JP2019074489A - Workpiece inspection device and workpiece inspection method - Google Patents

Abstract

To provide workpiece inspection technology with which it is possible to compare internal defects of a plurality of workpieces formed by the same die with one another in a short time.SOLUTION: A work-piece inspection device 1 is designed to inspect a plurality of workpieces W formed by the same die, and includes a processing device 20 for processing an X-ray photographed image C of each of the plurality of workpieces W. The processing device 20 comprises: a data extraction unit 23 for extracting blow hole voxel data D2 relating to a blow hole B as an internal defect from among pieces of three-dimensional volume data D1 based on the X-ray photographed image C; and a data combination unit 25 for combining spherical graphic voxel data Ds obtained from the blow hole voxel data D2 extracted for each of the plurality of workpieces W by the data extraction unit 23 with workpiece body data that indicates the whole shape common to the plurality of workpieces W at a combination position that is the original position of the blow hole B.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for inspecting a work.

  Conventionally, internal defects formed inside the workpiece are nondestructively for workpieces such as cast products in which metal materials such as aluminum, zinc and magnesium are cast in molds, and resin molded products in which resin materials are molded using molds Various devices for testing have been proposed.

  Patent Document 1 below discloses a technique for inspecting a casting, which is one of internal defects, by imaging a cast product with an X-ray CT imaging apparatus. This technology is capable of inspecting the pores formed inside the cast product regardless of its size by executing specific filtering on the 3D CT volume data obtained by CT imaging of the cast product. It is said that.

JP 2005-351875 A

  By the way, it is known that the cavities generated inside the cast product include the cavities resulting from the structure of the casting mold and the cavities generated according to the minute change of the actual casting conditions and the like. It is done. In the case of casting a plurality of castings with the same mold, the cavities generated randomly according to the change of the casting conditions are generated in the interior of the castings, whereas the cavities due to the structure of the casting die Occurs concentrated in a generally common area inside the casting.

  Therefore, if multiple samples cast with the same mold can be photographed with the above-mentioned X-ray CT imaging apparatus, and the cavities contained in the three-dimensional volume data from the radiographic image can be accurately compared among the samples, the cavities will be generated Can be related to the factors of the mold and the structure of the mold. As a result, it becomes possible to reflect the occurrence of the cavities in the structure of the mold, and the cost required for the development and improvement of the expensive mold can be kept low.

  However, when the three-dimensional volume data of a plurality of samples are statistically collected manually to compare the cavities, a large number of man-hours are required. For this reason, in order to keep man-hour low, it is possible to combine three-dimensional volume data of a plurality of samples on a system.

However, since the data volume of three-dimensional volume data per sample is large, it takes time to process a plurality of three-dimensional volume data on a system. In particular, since a large number of samples are required to ensure the above-mentioned association, it takes an enormous amount of time to combine three-dimensional volume data of these samples on the system.
Therefore, it is difficult to compare cavities between samples in a limited development period or improvement period of a mold. However, if the number of samples is reduced, then the accuracy of the above-mentioned association is lowered, so it is difficult to fulfill the original purpose of reflecting the occurrence of the cavities in the structure of the mold.
In addition, the above-mentioned defects may occur not only in the case of a cast product obtained by casting a metal material with a mold but also with a resin molded product obtained by molding a resin material with a mold.

  The present invention has been made in view of such problems, and an object of the present invention is to provide a workpiece inspection technique capable of comparing internal defects of a plurality of workpieces formed by the same mold in a short time. It is.

One aspect of the present invention is
A work inspection apparatus (1, 101) for inspecting a plurality of works (W) formed by the same mold (M),
A processing device (20, 120) for processing an X-ray image (C) of each of the plurality of works;
The above processing unit
A data extraction unit (23) for extracting internal defect voxel data (D2) relating to internal defects (B, B1, B2) from the three-dimensional volume data (D1) from the radiographic image;
The comparison image data (D3) obtained from the internal defect voxel data extracted by the data extraction unit for each of the plurality of workpieces is used as the workpiece body data (Dw) indicating the entire shape common to the plurality of workpieces. A data synthesizing unit (25) for synthesizing at a synthesis position (P, P1, P2) which is the original position of the internal defect;
A workpiece inspection apparatus (1, 101),
It is in.

Moreover, another aspect of the present invention is
A workpiece inspection method for inspecting a plurality of workpieces (W) formed by the same mold (M), comprising:
A data extraction step (S106) for extracting internal defect voxel data (D2) related to internal defects (B, B1, B2) from three-dimensional volume data (D1) based on X-ray images (C) of each of the plurality of workpieces S206),
The comparison image data (D3) obtained from the internal defect voxel data extracted in the data extraction step for each of the plurality of workpieces is compared to the workpiece body data (Dw) indicating the overall shape common to the plurality of workpieces Data synthesizing step (S112, S210) for synthesizing at the synthesis position (P, P1, P2) which is the original position of the internal defect;
Work inspection method,
It is in.

In the above-described workpiece inspection apparatus or workpiece inspection method, internal defect voxel data relating to internal defects is extracted from three-dimensional volume data of a radiographic image of the workpiece. The internal defect voxel data has a smaller data capacity than the three-dimensional volume data of the work itself.
Therefore, compared with the case of combining a plurality of three-dimensional volume data, the processing load when combining the comparison image data obtained from the internal defect voxel data with the work body data in the system can be reduced. . Then, by comparing the comparison image data synthesized with the workpiece body data, the occurrence status of the internal defect can be compared among the plurality of workpieces.
In this case, since the processing load for combining the comparison image data is low, a series of processes from acquisition of the respective X-ray images of the plurality of workpieces to comparison of internal defects included in the X-ray images Can be done in a short time.

As described above, according to the above-described aspect, it is possible to provide a workpiece inspection technique capable of comparing internal defects of a plurality of workpieces molded by the same mold in a short time.
The reference numerals in parentheses described in the claims and the means for solving the problems indicate the correspondence with the specific means described in the embodiments described later, and the technical scope of the present invention is limited. It is not a thing.

FIG. 1 is a block diagram of a work inspection apparatus according to a first embodiment. FIG. 2 is a perspective view showing the peripheral structure of the rotation base of the CT imaging apparatus in FIG. 1; 3 is a flowchart of a workpiece inspection method according to the first embodiment. FIG. 4 is a view schematically showing a process of processing a radiographic image of each of a plurality of works according to the flowchart of FIG. 3; The figure which shows typically three-dimensional volume data by the X-ray image of a workpiece | work. The figure which shows the voxel assembly of the 1st spout contained in the three-dimensional volume data of FIG. FIG. 6 is a view showing a second voxel collection of voxels included in the three-dimensional volume data of FIG. 5; FIG. 7 is a diagram showing graphic voxel data created from the voxel assembly of the first spout in FIG. 6. FIG. 8 is a diagram showing graphic voxel data created from the voxel assembly of the second spout in FIG. 7; The figure which shows a mode that several figure voxel data were synthesize | combined with work main body data. The figure which shows the state to which the voxel of each figure voxel data was color-coded in FIG. FIG. 5 is a configuration diagram of a workpiece inspection device of a second embodiment. 6 is a flowchart of a workpiece inspection method according to a second embodiment.

  Hereinafter, embodiments of a workpiece inspection apparatus and a workpiece inspection method will be described with reference to the drawings.

  In the drawings of this specification, unless otherwise specified, the left and right direction of the work is indicated by arrow X, the depth direction of the work is indicated by arrow Y, and the height direction of the work is indicated by arrow Z.

(Embodiment 1)
As shown in FIG. 1, a workpiece inspection apparatus (hereinafter, also simply referred to as “inspection apparatus”) 1 according to the first embodiment is an apparatus for inspecting a plurality of workpieces W formed by the same mold. In this embodiment, a cast product (die casting) made of a metal material such as aluminum, zinc, magnesium or copper is assumed as the work W. In the work W, a casting cavity B is generated as an internal defect which is a spatial defect at the time of casting.

  The inspection device 1 includes a CT imaging device 10, a processing device 20, an external storage device 30, and a display device 40.

  The CT imaging apparatus 10 is for imaging a workpiece W using X-rays. The CT imaging apparatus 10 rotates around the rotation axis L between the X-ray generator 11 and the X-ray detector 12 in order to install the X-ray generator 11, the X-ray detector 12, and the work W And a rotatable base 13 which is arranged as possible. The CT imaging apparatus 10 is also referred to as a "CT scanner".

  The X-ray generator 11 is a known device, and the description of the structure thereof is omitted. However, the X-ray generator 11 is configured to generate X-rays by driving electrons accelerated by voltage into metal. The X-ray detector 12 has a function of detecting an X-ray, and is configured of an X-ray two-dimensional camera in this embodiment. This X-ray two-dimensional camera is a flat panel detector (FPD) using CMOS or amorphous as an imaging element. The X-ray detector 12 is electrically and always connected to the processing device 20 via connection means (not shown) such as Ethernet (LAN) or camera link.

  In CT imaging of the workpiece W, X-rays generated by the X-ray detector 12 reach the X-ray detector 12 after being transmitted while being absorbed energy through the workpiece W placed on the rotation base 13. Then, the transmission image at this time is detected by the X-ray detector 12. By performing CT imaging while rotating the rotation base 13, a plurality of two-dimensional radiograph images C can be obtained for one work W. The plurality of X-ray images C are continuously transmitted from the X-ray detector 12 to the storage unit 21 of the processing device 20 through the above-described connection means in the order of imaging. This X-ray image C can also be referred to as “X-ray fluoroscopic image” or “X-ray fluoroscopic image”.

  Preferably, a positioning jig 14 is attached to the rotary base 13 as shown in FIG. The positioning jig 14 is configured to be able to position each of a plurality of works W formed by the same mold M at a common reference position Q. For this purpose, the positioning jig 14 comprises a fixed leg 15 to which the rotary base 13 is fixed, and a plate-like holding plate 16 disposed above the fixed leg 15. There are provided two projections 17 for supporting the work W from below and two projections 18 engaged with the work W at the common reference position Q.

  According to the positioning jig 14, for example, when sequentially switching the workpiece W from the first workpiece W1 to the n-th workpiece Wn molded by the same mold M and holding CT, each workpiece W is held The plate 16 can be easily positioned and set at the common reference position Q.

  Here, the holding plate 16 of the positioning jig 14 is preferably configured to be inclined at an acute angle θ with respect to the vertical direction. According to this configuration, the work W set on the holding plate 16 is biased toward the front surface of the holding plate 16 by its weight, and is stably held by the holding plate 16.

  Moreover, it is preferable that the positioning jig 14 be configured to be attachable to and detachable from the rotation base 13 and be prepared for each type of the work W. In this case, the positioning work of each work W can be easily performed when the work W is switched for each type and CT imaging is performed by replacing the positioning jig 14 as necessary.

  Note that, instead of using the positioning jig 14 described above for positioning a plurality of workpieces W, absolute coordinates common to a plurality of X-ray images C may be designated in processing on the processing apparatus 20 side.

  Referring back to FIG. 1, the processing device 20 is configured using a personal computer (hereinafter, referred to as “PC”) having a function of processing a radiographic image C of each of a plurality of workpieces W.

  In order to realize the functions, the processing device 20 has a storage unit 21, a reconstruction calculation unit 22, a data extraction unit 23, a data creation unit 24, a data combining unit 25, an overlap determination unit 26, And a hue setting unit 27. With regard to these elements constituting the processing device 20, a plurality of elements can be integrated into one element, or one element can be divided into a plurality of elements, as necessary.

  The storage unit 21 has a function of temporarily storing a plurality of radiographed images C transmitted from the X-ray detector 12 of the CT imaging apparatus 10. The storage unit 21 is configured by a memory on the motherboard of the PC. An external storage device 30 having the same function as the storage unit 21 is attached to the processing device 20. The external storage device 30 is configured by at least one of a hard disk drive (HDD) having a larger capacity than the storage unit 21 and a solid state drive (SSD).

  The reconstruction operation unit 22 performs a reconstruction operation to reconstruct a plurality of X-ray images C into three-dimensional volume data D1 using a CPU or GPU (image board) and CT reconstruction software. It is configured.

  The data extraction unit 23 has a function of extracting, from the three-dimensional volume data D1 reconstructed by the reconstruction calculation unit 22, the hollow voxel data D2 regarding the hollow B, which is internal defect voxel data.

  Here, the three-dimensional volume data D1 is created by performing a reconstruction operation on a plurality of roentgen photographed images C. That is, this three-dimensional volume data D1 is three-dimensional CT volume data created directly by the reconstruction operation. This three-dimensional volume data D1 is obtained for each work W. This three-dimensional volume data D1 is represented by a voxel collection which is a collection of voxels (minimum cubes (normal grids)) finely dividing a three-dimensional space into small portions, and the density distribution of the image in the three-dimensional space It is expressed by a three-dimensional data array. Specifically, data in which the amount of absorption of X-rays is stored for each voxel is obtained, and one pixel value (value indicating the amount of absorption of X-rays) is given to each voxel.

  The data creation unit 24 obtains the digitized data Dc representing the voxel assembly A that constitutes the nest B on the basis of the nest voxel data D2 extracted by the data extraction unit 23, and generates the voxel collection from this digitized data Dc. It has a function of creating figure voxel data Ds of a sphere S as a simple figure having the same volume as the body A as comparison image data D3.

  In the present embodiment, the numerical data Dc is configured by numerical values for specifying the position and size of the sphere S. Specific numerical data Dc includes the respective numerical values of the three-dimensional coordinates of the gravity center G of the voxel aggregate A constituting the cavity B in the three-dimensional volume data D1 and the volume of the voxel aggregate A. In this case, it is possible to use the values of xyz coordinates of the center of gravity G of the voxel assembly A and the volume value of the voxel assembly A. Note that the volume value of the voxel assembly A can also be represented by the diameter value of the sphere S obtained from this volume value.

  According to the data creation unit 24, the casting cavity B is re-voxelized as the figure voxel data Ds of the sphere S by using the digitization data Dc. The process at this time can also be referred to as "revoxelization" or "inverse voxelization".

  It should be noted that, in place of the simple figure sphere S, other simple three-dimensional figures such as prisms such as a cube, cones and ellipsoids are used to create figure voxel data Ds of these figures. You may do so. In short, it is sufficient if it is a simple figure whose shape is simpler than that of the cavity B (voxel aggregate A) whose shape is complicated.

  The data synthesis unit 25 casts the figure voxel data Ds (comparison image data D3) obtained from the nest voxel data D2 extracted by the data extraction unit 23 for each of the plurality of works W with respect to the work main body data Dw. It has a function of combining at the combining position P which is the original position of the nest B, that is, the position of the center of gravity G of the voxel aggregate A.

Here, the work main body data Dw is data obtained by removing three-dimensional volume data corresponding to the cavity B from any one three-dimensional volume data D1 of a plurality of works W.
Alternatively, CAD data indicating the overall shape of the workpiece W can be used as the workpiece body data Dw.

  The overlap determination unit 26 has a function of determining an overlap degree in which a plurality of pieces of graphic voxel data Ds synthesized with respect to the work main body data Dw by the data synthesis unit 25 overlap each other. The hue setting unit 27 has a function of causing the display device 40 to display the graphic voxel data Ds with a hue set in advance according to the degree of overlap determined by the overlap determination unit 26.

  The display device 40 is configured as a PC display (monitor) capable of displaying information generated by processing of data and images by the processing device 20, and is electrically connected to the processing device 20 via connection means (not shown). ing.

  Next, an inspection method using the inspection apparatus 1 configured as described above will be described with reference to FIGS.

  As shown in the flowchart of FIG. 3, the workpiece inspection method for inspecting a plurality of workpieces W formed by the same mold M is achieved by sequentially executing the processing from step S101 to step S114. Note that another step may be added to this flowchart as needed, or one step of this flowchart may be divided into a plurality of steps.

  Step S101 is a step of setting one work W on the rotation base 13 of the CT imaging apparatus 10. In this step S101, in order to easily position the workpiece W, it is preferable to use the above-mentioned positioning jig 14 (see FIG. 2). According to step S101, preparation before operating the CT imaging apparatus 10 is performed.

  Step S102 is a step of operating the CT imaging apparatus 10 to perform CT imaging of the workpiece W next to step S101. In step S102, the rotation base 13 is rotationally driven about the rotation axis L while the X-ray generator 11 and the X-ray detector 12 are operated. According to step S102, actual CT imaging of the work W is performed, and a plurality of X-ray images C of the work W are obtained.

  Step S103 is a step of storing a plurality of X-ray images C of the work W obtained in step S102. According to step S103, a plurality of X-ray images C are transmitted from the CT imaging apparatus 10 to the storage unit 21 of the processing device 20 and temporarily stored. In this step S103, a plurality of X-ray images C may be stored in the external storage device 30 instead of or in addition to the storage unit 21.

  Step S104 is a step of reconstructing a plurality of radiographic images C into three-dimensional volume data D1. In step S104, a plurality of X-ray radiographed images C temporarily stored in the storage unit 21 are read out, and the plurality of X-ray radiographed images C are three-dimensional volume by CT reconstruction software of the reconstruction calculation unit 22. Restructure into data D1. This step S104 is executed for all the works W.

  For example, as shown in FIG. 4, a plurality of X-ray images C1 for the first work W1 are reconstructed into three-dimensional volume data D1, and a plurality of X-ray images Cn for the n-th work Wn Reconstructed into three-dimensional volume data D1.

  Step S105 is a step of displaying the three-dimensional volume data D1 reconstructed in step S104 on the display device 40. According to this step S105, the user can observe each work W as a solid on the display device 40, or can observe the cross section of each work W.

  Here, as shown in FIG. 5, the three-dimensional volume data D1 is composed of an assembly A of voxels a (also referred to as a “void”) that constitutes an internal space. By dividing the luminance value of each voxel a based on the threshold value, it is possible to sort a plurality of cavities B in the three-dimensional volume data D1.

  Step S106 is a data extraction step of extracting the void voxel data D2 related to the cavity B from the three-dimensional volume data D1 of the X-ray image C of each work W. This step S106 is executed by the data extraction unit 23 for the cavities B of all the workpieces W.

For example, as shown in FIG. 5, a first cavity B1 which is one of the cavities B is formed at the position P1 of the work W, and is a voxel assembly which is one of the voxel assembly A. It is configured by A1. Similarly, a second cavity B2 which is one of the cavities B is formed by the voxel aggregate A2 which is formed at the position P2 of the work W and which is one of the voxel assembly A.
Therefore, according to step S106, these voxel aggregates A1 and A2 are extracted as the three-dimensional voxel data D2 from the three-dimensional volume data D1.

  Step S <b> 107 is a step of digitizing the nest cavity voxel data D <b> 2 extracted in step S <b> 106. Specifically, numerical data Dc representing the voxel aggregate A is obtained. This step S107 is performed by the data creation unit 24 for all the holes B of the workpiece W.

  For example, as shown in FIG. 4 and FIG. 6, for the cavity B1, three-dimensional coordinates (x1, y1, z1) of the center of gravity G of the voxel assembly A1 constituting the cavity B1 and the voxel assembly A1 Each numerical value with the volume V1 is used as numerical data Dc representing a voxel assembly A1. In this case, the volume of the voxel aggregate A1 is derived from the number of voxels a constituting the voxel aggregate A1.

  Similarly, as shown in FIG. 4 and FIG. 7, for the cavity B2, three-dimensional coordinates (x2, y2, z2) of the center of gravity G of the voxel assembly A2 constituting the cavity B2 and the voxel assembly A2 The respective numerical values of the volume V2 and the volume V2 are used as numerical data Dc representing the voxel aggregate A2. In this case, the volume of the voxel aggregate A2 is derived from the number of voxels a constituting the voxel aggregate A2.

  The data capacity of the above-mentioned numerical data Dc is significantly less than the data capacity of the three-dimensional volume data D1 of the work W. For example, while the data capacity of the three-dimensional volume data D1 is several gigabytes, the data capacity of the numerical data Dc is several kilobytes.

  Step S108 is a step of temporarily storing the numerical data Dc obtained in step S107 in the storage unit 21.

  Step S109 determines whether the inspection number n of the workpiece W has reached a predetermined number k. In this S109, when the inspection number n reaches the predetermined number k (in the case of "Yes" in step S109), the process proceeds to step S110, otherwise (in the case of "No" in step S109) the process proceeds to step S101. Return. In addition, in order to raise the comparison accuracy of the hollow B of the workpiece | work W, it is preferable to make predetermined number k at this time into about 30 at least.

  Step S110 is a step of reading the digitized data Dc from the storage unit 21.

  Step S111 is a data creation step of creating, as comparison image data D3, graphic voxel data Ds of a sphere S having the same volume as that of the voxel assembly A based on the numerical data Dc read out in step S110. The graphic voxel data Ds is three-dimensional volume data of a sphere S, that is, a voxel aggregate. Thereby, the nest B is re-voxelized to create figure voxel data Ds of the sphere S. This step S111 can also be called "re-voxelization step" or "inverse voxelization step". This step S111 is executed by the data creation unit 24 for all the cavities B of the workpiece W.

  For example, as shown in FIG. 4 and FIG. 8, with respect to the cavity B1, graphic voxel data Ds1 of a sphere S1 having the same volume as the voxel assembly A1 is created as comparison image data D3. Similarly, as shown in FIG. 4 and FIG. 9, as for the hollow B2, graphic voxel data Ds2 of a sphere S2 having the same volume as that of the voxel assembly A2 is created as comparison image data D3.

  In step S112, the figure voxel data Ds created in step S111 is synthesized at the synthesis position P, which is the original position of the spit B, with respect to the workpiece body data Dw indicating the entire shape common to the plurality of workpieces W Data synthesis step). This step S112 is executed by the data combining unit 25 for all the holes B of the work W.

  For example, as shown in FIG. 4 and FIG. 10, with regard to the first cavity B1, the figure voxel data Ds1 of the sphere S1 corresponding to the voxel assembly A1 is the original of this cavity B1 among the workpiece body data Dw. It synthesize | combines to the synthetic | combination position P1 (refer FIG. 5) which is a position of. The combined position P1 is the position of the center of gravity G of the voxel aggregate A1.

  Similarly, for the second spit B2, the figure voxel data Ds2 of the sphere S2 corresponding to the voxel aggregate A2 is combined position P2 which is the original position of the spit B2 in the work main body data Dw (FIG. 5) (See reference). The combined position P2 is the position of the center of gravity G of the voxel assembly A2.

  It is preferable to display the state in which the graphic voxel data Ds is synthesized with the workpiece body data Dw on a PC display as the display device 40. In this case, since both the work body data Dw and the graphic voxel data Ds are three-dimensional volume data, the user can observe the cross section of the work W.

  Step S113 is an overlap determining step of determining an overlapping degree in which a plurality of pieces of graphic voxel data Ds synthesized with respect to the workpiece body data Dw in step S112 overlap each other. This step S113 is executed by the overlap determination unit 26 for all the cavities B of the workpiece W.

  According to this step S113, it is possible to determine the degree of overlap of the figure voxel data Ds for a plurality of works W. In the present embodiment, the overlap number of voxels a of each figure voxel data Ds overlapping the voxel a of the work main body data Dw (hereinafter referred to as “voxel overlap number”) is determined as the overlap degree. Information on the number of overlaps is stored in the storage unit 21.

  Step S114 is a hue setting step for causing the display device 40 to display the graphic voxel data Ds with the hue set in advance in accordance with the degree of overlap determined in step S113. This step S114 is executed by the hue setting unit 27 for all the holes B of the work W.

  In this step S114, for example, the work main body data Dw is displayed in a base color such as black or white, while the voxel a in which the number of voxel overlap is relatively large in each figure voxel data Ds is relatively small in the number of voxel overlap. Display in a color that is more noticeable than voxels. According to this step S114, the user can visually recognize the occurrence state of the cavities B in each work W and the occurrence probability of the cavities B in all the works W.

  Here, an example of the display mode of the voxel overlap number of the graphic voxel data Ds will be described with reference to FIG.

  In the example shown in FIG. 11, one graphic voxel data Ds1 overlapping only with the workpiece body data Dw, two graphic voxel data Ds2 overlapping with the workpiece body data Dw and overlapping each other, and the workpiece body data Dw overlapping with each other 3 One graphic voxel data Ds3 is shown. In this case, the voxel overlap number of voxels a belonging to the first region R1 is “1”, the voxel overlap number of voxels a belonging to the second region R2 is “2”, and the voxel overlap number of the voxels a belonging to the third region R3 The voxel overlap number is “3”.

  Therefore, in one display mode, the voxel a in the second region R2 is displayed with a hue that is more noticeable than the voxel a in the first region R1, and the voxel a in the third region R3 is more than the voxel a in the second region R2. It can be displayed in a noticeable hue. At this time, the hue of voxel a may be changed continuously between the regions, or the hue of voxel a may be changed stepwise between the regions.

  For example, when the number of works W is 30, the voxel overlap number can be a value from 1 to 30. Therefore, when the hue displayed on the PC display as the display device 40 is a combination of R (red), G (green) and B (blue) and each of the RGB has 8 bits (256 gradations) of information, RGB The hues of voxel a can be changed continuously or stepwise between voxel numbers 1 to 30 by setting the respective gradations appropriately and combining them.

  Note that instead of the display form in which voxels a are color-coded according to the voxel overlap number as described above, a display form in which voxels a are color-coded according to parameters other than voxel overlap number such as overlap density and overlap area of voxel a Alternatively, a display form may be adopted in which the graphic voxel data Ds are color-coded according to the number of overlapping graphic voxel data Ds.

  In addition, when the figure voxel data Ds displayed on the display device 40 is selected by the selection means such as a mouse, the number of voxel overlap of the voxel a at the selected position is displayed numerically on the display device 40. preferable. Thus, the user can quantitatively understand the number of voxels of the voxel a quantitatively as well as numerically, in addition to qualitatively grasping the difference in hue.

  Next, the operation and effect of the above-described first embodiment will be described.

According to the first embodiment, from the three-dimensional volume data D1 of the X-ray image C of each work W, the void voxel data D2 related to the void B is extracted. Further, graphic voxel data Ds is created using numerical data Dc which is obtained by digitizing the spout voxel data D2. The numerical data Dc at this time has a smaller data capacity than the three-dimensional volume data D1 of the work W itself. For example, with respect to three-dimensional volume data D1 having a data capacity of several gigabytes, the data capacity of the numerical data Dc can be suppressed to several kilobytes.
Therefore, compared with the case where a plurality of three-dimensional volume data D1 are combined, the processing load when combining the figure voxel data Ds obtained from the digitized data Dc with the work main body data Dw in the system is reduced. Can. Then, by comparing the figure voxel data Ds synthesized with the work main body data Dw, it is possible to compare the occurrence of the cavities B among the plurality of works W.
In this case, since the processing load for synthesizing the graphic voxel data Ds is low, it is necessary to obtain the X-ray images C of each of the plurality of works W and compare the foci B contained in the X-ray images C A series of processes can be performed in a short time.

  As described above, according to the first embodiment, it is possible to compare the respective cavities B of the plurality of works W formed by the same mold M in a short time.

  Further, according to the first embodiment, the occurrence probability of the void B can be grasped by determining the degree of overlap (the number of voxel overlap) in which a plurality of graphic voxel data Ds synthesized with the work main body data Dw overlap each other. . In particular, by displaying the graphic voxel data Ds on the display device 40 with the hue set in advance in accordance with the number of overlapping voxels, the user visually recognizes the location where the blobs B are concentrated among the plurality of works W it can.

  As a result, the user can determine that the cavities B occurring in a concentrated area generally common to the plurality of workpieces W are caused by the structure of the mold M, and the cavities B are generated. Can be related to the structure of the mold M. On the other hand, this cavity B can be distinguished from internal defects randomly generated according to a minute change of the actual casting conditions and the like. As a result, it becomes possible to reflect the generation | occurrence | production condition of the cavity B in the structure of the metal mold | die M, and can hold down the cost which the development and improvement of the expensive metal mold | die M require.

  Further, according to the first embodiment, by using any one three-dimensional volume data D1 of a plurality of works W as the work main body data Dw, the figure voxel data Ds is synthesized with the work main body data Dw. Cross section observation of the work W becomes possible.

  Hereinafter, other embodiments relating to the above-described first embodiment will be described with reference to the drawings. In the other embodiments, the same elements as the elements of the first embodiment are denoted by the same reference numerals, and the description of the same elements is omitted.

Second Embodiment
As shown in FIG. 12, the workpiece inspection apparatus 101 of the second embodiment differs from the workpiece inspection apparatus 1 of the first embodiment only in the configuration of the processing device 120.

The processing device 120 includes a storage unit 21, a reconstruction calculation unit 22, a data extraction unit 23, a data combining unit 25, an overlap determination unit 26, and a hue setting unit 27. That is, the processing device 120 does not include an element corresponding to the data creation unit 24 of the processing device 20 of the first embodiment.
The other configuration is the same as that of the first embodiment.

  As shown in FIG. 13, the inspection method using the workpiece inspection apparatus 101 of the second embodiment is achieved by sequentially executing the processing from step S201 to step S212. Note that another step may be added to this flowchart as needed, or one step of this flowchart may be divided into a plurality of steps.

  Here, the process from step S201 to step S206 and the process of step S208 are the same as the process from step S101 to step S106 of the first embodiment and the process of step S108 (see FIG. 3). Therefore, in the following, only processing of other steps will be described.

  Step S207 is a step of temporarily storing the nest cavity voxel data D2 extracted in step S206 in the storage unit 21. In this case, the nest cavity voxel data D2 is stored as three-dimensional coordinates of all the voxels a constituting the voxel assembly A.

  Step S209 is a step of reading the nest cavity voxel data D2 from the storage unit 21.

  Step S210 is a data combining step of combining the hollow voxel data D2 read out in step S209 at the combining position P, which is the original position of the hollow B, with respect to the workpiece body data Dw. In this case, the nest cavity voxel data D2 itself becomes the comparison image data D3. This step S210 is executed by the data combining unit 25 for the cavities B of all the workpieces W.

  Step S211 is an overlap determination step of determining an overlap degree (the number of overlap of voxels a) in which a plurality of the nest cavity voxel data D2 synthesized with the work body data Dw in step S210 overlap each other. This step S211 is performed by the overlap determination unit 26 for all the cavities B of the workpiece W.

  Step S212 is a hue setting step which causes the display device 40 to display the sprinkling voxel data D2 with a hue set in advance according to the degree of overlap determined in step S210. This step S212 is performed by the hue setting unit 27 for all the holes B of the workpiece W.

  According to the second embodiment, since the hollow voxel data D2 (the voxel aggregate A of the hollow B) is synthesized as it is with respect to the work main body data Dw, the hollow B after synthesis in the display device 40 Can be displayed accurately. As a result, the user can observe the shape of the actual cavity B, not the simple spherical shape as in the first embodiment.

In addition, although the data volume of the hollow voxel data D2 to be synthesized with respect to the work main body data Dw is larger than the numerical data Dc in the first embodiment, the volume of the hollow B is considerably small. The data volume of the dimensional volume data D1 is still far below this data volume.
For this reason, the processing load for synthesizing the void voxel data D2 is low, and until the roving radiographic images C of the plurality of works W are acquired and the voids B included in the radiographic imaging images C are compared A series of processes can be performed in a short time.

  Further, it is possible to omit both the data creation unit 24 of the first embodiment, the digitizing step as in step S107, and the data creation step as in step S111.

  Other effects and effects similar to those of the first embodiment are achieved.

  The invention is not limited to the exemplary embodiments described above, but various applications and modifications are conceivable without departing from the object of the invention. For example, the following embodiments to which the above embodiment is applied can be implemented.

  Although the above embodiment exemplifies the case where the processing devices 20 and 120 include at least the three elements of the reconstruction calculation unit 22, the overlap determination unit 26, and the hue setting unit 27, instead of this, A configuration in which all are omitted (hereinafter referred to as “first configuration”), a configuration in which only the reconstruction calculation unit 22 is omitted (hereinafter referred to as “second configuration”), or only the hue setting unit 27 A configuration (hereinafter referred to as a “third configuration”) in which is omitted and a configuration (hereinafter referred to as a “fourth configuration”) in which both the overlap determination unit 26 and the hue setting unit 27 are omitted are employed. It can also be done.

  For example, in the case of the first configuration, the three-dimensional volume data D1 of each work W is stored in advance in the storage unit 21, and the three-dimensional volume data D1 is read from the storage unit 21 by the data extraction unit 23 Voting voxel data D2 may be extracted from the data D1. Further, by displaying the state in which the comparison image data D3 is combined with the workpiece body data Dw on the display device 40, the user can compare the respective cavities B of the plurality of workpieces W.

  Further, in the case of the second configuration, the three-dimensional volume data D1 of each work W is stored in advance in the storage unit 21, and the three-dimensional volume data D1 is read from the storage unit 21 by the data extraction unit 23 Voting voxel data D2 may be extracted from the data D1.

  Further, in the case of the third configuration, the numerical value information indicating the degree of overlap is stored in the storage unit 21 without color-coding the degree of overlap of the comparison image data D3 determined by the overlap determination unit 26. it can.

  Further, in the case of the fourth configuration, the user can display the state in which the comparison image data D3 is combined with the work main body data Dw on the display device 40, so that the user can store each of the plurality of You can compare each other.

  Although the above embodiment exemplifies the case where the processing apparatus 20 performs the image processing while photographing the work W by the CT imaging apparatus 10, instead of this, the image processing in the processing apparatus 20 and the work W as a CT imaging apparatus It can also be executed at a timing different from the operation of photographing at 10.

  Although the above embodiment exemplifies the case where the X-ray detector 12 of the CT imaging apparatus 10 is always connected to the processing apparatuses 20 and 120, the CT imaging apparatus 10 and the processing apparatuses 20 and 120 may be replaced instead. It can also be used separately from one another.

  In the above embodiment, although the case where the internal cavity of the work W is inspected as the cavity B is illustrated, it is also possible to inspect a spatial cavity B such as a crack other than the cavity.

  Although the above embodiment exemplifies a case where an X-ray is used to obtain an X-ray image of the work W, X-ray imaging of the work W using radiation other than X-rays, such as gamma rays, instead of this. An image may be obtained.

  In the above embodiment, the case of inspecting a plurality of cast products (die castings) cast by the same mold M from a metal material is exemplified, but this inspection technique is molded by the same mold from a resin material The present invention can also be applied to techniques for inspecting a plurality of resin molded articles.

1,101 Workpiece inspection device (inspection device)
DESCRIPTION OF SYMBOLS 10 CT imaging apparatus 11 X-ray generator 12 X-ray detector 13 Rotation base 14 Positioning jig 20, 120 Processing apparatus 22 Reconfiguration operation part 23 Data extraction part 24 Data creation part 25 Data combining part 26 Overlap determination part 27 Hue setting Part 40 Display a Voxel A, A1, A2 Voxel assembly B, B1, B2 Cavity (internal defect)
C roentgen radiographed image D1 3D volume data D2 homing voxel data (internal defect voxel data)
D3 Comparison image data Dc Numerical data Ds, Ds1, Ds2, Ds3 Graphic voxel data (comparison image data)
Dw Workpiece body data G Center of gravity M Mold P, P1, P2 Composite position Q Common reference position S, S1, S2 Sphere (simple figure)
W, W1, Wn Workpieces S106, S206 Data extraction step S111 Data creation step S112, S210 Data synthesis step S113, S211 Overlap determination step S114, S212 Hue setting step

Claims (11)

A work inspection apparatus (1, 101) for inspecting a plurality of works (W) formed by the same mold (M),
A processing device (20, 120) for processing an X-ray image (C) of each of the plurality of works;
The above processing unit
A data extraction unit (23) for extracting internal defect voxel data (D2) relating to internal defects (B, B1, B2) from the three-dimensional volume data (D1) from the radiographic image;
The comparison image data (D3) obtained from the internal defect voxel data extracted by the data extraction unit for each of the plurality of workpieces is used as the workpiece body data (Dw) indicating the entire shape common to the plurality of workpieces. A data synthesizing unit (25) for synthesizing at a synthesis position (P, P1, P2) which is the original position of the internal defect;
Work inspection apparatus (1, 101). The processing apparatus generates voxel data from numerical data (Dc) representing a voxel assembly (A, A1, A2) constituting the internal defect based on the internal defect voxel data extracted by the data extraction unit. It has a data creation unit (24) that creates graphic voxel data (Ds, Ds1, Ds2, Ds3) of simple figures (S, S1, S2) having the same volume as the body as the comparison image data,
The data combining unit combines the figure voxel data created by the data creating unit with the work main body data at the position of the center of gravity (G) of the voxel aggregate which is the combining position. Work inspection device as described.   The processing apparatus according to claim 1 or 2, further comprising: an overlap determining unit (26) configured to determine an overlapping degree in which a plurality of the comparison image data combined with the work body data by the data combining unit overlap each other. Work inspection device as described.   The processing device comprises a hue setting unit (27) for causing the display device (40) to display the comparison image data with a hue preset according to the degree of overlap determined by the overlap determination unit. The workpiece inspection device according to 3.   5. The work main body data according to claim 1, wherein three-dimensional volume data corresponding to the internal defect is excluded from the three-dimensional volume data of any one of the plurality of works. The work inspection apparatus described in.   The work inspection apparatus according to any one of claims 1 to 5, wherein the processing device has a reconstruction operation unit (22) that reconstructs the X-ray image into the three-dimensional volume data. A CT imaging device (10) for obtaining the X-ray image;
The CT imaging apparatus includes an X-ray generator (11), an X-ray detector (12), a rotation base (13) rotatably disposed between the X-ray generator and the X-ray detector. The positioning base according to any one of claims 1 to 6, wherein the rotation base is attached with a positioning jig (14) capable of positioning each of the plurality of workpieces at a common reference position (Q). The work inspection apparatus described in. A workpiece inspection method for inspecting a plurality of workpieces (W) formed by the same mold (M), comprising:
A data extraction step (S106) for extracting internal defect voxel data (D2) related to internal defects (B, B1, B2) from three-dimensional volume data (D1) based on X-ray images (C) of each of the plurality of workpieces S206),
The comparison image data (D3) obtained from the internal defect voxel data extracted in the data extraction step for each of the plurality of workpieces is compared to the workpiece body data (Dw) indicating the overall shape common to the plurality of workpieces Data synthesizing step (S112, S210) for synthesizing at the synthesis position (P, P1, P2) which is the original position of the internal defect;
Have a work inspection method. Based on the internal defect voxel data extracted in the data extraction step, the same volume as the voxel aggregate is calculated from the digitized data (Dc) representing the voxel aggregate (A, A1, A2) constituting the internal defect. A data creation step (S111) of creating figure voxel data (Ds, Ds1, Ds2, Ds3) of the simple figure (S, S1, S2) that it has as the comparison image data;
9. The data synthesizing step according to claim 8, wherein said graphic voxel data generated by said data generating step is synthesized with respect to said work main body data at the position of the center of gravity (G) of said voxel aggregate which is said synthesizing position. Work inspection method.   The workpiece inspection according to claim 8 or 9, further comprising an overlap determination step (S113, S211) for determining an overlap degree between the plurality of comparison image data combined with the workpiece body data in the data combining step. Method.   11. The method according to claim 10, further comprising a hue setting step (S114, S212) for causing the display device (40) to display the comparison image data with a hue set in advance according to the overlap degree determined in the overlap determination step. Work inspection method.

Images