can be set. For example, the reference position can be set to the contour of the cast product. That is, if the defect and the outline of the cast product are separated from each other by a predetermined reference distance or more, a non-defective product can be obtained. The predetermined reference distance may be determined in advance and can be adjusted in view of the size, material, and reliability of the cast product.

  The contour of the cast product corresponds to the end face or the outer surface of the cast product. If it is determined that the predetermined reference position is set to the contour, it is not necessary to determine that the specific position of the contour is the predetermined reference position. In other words, defects in the cast product can exist throughout the cast product, but in order to evaluate whether this defect causes fracture in relation to the contour, the position of the defect and the contour of the cast product It is preferable to calculate the shortest distance. Therefore, when calculating the shortest distance, when the position of the defect and the outline of the cast product are connected by a line segment corresponding to the shortest distance, the end on the outline side of the line segment is the predetermined reference position. Become.

  Furthermore, in order to make an accurate pass / fail judgment, the size of the defect may be considered as in the fifth aspect. That is, in the determination means, when the distance between the detected defect and the predetermined reference position in the cast product is not more than the predetermined reference distance and the defect is not less than a predetermined size, the cast product is It is determined that the product is defective. Defects can cause destruction in the cast product, but small defects can be excluded from the cause of destruction. Therefore, it is preferable that a defective product is formed when the defect and the predetermined reference position are not more than a predetermined reference distance and the defect is not less than a predetermined size. Here, the predetermined size may be determined in advance, and can be adjusted in view of the size, material, and reliability of the cast product.

  Furthermore, in order to secure the quality of the cast product and suppress an increase in the number of processes in the X-ray inspection, a configuration in which a predetermined position is inspected as in claim 6 may be adopted. That is, in the X-ray irradiation means, an irradiation region adjustment unit that adjusts the relative positional relationship between the cast product and the X-ray irradiation region is configured, and a predetermined region in the cast product is determined by the irradiation region adjustment unit. Adjust so that it is located in the irradiation area. Here, the predetermined area is an area where the cast product can be destroyed by the presence of defects in the area. With this configuration, it is possible to determine the quality of a cast product with only necessary and sufficient steps.

  Furthermore, when determining the quality of a cast product based on the distance between the detected defect and a predetermined reference position in the cast product, it is possible to employ a configuration for easily performing an accurate quality determination. For example, in a configuration for analyzing a two-dimensional distribution of transmitted X-rays as in claim 7, a configuration in which a transmitted X-ray image is moved so that defect detection processing can be easily performed can be employed.

  That is, when the transmission X-ray detection means detects a transmission X-ray image showing a two-dimensional distribution of transmitted X-rays, it is possible to detect a two-dimensional distribution of defects based on the two-dimensional distribution of transmitted X-rays. After detecting the two-dimensional distribution of defects, it is necessary to compare the relative relationship between the defect and a predetermined reference position. If the predetermined reference position in the cast product exists at an arbitrary position, it is necessary to detect the predetermined reference position when performing this comparison.

  Therefore, the transmitted X-ray image is moved so that a portion corresponding to a predetermined reference position in the cast product in the transmitted X-ray image becomes a predetermined inspection position. That is, a coordinate system in which coordinate axes are fixed on a two-dimensional plane is set, and a fixed inspection position is determined in advance in this coordinate system. Then, by moving the transmitted X-ray image relative to the coordinate system, the inspection position and the predetermined reference position in the cast product are superimposed. As a result, the distance between the defect and the predetermined reference position can always be calculated by the same formula, and the distance between the two can be calculated very easily.

  For example, when a predetermined reference position in a cast product is set to a certain end face in the cast product, if the end face always exists at a specific inspection position, the straight line of the end face is always expressed by the same formula in the coordinate system. Can do. Therefore, if the position of the defect is expressed by the coordinate value of this coordinate system, it is possible to always calculate the distance between the defect by the same formula using the coordinate value of the defect and the formula indicating the straight line of the end face. . As described above, when the transmitted X-ray image is moved, it may be moved in any manner on the two-dimensional plane. That is, since the state of the object is expressed by the position and the orientation angle on the two-dimensional plane, the image can be moved by changing either or both of the position and the orientation angle.

  As a preferred configuration example for moving the image, a deviation between a predetermined reference position and the inspection position in the cast product may be detected by template matching as described in claim 8. That is, a reference X-ray image for a good casting product is created in advance, and this image is compared with the transmitted X-ray image. At this time, a coordinate system in which a predetermined reference position in the reference X-ray image and the predetermined inspection position coincide with each other as an initial position of the reference X-ray image and the coordinate axes are fixed on the two-dimensional plane. Confirm in advance.

  Then, if the reference X-ray image is sequentially moved in this coordinate system to detect whether or not it matches the transmitted X-ray image, the reference X-ray image is moved until they match. The amount corresponds to a deviation between a predetermined reference position in the cast product and the inspection position. Therefore, if the transmitted X-ray image is moved so as to compensate for this deviation, the transmitted X-ray image is adjusted so that the predetermined reference position in the transmitted X-ray image becomes a predetermined inspection position. It is possible to move.

  Furthermore, in order to perform a more accurate pass / fail judgment when performing the pass / fail judgment of the cast product based on the distance between the detected defect and the predetermined reference position in the cast product, the cast product and X It is possible to employ a configuration that corrects a shift in the relative positional relationship with the irradiation region of the line. That is, an X-ray irradiation means is provided with an irradiation area adjustment unit, and the relative positional relationship between the cast product and the X-ray irradiation area can be adjusted, and a transmission X-ray detection means is provided with a position shift detection part, A shift in the relative positional relationship between the cast product and the X-ray irradiation area is detected. Then, when a shift in the relative positional relationship is detected, the irradiation area adjustment unit performs adjustment to compensate for this shift.

  As a result, even if there is a deviation between the actual position and the predetermined position to be placed when the cast product is irradiated with X-rays, the relative positional relationship can be corrected and X-rays can be irradiated. Therefore, it is possible to detect a defect with high accuracy and to perform pass / fail judgment. Note that the two-dimensional positional deviation can be corrected by moving the transmitted X-ray image as described above. Therefore, it is preferable to correct the relative positional relationship between the cast product and the X-ray irradiation area when there is a deviation in the direction orthogonal to the two-dimensional plane in the transmission X-ray image.

  The position shift detection unit only needs to be able to detect that the relative positional relationship between the cast product and the X-ray irradiation region is shifted from a predetermined reference. Therefore, various configurations such as providing a sensor for detecting this shift can be adopted. However, as a preferred mode, a configuration for detecting the shift based on the transmitted X-ray image as in claim 10 can be used. . That is, since the transmitted X-ray image shows a two-dimensional distribution of transmitted X-rays, it is easy to detect a deviation in the two-dimensional plane, and an image of a cast product in the transmitted X-ray image. Based on this, it is also possible to detect a shift in a direction orthogonal to the two-dimensional plane.

  For example, as described above, when performing template matching for comparing a reference X-ray image of a non-defective casting product with the transmission X-ray image, the casting product transmits the transmission X when the degree of coincidence between the two does not exceed a predetermined reference. It is determined that the line image is shifted in a direction orthogonal to the two-dimensional plane. If a shift in this direction can be detected, it is possible to reacquire a transmission X-ray image in which the influence of the shift is suppressed by correcting the posture of the cast product. As a result, it becomes possible to carry out defect detection and pass / fail judgment more accurately.

  Furthermore, when it is difficult to detect a defect as in claim 11, it is possible to detect the defect again by changing the conditions for the analysis. That is, the X-ray irradiation condition, the X-ray detection condition in the transmission X-ray detection means, the transmission X-ray calculation method when there is no defect, and the relative positional relationship between the cast product and the X-ray irradiation region Any of the above or a combination thereof can be adjusted, and this adjustment is performed when a difficult detection condition that makes it difficult to detect a defect occurs.

  Various conditions can be assumed as the difficult detection conditions. For example, calculation is performed when the distance between the defect detected by the determination unit and the predetermined reference position in the cast product is compared with the predetermined reference distance. Including measurement variation of the measured distance may not be able to determine whether the distance is longer or closer than the reference distance. In this case, it can be said that it is difficult to detect defects. Moreover, it may be difficult to detect depending on the position of the defect. That is, the transmitted X-ray reflects the thickness of the cast product, but if there is a defect at a site where the thickness of the transmitted X-ray changes greatly, the change in the transmitted X-ray is caused by either the defect or the thickness of the cast product. It becomes difficult to determine whether it is caused. Therefore, in this case as well, it can be said that it is difficult to detect defects.

  Thus, when it is difficult to detect a defect, the S / N ratio can be adjusted by changing the X-ray irradiation conditions, for example, the irradiation time, the X-ray energy, etc., so that the above-described measurement variation is suppressed. It becomes possible. In addition, the measurement variation can be suppressed even by adjusting the X-ray detection conditions, for example, the resolution at the time of detection. Further, if the calculation method of the transmitted X-ray in the defect-free X-ray calculating means is changed, the result calculated as the transmitted X-ray when there is no defect is different, so that it is possible to easily determine the defect that is difficult to determine. It is also possible to make it.

  Furthermore, since the posture of the cast product can be changed by changing the relative positional relationship set by the irradiation area adjustment unit, the cast product can be oriented so that defects can be easily identified. In other words, transmission X-rays provide information that reflects the two-dimensional thickness of the cast product, so that even if it is difficult to detect defects when X-rays are irradiated to a cast product in a certain posture, the posture of the cast product is changed. If X-rays are irradiated instead, the defect can be moved to a position where it can be easily detected.

  Although the case where the present invention is realized as an apparatus has been described above, the present invention can also be applied to a method for realizing such an apparatus. As an example thereof, the invention according to claim 12 is configured to implement the method corresponding to claim 1. Of course, the substantial operation is the same as that of the apparatus described above. A method corresponding to claims 2 to 11 can also be configured. Such an X-ray inspection apparatus may be realized by itself, applied to a certain method, or used in a state where the same method is incorporated in another device. Not only this but various aspects are included. Therefore, it can be changed as appropriate, such as software or hardware.

  In the case of software for controlling the above method as an embodiment of the idea of the invention, it naturally exists and is used also on the recording medium on which the software or software is recorded. As an example, the invention according to claim 13 is configured to realize the function corresponding to claim 1 by software. Of course, software corresponding to claims 2 to 11 can also be configured.

  The software recording medium may be a magnetic recording medium, a magneto-optical recording medium, or any recording medium that will be developed in the future. The duplication stage of the primary reproduction product and the secondary reproduction product is equivalent without any question. In addition, even when the communication apparatus is used as the supply device, the present invention is not changed. Further, even when a part is software and a part is realized by hardware, the idea of the invention is not completely different, and a part is stored on a recording medium and is appropriately changed as necessary. It may be in the form of being read.

Here, embodiments of the present invention will be described in the following order.
(1) Configuration of the present invention:
(2) X-ray inspection process:
(3) Inspection position correction processing:
(4) Pass / fail judgment processing:
(5) Other embodiments:

(1) Configuration of the present invention:
FIG. 1 is a schematic block diagram of an X-ray inspection apparatus according to the present invention. In the figure, the X-ray inspection apparatus is composed of an X-ray imaging mechanism unit 10 and an X-ray imaging control unit 20. The X-ray imaging mechanism unit 10 includes an X-ray generator 11, a positioning mechanism 12, and an X-ray detector 13. The X-ray imaging control unit 20 includes an X-ray control unit 21, a positioning mechanism control unit 22, a transmitted X-ray image acquisition unit 23, a CPU 24, an input unit 25, an output unit 26, and a memory 27. In this configuration, the CPU 24 can execute a program (not shown) recorded in the memory 27, control each unit, and perform predetermined arithmetic processing.

  The memory 27 is a storage medium capable of storing data, in which threshold data 27a, examination site data 27b, imaging condition data 27c, and reference X-ray image data 27d are recorded. The threshold data 27 a is data indicating various threshold values that are referred to in the processing described later, and is determined in advance and recorded in the memory 27. The inspection part data 27b is data indicating a part where a defect is to be detected in a casting product to be inspected. The imaging condition data 27c is data indicating conditions when X-rays are generated by the X-ray generator 11, and includes an applied voltage to the X-ray tube, imaging time, and the like. The reference X-ray image data 27d is X-ray image data that is referred to in an inspection position correction process to be described later, and a transmission X-ray image is generated by accurately arranging each inspection site of the cast product 12a in the X-ray irradiation region. This is the image data when The memory 27 only needs to be able to store data, and various storage media such as a RAM, an EEPROM, and an HDD can be employed.

  The X-ray control unit 21 can generate predetermined X-rays by referring to the imaging condition data 27c and controlling the X-ray generator 11. The positioning mechanism control unit 22 is connected to the positioning mechanism 12 and controls the positioning mechanism 12. The positioning mechanism 12 is a multi-axis robot, clamps the cast product 12a, arranges the cast product 12a at a desired position, and takes a desired posture. The multi-axis robot in the present embodiment is configured to perform at least an operation of transporting the cast product 12a to a desired position and an operation of changing the posture. Further, the positioning mechanism control unit 22 refers to the inspection part data 27b stored in the memory 27 and conveys the casting product 12a so that the designated part in the casting product 12a is included in the X-ray irradiation region. be able to.

  The transmitted X-ray image acquisition unit 23 is connected to the X-ray detector 13, and detects the intensity of the X-ray transmitted through the casting product 12 a based on the detection value output from the X-ray detector 13. The X-ray detector 13 in the present embodiment includes a two-dimensionally distributed sensor, and can generate transmission X-ray image data indicating a two-dimensional X-ray distribution from the detected X-rays. In the present embodiment, the transmitted X-ray image acquisition unit 23 expresses the intensity of X-rays in shades, and performs image processing so that the intensity decreases as the intensity decreases.

  The output unit 26 is a display that displays the transmitted X-ray image and the like in the CPU 24, and the input unit 25 is an operation input device that receives user input. That is, the user can execute various inputs via the input unit 25, and various output results obtained by the processing of the CPU 24, transmitted X-ray image data, quality determination results of the cast product 12a, and the like are output to the output unit 26. Can be displayed.

  The CPU 24 can execute predetermined arithmetic processing in accordance with various control programs stored in the memory 27, and in order to perform defect inspection and pass / fail judgment of the cast product 12a, the inspection position correction processing unit 24a and the model image shown in FIG. The calculation unit 24b, the defect detection unit 24c, and the pass / fail determination unit 24d are operated. The inspection position correction processing unit 24a performs processing for compensating for position control deviation caused by the positioning mechanism 12. Based on the transmission X-ray image data acquired by the transmission X-ray image acquisition unit 23, the model image calculation unit 24b generates a transmission X-ray image (hereinafter referred to as a model image) when the casting product 12a is defect-free. The model image data shown is generated. The defect detector 24c detects a defect in the cast product 12a based on the difference between the transmission X-ray image data and the model image data. The quality determination unit 24d determines whether the cast product 12a is a good product or a defective product based on the detected defect.

(2) X-ray inspection process:
In the present embodiment, defect detection and casting product quality determination are performed according to the flowchart shown in FIG. When the casting product 12a is clamped in the positioning mechanism 12 and the inspection process is started, the CPU 24 acquires the inspection site data 27b and transfers it to the positioning mechanism control unit 22 in step S100. The positioning mechanism control unit 22 refers to the inspection site data 27b, and sets so that one of a plurality of predetermined inspection sites in the cast product 12a is located in the X-ray irradiation region.

  When one place to be examined is set in the X-ray irradiation area in step S100, the CPU 24 acquires the imaging condition data 27c in step S105 and transfers it to the X-ray control unit 21. The X-ray control unit 21 sets conditions in the X-ray generator 11 according to the imaging condition data 27c, and irradiates X-rays. The X-rays irradiated by the setting in step S105 pass through the cast product 12a and reach the X-ray detector 13. Therefore, the CPU 24 controls the transmission X-ray image acquisition unit 23 to acquire transmission X-ray image data.

  The position where the cast product 12a is set in step S100 is determined in advance by the inspection part data 27b. However, the inspection part is present in all the cast products 12a due to the control accuracy by the positioning mechanism 12 and the product variation in the cast product 12a. It is not always accurately set at a specific position. Therefore, in order to eliminate the influence of such misalignment in step S110, an inspection position correction process is performed. As a result, it is no longer required that the examination site be arranged at a specific position with high accuracy, and the positioning mechanism 12 can be configured simply. Details of the inspection position correction process will be described later.

  After the position accuracy is compensated as necessary by the above inspection position correction processing, a part of the captured transmission X-ray image data is cut out as an analysis target. FIG. 3 is an explanatory diagram for explaining an example in which an image obtained by capturing an image of a cylindrical boss with X-rays is cut out. Hereinafter, the processing shown in FIG. 2 will be explained based on this example. FIG. 3A schematically shows a part of a transmission X-ray image obtained by imaging a cylindrical boss from a direction orthogonal to the cylinder axis. In the transmitted X-ray image, the thicker the cast product, the smaller the transmitted X-ray dose. In the present embodiment, as described above, the image is created such that the lower the X-ray dose, the brighter the image.

  Accordingly, when a cylindrical boss is imaged from a direction perpendicular to the cylinder axis, the central portion corresponding to the hollow portion becomes darker than the left and right portions. The right side of FIG. 3A shows a one-dimensional change of the transmitted X-ray image at the position P. As shown in this graph, the transmitted X-ray dose is large at the center and the transmitted X-ray dose is decreased at the left and right. In addition, if there is a defect in the cast product 12a, X-rays are not absorbed when passing through the defect, so that the transmitted X-ray dose is larger (darker) than the surroundings. That is, in the transmitted X-ray image, an image like V shown in FIG. 3A or a corresponding graph is formed.

  In a situation where such a transmitted X-ray image is cut out, the CPU 24 executes processing in the model image calculation unit 24b in step S115 to create model image data. That is, the transmission X-ray image includes a change that is steeper than the change in thickness of the cast product 12a. By reducing this change and making it uniform, there is no defect, that is, only the change in thickness of the cast product 12a. Create a model image that reflects. Various kinds of processing can be adopted for this processing, and a low-pass filter may be applied to the transmission X-ray image shown in FIG. 3A, or an envelope is formed from a spatial change of the transmission X-ray image. Alternatively, the above-described expansion / contraction process may be performed on the spatial change of transmitted X-rays.

  FIG. 3B shows a model image created from FIG. 3A. The one-dimensional change of the model image at the position P is shown on the right side of FIG. 3B. Thus, in the model image, the spatial change including the influence of the defect is made uniform and smooth. Therefore, if the difference between this model image and the actually captured X-ray image is calculated, only a steep change in the transmitted X-ray image can be extracted. Therefore, the CPU 24 executes a process in the defect detection unit 24c, and calculates a difference between the model image data and the transmission X-ray image data in step S120.

  FIG. 3C shows the data thus obtained. As shown in this figure, in the difference result, a finite difference result is obtained at a plurality of positions reflecting the variation of the transmission X-ray image data. Therefore, in the present embodiment, it is assumed that a threshold value T is set in advance, and a defect has occurred at a position having a value exceeding the threshold value in the difference result. For this reason, in step S125, the defect detection unit 24c extracts a position having a finite value from the difference result, and compares the difference value of each position with the threshold value T. Then, the image is binarized with “1” for positions exceeding the threshold T, “0” for positions not exceeding the threshold T and positions where the difference value is “0” or less. The threshold value T is predetermined as the threshold value data 27a.

  FIG. 3D is a binarized image obtained as a result. The defect detection unit 24c further refers to the binarized image and measures the size and position of the defect in step S130. That is, the position of the defect is acquired from the position of the pixel that is “1” in the binarized image, and the maximum diameter of the pixel that is “1” in the binarized image is acquired as the defect size. As described above, in the present invention, the defect is detected using only the actually measured transmission X-ray image, and the dimensional error or the positional deviation between the transmission X-ray image and the model image when casting the cast product 12a. It is possible to detect a defect without being affected by this.

  That is, if a model image created in advance and an actually measured transmission X-ray image are compared, it is necessary to accurately specify the corresponding positions in the model image and the transmission X-ray image, and the corresponding positions of the two can be determined without deviation. It is difficult to identify. If a model image is created in advance, an average image must be created based on a non-defective product of the cast product 12a, and it is impossible to reflect a dimensional error in each cast product 12a. Therefore, it is difficult to accurately detect defects using a model image created in advance. However, in the present invention, the defect can be detected with very high accuracy because it is generated from the actually measured transmission X-ray image.

  When the size and position of the defect are measured, the CPU 24 executes processing in the quality determination unit 24d in step S135 to determine quality of the cast product 12a. Details of the pass / fail judgment will be described later. As described above, when a defect is detected with high accuracy by detecting the defect based on the actually measured transmission X-ray image, the accuracy of the pass / fail judgment can be significantly improved. That is, in the quality determination, as described later, the quality is determined based on the distance between the predetermined reference position and the defect in the cast product 12a. Therefore, the quality determination is performed accurately if the position of the defect is not accurate. However, since the position of the defect is accurately specified in this embodiment, the pass / fail judgment accuracy is also very high.

  As described above, after the pass / fail determination is performed, the CPU 24 refers to the inspection site data 27b in step S140, inspects the presence / absence of defects for all the predetermined inspection sites, and performs the pass / fail determination. Determine whether or not. If it is determined in step S140 that the inspection and the pass / fail determination have been performed for all the inspection parts, in step S100, the unexamined inspection part is set in the irradiation region, and the processes in and after step S105 are repeated.

(3) Inspection position correction processing:
Next, the inspection position correction process in step S110 will be described in detail. FIG. 4 is a flowchart of the inspection position correction process performed by the inspection position correction processing unit 24a. In this inspection position correction process, a part of the reference X-ray image data is extracted as a positioning area, and a template matching process is performed between the positioning area and the actually measured transmission X-ray image, whereby a deviation in the transmission X-ray image is obtained. Is detected. This process is performed while storing reference X-ray image data and transmission X-ray image data in a buffer secured in advance in the memory 27.

  The upper part of FIG. 5 shows an example of the reference X-ray image data, and the outer rectangle indicates the size of the buffer. Originally, the X-ray image should be a grayscale image, but FIG. 5 shows a schematic image so that the shape can be easily understood. The example shown in FIG. 5 shows a flange on a cylindrical boss Bo. It is a reference | standard X-ray image of the casting product 12a in which the shape site | part F and the reinforcement member R are connected. Note that a predetermined coordinate axis is set in advance in this buffer, and the position of each pixel in the image can be specified by coordinates. In the present embodiment, the left and lower end surfaces of the cast product 12a overlap the y-axis and the x-axis, respectively, and the cast product 12a is accurately placed in the X-ray irradiation region. Indicates that such transmission X-ray image data can be obtained.

  The lower part of FIG. 5 shows an example of actually measured transmission X-ray image data. Again, the outer rectangle indicates the size of the buffer and has the same capacity as the buffer storing the reference X-ray image data. Further, the x-axis and the y-axis are set as in the upper stage of FIG. In the lower part of FIG. 5, the cast product 12a is inclined with respect to the x-axis and the y-axis, and the left end face and the lower end face do not coincide with the axes. Therefore, the relative positional relationship between the cast product 12a and the X-ray irradiation area is deviated from the above-mentioned reference X-ray image data.

  In the inspection position correction process, the relative relationship between the transmission X-ray image of the cast product 12a and the x-axis and y-axis is such as shown in the upper part of FIG. It is the process which corrects to become. In order to perform this processing, in step S200, the inspection position correction processing unit 24a refers to the reference X-ray image data 27d, and generates reference X-ray image data corresponding to the inspection site that has been imaged in step S105. Extract and store in buffer. Then, a part thereof is extracted and set as a positioning region.

  Here, the positioning region may be extracted in order to compare the reference X-ray image data with the actually measured transmission X-ray image data, and a characteristic part as an image of the casting product 12a is extracted from the reference X-ray image data. That's fine. In the example shown in the upper part of FIG. 5, a rectangle T1 shown on the upper right side of the cast product 12a is the positioning region. In the present embodiment, template matching processing is performed in the transmitted X-ray image data while sequentially rotating the positioning region.

  Therefore, in step S205, an initial start angle (for example, 0 °) is set to the variable θ for defining the angle, and in step S210, the angle corresponding to the variable θ is rotated with respect to the reference X-ray image data. . In step S215, a rectangle circumscribing the rotated positioning region T1 is registered as a template T2 serving as a reference in the template matching process. This is shown in the middle part of FIG. That is, when the positioning area T1 indicated by the broken line rotates by θ, a rectangle circumscribing the positioning area T1 can be defined as a one-dot chain line. Therefore, the reference X-ray image data included in the one-dot chain line is registered as a template T2.

  In step S220, the transmission X-ray image data is scanned with the template T2, and template matching processing is performed. At each scan position, a score S indicating the degree of coincidence between the reference X-ray image data in the template T2 and the transmitted X-ray image data to be scanned is calculated. The score S may be an index for evaluating the degree of coincidence between the two, and in the present embodiment, the score S is set so that the numerical value increases as the two coincide. For example, it is possible to employ a configuration in which the difference between the two image data is calculated and the score S is the reciprocal of the sum of the differences. The scan range only needs to be set to a necessary and sufficient range, and the entire range of the transmitted X-ray image data need not be set as the scan range.

After calculating the score S at each position, in step S225, the maximum value is extracted from the score S, and the positions X _θ and Y _θ of the template T2 at that time are temporarily stored in the memory 27. Further, the score at that time is temporarily stored in the memory 27 as a score _Sθ . That is, the position where the template T2 and the transmitted X-ray image most closely coincide with the score is stored while being associated with the rotation angle θ. The position of the template T2 may be any position in the template T2 as long as the position can be specified. For example, coordinates indicating the center position of the template T2 can be adopted.

  In step S230, it is determined whether or not the variable θ has reached a preset maximum rotation angle. If it is not determined in step S230 that the variable θ has reached a preset maximum rotation angle, In step S235, a predetermined step angle is added to the current angle θ, and this is set as a new angle θ, and the processing in step S210 and subsequent steps is repeated.

When it is determined in step S230 that the variable θ has reached the preset maximum rotation angle, whether or not the score of the template matching process is within an allowable range, that is, the posture deviation of the measured casting product 12a is allowable. Whether it is a range or not is determined. Note that the deviation in the two-dimensional plane detected by the X-ray detector 13 can be compensated by the movement of the image, which will be described later. Therefore, here, whether or not the deviation occurs in a direction orthogonal to the two-dimensional plane. Is determined. Therefore, in step S240, the maximum value of the score S _theta at each angle is equal to or smaller than a predetermined threshold value. The threshold indicates the lower limit of the score S _theta, are recorded in the memory 27 as the threshold data 27a with is predetermined.

  FIG. 6 is an explanatory diagram for explaining that the accuracy of the quality determination can be lowered depending on the posture of the cast product 12a. In FIG. 6, the cast product 12 a is inclined by an angle α with respect to an ideal position I to be arranged by the positioning mechanism 12. In this case, the transmission X-ray image data of the cast product 12a changes gently at both ends as shown by the solid line in the lower part of FIG. On the other hand, when the cast product 12a is arranged at the ideal position I, the transmission X-ray image data has a sharp change at both ends as indicated by a broken line in the lower part of FIG.

  In this embodiment, the quality of the cast product 12a is determined based on the distance from the end surface of the cast product 12a to the defect, as will be described later. If the change of both ends is steep as shown by the broken line in the lower part of FIG. 6, the position of the end face can be specified accurately. However, if the influence of the end face exists over the range E as shown by the solid line, The position cannot be specified accurately. Therefore, the pass / fail judgment cannot be performed accurately.

Therefore, by setting the lower limit of the score _Sθ as described above, it is determined whether or not the cast product 12a is positioned at the ideal position I and has the ideal posture. That is, if a transmission X-ray image is captured at a position different from the ideal position I, a good score S should not be obtained even if the template matching process is performed. Using this, the attitude of the cast product 12a, which was determined by comparing the maximum value and the threshold value of the score S _theta template matching process is to determine whether a permissible range.

When the maximum value of the score S _theta at each angle is determined to be equal to or less than the predetermined threshold value at step S240, CPU 24 at step S245 to issue instructions to the positioning mechanism control unit 22, changes the attitude of the cast product 12a . And the process after step S105 shown in FIG. 2 is repeated. If it is not determined in step S240 that the maximum value of the score S _θ at each angle is equal to or less than a predetermined threshold, the angle θ at which the score S _θ is maximum is acquired in step S250, and the variable θ _max indicating the movement angle is obtained. Is temporarily stored in the memory 27.

In step S255, the transmitted X-ray image is moved based on the angle θ _max and the positions X _θ and Y _θ acquired as described above. That is, the transmitted X-ray image is rotated by “−θ _max ”, and further moved by “−X _θ ” in the x-axis direction and “−Y _θ ” in the y-axis direction. Thereby, as shown in the lower part of FIG. 5, the image of the cast product 12a inclined with respect to the x-axis and the y-axis can be brought into a state as shown in the upper part of FIG. After compensating for the deviation in the image of the cast product 12a in the two-dimensional plane in this way, in step S260, the image in the predetermined region to be inspected is cut out and the processing returns to the processing after step S115 in FIG. To do.

(4) Pass / fail judgment processing:
Next, the quality determination process in step S135 will be described in detail. FIG. 7 is a flowchart of the quality determination process performed by the quality determination unit 24d. When the pass / fail judgment is performed, an image in the region to be inspected is cut out from the transmission X-ray image, and a binary image indicating a defect and a portion other than the defect is obtained in the cut out image. ing. Therefore, the quality is determined by calculating the distance between this defect and a predetermined reference position in the cast product 12a. In the present embodiment, the end face of the cast product 12a is adopted as the predetermined reference position.

  FIG. 8 is a view showing a cut-out image and a binarized image in the cast product 12a as shown in FIG. That is, when the transmission X-ray image is rotated by the position correction process and the end surface of the cast product 12a is arranged so as to overlap the y-axis and the x-axis as shown in the upper part of FIG. The image in the area is cut out. In the upper part of FIG. 8, this area is illustrated by a bold rectangle. When the processing of steps S115 to S130 is performed on this image, a binary image from which defects are extracted is obtained as shown in the lower part of FIG.

In step S300, the outline of the defect is calculated based on such a binarized image. That is, all pixels in which “1” and “0” are adjacent to each other in the binarized image are extracted, and the positions of the respective pixels are temporarily stored as (X _E (t), Y _E (t)). . When the contour pixel is extracted, a distance D (t) from the position (X _E (t), Y _E (t)) of each pixel to the end face line is calculated in step S305. Since the position of the end face line is determined in advance, the formula of the end face line is fixed, and the distance D (t) between the position of each pixel and the end face line can be calculated by a predetermined formula.

The end face line is one side of a rectangle indicating the region to be inspected, and in the lower part of FIG. 8, a straight line Y = Co that coincides with the upper side of the binarized image. When the distance D (t) from the position of each pixel (X _E (t), Y _E (t)) to the end face line is calculated, the minimum value D (t) of these distances D (t) is calculated in step S310. ) Extract min. In step S315, the maximum defect diameter Ra is calculated. For example, the distance between each pixel is calculated, and the maximum value is set as the maximum diameter Ra.

Next, in step S320, it is determined whether or not a difficult detection condition is satisfied. Here, the difficulty detection conditions are set in advance for the D (t) min and the contour position (X _E (t), Y _E (t)) of the defect. That is, it is determined whether or not D (t) min is equal to or greater than a predetermined threshold value by the subsequent processing. Here, whether or not the distance D (t) is greater than or equal to the threshold value in consideration of measurement variation. May become unknown. Therefore, in this case, the inspection condition is changed in step S325, and the processing after step S105 is repeated.

Further, the contour position of the site and defects thickness of the cast product 12a changes significantly when the _{(X E (t), Y} E (t)) is overlapped, the change in transmission X-ray image due to the defective It is difficult to determine whether it is caused by the thickness change of the cast product 12a. Therefore, even when the part where the thickness of the cast product 12a greatly changes and the outline position (X _E (t), Y _E (t)) of the defect overlap, the inspection condition is changed in step S325, and after step S105. Repeat the process. In addition, Rt shown to FIG. 3A etc. corresponds as a site | part where the thickness of the casting product 12a changes a lot.

  Here, various conditions can be adopted as inspection conditions to be changed. For example, the X-ray irradiation time or X-ray energy in the X-ray generator 11 may be changed, or the resolution at the time of detection in the X-ray detector 13 may be adjusted. Further, the model image calculation method in the model image calculation unit 24b may be changed. That is, as a calculation method of the model image, a method using a low-pass filter according to the quality of the image, a method of forming an envelope from a spatial change of the transmission X-ray image, and a spatial change of the transmission X-ray Since a technique for applying the expansion / contraction process can be selected, this technique is changed.

Furthermore, the positioning mechanism control unit 22 may be instructed to change the posture of the cast product 12a. This change is preferably applied when a portion where the thickness of the cast product 12a greatly changes overlaps with the defect contour position (X _E (t), Y _E (t)). For example, when a defect exists in the above-described Rt, the example shown in FIG. 3 is a transmission X-ray image of a cylindrical boss, but if the posture is changed so as to rotate the boss with respect to this cylindrical axis, It becomes possible to change to a position where the defect can be easily identified.

  If it is not determined in step S320 that the hard detection condition is satisfied, it is determined in step S330 whether or not the minimum value D (t) min is equal to or greater than a predetermined threshold value. If it is not determined in step S330 that the minimum value D (t) min is greater than or equal to the predetermined threshold value, it is further determined in step S335 whether or not the maximum defect diameter Ra is equal to or greater than the predetermined threshold value. When it is determined in step S335 that the maximum diameter Ra is equal to or larger than the predetermined threshold value, it is determined in step S340 that the cast product 12a is defective.

  That is, when a defect having a size larger than a certain threshold is present closer to the end face line than a certain allowable range, it is determined that the defect causes the casting product 12a to be broken and is determined to be a defective product. If it is determined in step S330 that the minimum value D (t) min is greater than or equal to a predetermined threshold value, or if it is not determined in step S335 that the maximum diameter Ra is greater than or equal to the predetermined threshold value, the cast product is determined in step S345. It is determined that 12a is a non-defective product. The threshold values in steps S330 and S335 are specified in advance and recorded in the memory 27 as threshold data 27a.

(5) Other embodiments:
In the present invention, it is only necessary to create a model image based on the actually measured transmission X-ray image, detect the defect, and determine whether or not the defect is good based on the distance between the defect and a predetermined reference position in the cast product. The configuration as described in the above embodiment is not essential. For example, if the positioning accuracy by the positioning mechanism 12 is good and the shift in the direction orthogonal to the two-dimensional image created by the X-ray detector 13 does not affect the accuracy of the quality determination, the above steps S240 and S245 are omitted. be able to.

  In addition, it is not essential that the inspection target area is determined in advance when determining pass / fail. For example, by detecting an end face that is a predetermined reference position in a cast product in transmitted X-ray image data and determining a straight line indicating the end face in the image, the distance between the straight line and the defect can be calculated. is there. Therefore, if this process is adopted, the inspection position correction process as in steps S200 to S260 is unnecessary.

  Furthermore, when the defect detection accuracy is high and the pass / fail determination accuracy is high, the processing for detecting whether or not the detection condition is difficult and changing the inspection condition as described above is not essential, and steps S320 and S325 are not necessary. Can be omitted. Furthermore, the criteria for performing the pass / fail judgment are not limited to the above criteria. For example, it is not essential to perform the pass / fail determination based on both the distance between the defect and the end face and the size of the defect, and the pass / fail determination may be performed based on the distance between the defect and the end face. Further, the predetermined reference position in the cast product is not limited to the end face, and may be a specific part where stress is concentrated.

1 is a schematic block diagram of an X-ray inspection apparatus according to the present invention. It is a flowchart of a X-ray inspection process. It is explanatory drawing explaining a mode that a transmitted X-ray is analyzed. It is a flowchart of an inspection position correction process. It is explanatory drawing explaining a mode that an image is corrected in a test | inspection position correction process. It is explanatory drawing for demonstrating that the precision of a quality determination may fall. It is a flowchart of a quality determination process. It is a figure which shows the example at the time of performing quality determination.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... X-ray imaging mechanism part 11 ... X-ray generator 12 ... Positioning mechanism 12a ... Casting product 13 ... X-ray detector 20 ... X-ray imaging control part 21 ... X-ray control part 22 ... Positioning mechanism control part 23 ... Transmission X Line image acquisition unit 24 ... CPU
24a ... Inspection position correction processing unit 24b ... Model image calculation unit 24c ... Defect detection unit 24d ... Pass / fail judgment unit 25 ... Input unit 26 ... Output unit 27 ... Memory 27a ... Threshold data 27b ... Inspection site data 27c ... Imaging condition data 27d ... Reference X-ray image data

Claims (13)

X-ray irradiation means for irradiating cast products with X-rays;
Transmitted X-ray detection means for detecting transmitted X-rays transmitted through the cast product;
Defect-free X-ray calculating means for calculating transmitted X-rays when the cast product is defect-free based on the detected transmitted X-rays;
An X-ray inspection apparatus comprising: defect detection means for detecting the defect of the cast product by comparing the detected transmitted X-ray with the calculated transmitted X-ray when there is no defect. The defect-free X-ray calculating means detects a portion where the transmitted X-ray dose is changed due to a defect, and cancels the change caused by the defect, thereby transmitting the transmitted X when the cast product is defect-free. The X-ray inspection apparatus according to claim 1, wherein a line is calculated. Based on the detected transmitted X-ray, a distance between the detected defect and a predetermined reference position in the cast product is calculated, and the cast product is a non-defective product when the distance is equal to or greater than a predetermined reference distance. The X-ray inspection apparatus according to claim 1, further comprising a determination unit that determines that there is one. The X-ray inspection apparatus according to claim 3, wherein the predetermined reference position is set to an outline of a cast product. The determination means determines that the cast product is defective when the distance between the detected defect and the predetermined reference position in the cast product is not more than the predetermined reference distance and the defect is not less than a predetermined size. The X-ray inspection apparatus according to claim 3, wherein the X-ray inspection apparatus is determined to be a non-defective product. The X-ray irradiation means includes an irradiation region adjustment unit that adjusts a relative positional relationship between the casting product and an X-ray irradiation region, and arranges a predetermined region in the casting product in the irradiation region to emit X-rays. The X-ray inspection apparatus according to claim 1, wherein irradiation is performed. The transmitted X-ray detection means generates a transmitted X-ray image showing a two-dimensional distribution of transmitted X-rays, and the determination means determines in advance a portion corresponding to a predetermined reference position in the cast product in the transmitted X-ray image. The X-ray inspection apparatus according to claim 1, wherein the transmission X-ray image is moved so that the inspection position is set. A reference X-ray image in a non-defective casting product is prepared in advance, and the determination means compares the reference X-ray image with the transmitted X-ray image, thereby determining a predetermined reference position in the casting product and the above-mentioned X-ray image. The X-ray inspection apparatus according to claim 7, wherein a deviation from the inspection position is detected. The X-ray irradiation means includes an irradiation area adjusting unit that adjusts a relative positional relationship between the casting product and an X-ray irradiation area, and the transmitted X-ray detection means is a relative position between the casting product and the X-ray irradiation area. A positional deviation detection unit that detects a deviation in the relationship is provided, and when the deviation in the relative positional relationship is detected by the positional deviation detection unit, the irradiation area adjustment unit performs adjustment to compensate for the deviation. The X-ray inspection apparatus in any one of Claims 1-8. The misregistration detection unit generates a transmission X-ray image indicating a two-dimensional distribution of the detected transmission X-rays, and the cast product is orthogonal to a plane on which the cast product forms the two-dimensional distribution based on the transmission X-ray images. The X-ray inspection apparatus according to claim 9, wherein the X-ray inspection apparatus is configured to detect whether or not the direction is shifted. The X-ray irradiation means includes an irradiation area adjustment unit that adjusts the relative positional relationship between the cast product and the X-ray irradiation area, and the defect detection means has a difficult detection condition that makes it difficult to detect a defect. In addition, the X-ray irradiation conditions in the X-ray irradiation means, the X-ray detection conditions in the transmission X-ray detection means, the transmission X-ray calculation method in the defect-free X-ray calculation means, and the irradiation region adjustment unit The X-ray inspection apparatus according to any one of claims 1 to 10, wherein any one or combination with the set relative positional relationship is changed. An X-ray inspection method for detecting defects in a cast product,
An X-ray irradiation process for irradiating cast products with X-rays;
A transmitted X-ray detection step of detecting transmitted X-rays transmitted through the cast product;
A defect-free X-ray calculation step of calculating transmitted X-rays when the cast product is defect-free based on the detected transmitted X-rays;
An X-ray inspection method comprising: a defect detection step of detecting a defect of the cast product by comparing the detected transmitted X-ray with the calculated transmitted X-ray when there is no defect. X-ray irradiation means for irradiating cast products with X-rays;
In a computer for controlling an X-ray inspection apparatus comprising transmitted X-ray detection means for detecting transmitted X-rays transmitted through the cast product,
A defect-free X-ray calculating function for calculating transmitted X-rays when the cast product is defect-free based on the transmitted X-rays detected by the transmitted X-ray detection means;
Realizing a defect detection function for detecting defects in the cast product by comparing the transmitted X-rays detected by the transmitted X-ray detection means with the calculated transmitted X-rays when there is no defect. A featured X-ray inspection program.

Images