JP6521165B2 - X-ray examination processing apparatus and x-ray examination method - Google Patents

Description

  The present invention relates to an industrial X-ray inspection apparatus.

  There is known an industrial X-ray inspection apparatus that detects defects or defects in industrial products using image information obtained by X-ray imaging. Such an X-ray inspection apparatus is applied to, for example, inspection of the solder joint state of components on a surface mounted substrate, etc., taking advantage of nondestructive inspection of a portion which is difficult to visually recognize from the appearance and the state inside the object. ing.

  The imaging methods used for X-ray inspection are roughly classified into 2D imaging and 3D imaging. The 2D imaging is a method of projecting an X-ray on an object from one direction and acquiring the transmission image as 2D image information. On the other hand, 3D imaging is a method of acquiring 3D image information such as 3D volume data of an object or a tomogram at an arbitrary cross section based on a plurality of transmission images obtained by changing the X-ray projection direction, Techniques such as CT (computed tomography) and tomosynthesis (Tomosynthesis) are known. In this specification, an inspection using 2D image information obtained by 2D imaging is referred to as 2D inspection, and an inspection using 3D image information obtained by 3D imaging is referred to as 3D inspection.

  For example, in the case of a BGA (ball grid array) component, as shown in FIG. 16A, there is a space (open) between the solder ball 162 on the component 160 side and the cream solder 163 on the substrate 161 side. Defects such as a state in which the solder 163 is not fused (unfused / Head In Pillow) may occur. However, in a transmission image obtained by 2D imaging, as shown in FIG. 16B, it is difficult to distinguish between the solder ball 162 and the cream solder 163, so it is difficult to distinguish these defective states from non-defective products. Therefore, in 2D inspection, problems such as the outflow of defective products due to missing, and the drop in the direct rate due to oversight occur. Therefore, for this type of defect that is difficult to detect in 2D inspection, it is desirable to use 3D inspection. For example, Patent Document 1 discloses a method of inspecting a solder ball of a BGA part with high accuracy by using a tomographic image obtained by tomosynthesis.

Japanese Patent Laid-Open No. 7-221151

  As described above, 3D inspection is an effective method for defects that are difficult to detect with 2D transmission images alone. However, 3D inspection has the following problems.

The first issue is the time required for imaging and examination. In the case of 2D inspection, a transmission image of one field of view (FOV; Field of View) can be obtained by one imaging, while in the case of 3D inspection, several to several tens of imagings per field of view Is required. Therefore, the imaging time is longer than that of the 2D inspection, resulting in a decrease in inspection throughput. This problem becomes more serious with the recent miniaturization and densification of component packages. When the resolution of imaging is increased in accordance with the narrowing of the diameter and the pitch of the solder ball, one component has to be divided into a plurality of fields of view and imaged, and the imaging time increases in proportion to the number of fields of view. FIGS. 17A and 17B show that when imaging a BGA component with a ball diameter of 300 microns with a detector of 2000 × 2000 pix with a resolution of 20 microns, the entire component can be imaged in one field of view (field size: 40 mm × 40 mm) On the other hand, when imaging a BGA part with a ball diameter of 80 microns using the same detector with a resolution of 3 microns, it must be divided into 30 fields of view (field size: 6 mm x 6 mm) It shows.

  The second issue is the exposure of parts. Even in the case of electronic components, if the exposure dose exceeds the allowable limit, the possibility of performance degradation and failure increases. Therefore, it is desirable that the number of times of imaging of parts (number of X-ray irradiations) be as small as possible.

  The present invention has been made in view of the above-described circumstances, and an object thereof is to provide a technique for enabling inspection with a short exposure time and low exposure even for defects that are difficult to detect in a 2D transmission image. .

  In order to achieve the above object, the present invention is characterized in that inspection is performed by combining 2D imaging information and 3D imaging information.

  Specifically, the present invention is a processing apparatus for X-ray inspection which inspects an object including a plurality of inspection objects by X-rays, and a part of the inspection area set in the object. A 3D image acquisition unit that acquires a 3D image based on a plurality of X-ray transmission images in different projection directions for a first area which is an area, and an area at least a part of which is different from the first area of the inspection area And a 2D image acquisition unit for acquiring a 2D image that is an X-ray transmission image, and a 3D image of the first region, the first inspection object being an inspection object in the first region. The 3D information is extracted, and from the 2D image of the second region, the 2D information of the second inspection object which is an inspection object different from the first inspection object among the inspection objects in the second region An extracting unit for extracting, and the first inspection object extracted by the extracting unit A 3D information estimation unit for estimating 3D information of the second inspection object based on 3D information, 2D information of the second inspection object extracted by the extraction unit, and the 3D information estimation unit estimated by the 3D information estimation unit An inspection unit for inspecting the second inspection object using 3D information of the second inspection object, and an X-ray inspection processing apparatus.

  According to this configuration, since the area for performing 3D imaging is limited to a part (only the first area) of the inspection area, the number of imaging times and X-rays are compared to imaging the entire inspection area in 3D. The number of times of irradiation can be reduced, and the time required for imaging and inspection can be shortened, and the exposure dose can be reduced. Also, by estimating 3D information of the second inspection object in the second region based on 3D information extracted from the 3D image of the first region, and using the 3D information for inspection of the second inspection object, The pseudo 3D inspection can also be performed on the second inspection object which is performing only 2D imaging. This enables inspection of defects that are difficult to detect by conventional 2D inspection.

  "2D information" is geometrical information in 2D space (2D plane), and is, for example, position, size, width, distance, area, shape, etc. in 2D plane. "3D information" is geometrical information in 3D space, and is, for example, position, size, height, width, distance, area, volume, shape, etc. in 3D space.

  There is no limitation on the method of estimating 3D information. For example, when a plurality of first test objects are included in the first region, the 3D information estimation unit interpolates or extrapolates 3D information of the plurality of first test objects. Thus, 3D information of the second inspection object can be calculated.

  Here, the XYZ coordinate system is placed so that the X-ray irradiation direction of the 2D imaging is perpendicular to the XY plane, and both ends of the inspection object in the Z direction are referred to as a first end and a second end.

For example, the extraction unit extracts an XY position and a Z-direction height of each of the plurality of first inspection objects from the 3D image of the first region, and the second inspection from the 2D image of the second region The XY position of the object is extracted, and the 3D information estimation unit interpolates or extrapolates the heights of the plurality of first inspection objects in the Z direction, whereby the Z direction at the XY positions of the second inspection object is extracted. It is good to calculate the height. As a result, a pseudo 3D inspection using the estimated Z-direction height can be performed on the second object to be inspected.

  The extraction unit extracts the XY position and volume information of each of the plurality of first inspection objects from the 3D image of the first region, and the XY position of the second inspection object from the 2D image of the second region The 3D information estimation unit may calculate volume information of the second inspection object at the XY position by interpolating or extrapolating the volume information of the plurality of first inspection objects. Thereby, the pseudo 3D inspection using the estimated volume information can be carried out also for the second object to be inspected.

  The extraction unit extracts the XY position of each of the plurality of first inspection objects from the 3D image of the first region, and the 3D information estimation unit interpolates the XY positions of the plurality of first inspection objects. Alternatively, extrapolation may be performed to calculate the XY position where the second inspection object should exist.

  The XY position (hereinafter referred to as a theoretical position) where the second inspection object should be present may be acquired from, for example, CAD information of the object. However, if there is a manufacturing error in the object or if the position of the entire object is shifted, the position of the second inspection object also does not conform to the CAD information. In that case, if the inspection of the second object is performed with the theoretical position obtained from the CAD information as a reference (correct), there is a possibility that an invalid result may be obtained, which is not preferable.

  In that respect, in the present invention, the theoretical position of the second inspected object is estimated (from the relative positional relationship) based on the XY position of the first inspected object obtained by 3D imaging. Therefore, it is possible to obtain the theoretical position of the second inspection object in which the actual state of the object (for example, manufacturing error, shift of the entire object, etc.) is taken into consideration. If the theoretical position thus determined is used for the inspection of the second inspection object, more appropriate inspection results can be obtained, and the inspection accuracy and reliability can be improved.

  The extraction unit extracts the XYZ positions of the first end and the XYZ positions of the second end of each of the plurality of first inspection objects from the 3D image of the first area, and the 3D information estimation unit is configured to While calculating the XYZ positions where the first end of the second inspection object should be present by interpolating or extrapolating the XYZ positions of the first end of the first inspection object, the plurality of first inspection objects The XYZ position at which the second end of the second inspection object should be present may be calculated by interpolating or extrapolating the XYZ position of the second end of.

  According to this configuration, it is possible to obtain theoretical positions of both ends of the second inspection object in which the actual state of the object (for example, manufacturing error, shift of the entire object, etc.) is considered. If the theoretical position thus determined is used for the inspection of the second inspection object, more appropriate inspection results can be obtained, and the inspection accuracy and reliability can be improved.

  The 3D information estimation unit determines the first end and the second end of the second inspection object from the estimated XYZ position where the first end should exist and the XYZ position where the second end should exist. The deviation amount in the XY plane may be calculated. In addition, the 3D information estimation unit averages the estimated XYZ position where the first end should exist and the XYZ position where the second end should exist, so that the second inspection object should exist. The position may be calculated.

According to this configuration, it is possible to obtain the theoretical position of the second inspection object in which the actual state of the object (for example, manufacturing error, shift of the entire object, etc.) is considered. If the theoretical position thus determined is used for the inspection of the second inspection object, more appropriate inspection results can be obtained, and the inspection accuracy and reliability can be improved.

  For example, the extraction unit extracts the actual XY position of the second inspection object from the 2D image of the second region, and the inspection unit is configured to extract the actual state of the second inspection object extracted by the extraction unit. The quality of the second test object may be determined by comparing the XY position of the second test object with the XY position where the second test object should be estimated by the 3D information estimation unit.

  According to this configuration, the actual object is obtained by comparing the theoretical position of the second inspection object estimated from the 3D imaging result with the actual position of the second inspection object calculated from the 2D imaging result. It is possible to obtain appropriate inspection results in which the state of (for example, manufacturing error, shift of the entire object, etc.) is taken into consideration, and inspection accuracy and reliability can be improved.

  The extraction unit extracts a distance between the second inspection object and an adjacent inspection object adjacent to the second inspection object from the 2D image of the second area, and the inspection unit is configured to extract the extraction unit. The second object under inspection based on the distance extracted by the second object and the deviation between the first end and the second end of the second object within the XY plane estimated by the 3D information estimation unit. It is good to judge the quality of the

  Since the second object to be inspected in the 2D image is a projected image on the XY plane, the projected image becomes larger as the second object to be inspected is inclined to the Z axis, and the adjacent object to be inspected on the 2D image is The apparent distance (distance) to the object decreases. The index to be evaluated in the inspection is not the apparent distance between the projected images but the three-dimensional actual distance, but the three-dimensional distance can not be determined from only the 2D image. Therefore, in the present invention, the deviation amount in the XY plane of the first end and the second end of the second inspection object estimated from the result of the 3D imaging is considered. The amount of deviation is an index that correlates with the inclination of the second test object relative to the Z axis. Therefore, according to the method of the present invention, it is possible to obtain a reasonable inspection result in which the inclination of the second inspection object is considered, and to improve the inspection accuracy and reliability.

  The extraction unit extracts the actual roundness of the second inspection object from the 2D image of the second area, and the inspection unit determines the first end and the second end of the second inspection object. The roundness of the second object to be inspected is estimated based on the deviation amount in the XY plane of the second embodiment, and the estimated roundness and the actual circularity of the second object to be inspected extracted by the extraction unit It is preferable to determine the quality of the second inspection object by comparing.

  The second object to be inspected in the 2D image is a projected image on the XY plane, so the projected image is deformed as the second object to be inspected is inclined to the Z axis, and the roundness changes. The above inspection focuses on such properties, and the theoretical roundness of the second inspection object estimated from the result of 3D imaging and the actuality of the second inspection object calculated from the result of 2D imaging By comparing the roundness of, it is possible to perform inspection of defects that can not be detected only by 2D inspection.

  The extraction unit extracts the XY position, the Z direction height, and the volume information of each of the plurality of first inspection objects from the 3D image of the first area, and the second information from the 2D image of the second area. The XY position and area of the object to be inspected are extracted, and the 3D information estimation unit interpolates or extrapolates Z-direction height and volume information of the plurality of first objects to be inspected, thereby the second object to be inspected Z-direction height and volume information at the XY position of the object are calculated, and the inspection unit is configured to calculate the area of the second inspection object extracted by the extraction unit and the second object estimated by the 3D information estimation unit. By comparing the volume information of the second inspection object calculated from the Z-direction height of the inspection object with the volume information of the second inspection object estimated by the 3D information estimation unit, the second object It is preferable to determine the quality of the inspection object.

  According to this configuration, the volume information of the second inspection object estimated from the result of the 2D imaging and the volume information of the second inspection object estimated from the result of the 3D imaging are detected only by the 2D inspection. It is possible to do an inspection of impossible defects.

  The object may be an electronic component, and the inspection object may be a solder that joins the electronic component and the substrate. For example, the present invention is suitable for inspection of electronic parts such as BGA, in which a plurality of solders of a certain size and shape are regularly arranged.

  The present invention can be grasped as an X-ray inspection apparatus having at least a part of the above configuration or function. Further, the present invention can also be grasped as a control method or an X-ray inspection method of an X-ray inspection apparatus including at least a part of the above processing. Further, the present invention is regarded as a program for causing a processor or a computer of an X-ray examination apparatus to execute each step of these methods, or a computer readable recording medium which records such a program non-temporarily. You can also. Each of the above configurations and processes can be combined with each other as long as there is no technical contradiction.

  According to the present invention, it is possible to inspect a defect which is difficult to detect with a 2D transmission image in a short time and with low exposure.

FIG. 1 is a view schematically showing 2D / 3D combination inspection. 2A to 2C show variations of visual field assignment. FIG. 3 is a diagram showing a hardware configuration of the X-ray inspection apparatus. FIG. 4 is a diagram showing a functional configuration of the X-ray inspection apparatus. FIG. 5 is a flowchart of 2D / 3D combination inspection. FIG. 6 is a diagram showing an example of 3D information of a 3D imaging ball extracted from a 3D image. FIG. 7 is a diagram showing an example of generating a projected image of a solder ball from a 3D image. FIG. 8 is a diagram showing an example of 2D information of a 2D imaging ball extracted from a 2D image. 9A-9C illustrate a method of approximately calculating the Z-direction height of a 2D imaging ball. FIG. 10 is a diagram showing a method of approximate calculation of volume information of a 2D imaging ball. 11A-11B illustrate a method of approximately calculating the theoretical position of a 2D imaging ball. FIG. 12 is a flowchart of misalignment inspection of the 2D imaging ball. FIG. 13 is a flowchart of a non-wetting test of the 2D imaging ball. 14A to 14C are diagrams for explaining the case where it is difficult to determine the bridge inspection. FIG. 15 is a flowchart of a bridge inspection of the 2D imaging ball. FIG. 16A is a side view of a BGA component, and FIG. 16B is an example of an X-ray transmission image of the BGA component. FIG. 17A shows an example of imaging an entire part in one field of view, and FIG. 17B shows an example of imaging a part in a plurality of fields of view.

  The present invention relates to an industrial X-ray inspection apparatus, and more particularly to an X-ray inspection apparatus for inspecting a 3D shape of an object including a plurality of inspection objects. There is no limitation on the type of object, but it is an object having a structure in which a plurality of inspection objects are regularly arranged, and the entire inspection region can not be imaged in one shot (that is, in a plurality of regions) The invention is preferably applicable to objects of a size which must be divided and imaged.

  Hereinafter, as a preferred application example of the present invention, an X-ray board inspection apparatus for inspecting the solder joint state of an electronic component on a surface mounting board will be described. In the case of the X-ray board inspection apparatus, the object is an electronic component (hereinafter, also simply referred to as a "component"), and the inspection object is a solder for joining the electronic component and the substrate. For example, in a component such as a BGA (ball grid array), so-called appearance inspection with a visible light camera is difficult because the solder is hidden under the component. On the other hand, in parts such as BGA, solder balls of a certain size and shape are regularly arranged. Therefore, such parts are one of the objects to which the present invention is preferably applied.

(2D ・ 3D combination inspection)
First, an outline of “2D · 3D combination inspection”, which is a feature of the X-ray inspection apparatus of the present embodiment, will be described.

  As described in the section of the background art, 2D inspection has the advantage that the number of imaging times is short and the inspection time is short, but there is a problem that there are some types of defects that are difficult to detect. On the other hand, 3D inspection is capable of higher-level inspection than 2D inspection, but has the disadvantage that the number of imaging times is much more than that of 2D inspection and the inspection time is long.

  Therefore, in the present embodiment, the area for performing 3D imaging is limited to a part of the area to be inspected, and only the 2D imaging is performed for the remaining area. Then, 3D information on the remaining area is estimated from 3D information obtained on a part of the area, and a pseudo 3D inspection is performed using the estimated 3D information. This inspection method is referred to herein as "2D / 3D combination inspection".

  FIG. 1 schematically shows an example in which 2D / 3D combination inspection is performed on a BGA component. In this example, a region to be inspected (broken line) set to include the entire part is divided into 5 rows × 4 columns = 20 regions. Each divided area corresponds to the field of view of the imaging system. The 3D imaging is performed only on the four corners and the five central areas A1 to A5 (hatched rectangles), and the remaining areas B1 to B15 are only subjected to the 2D imaging. The solder balls of the BGA part have substantially fixed size and shape, and are arranged according to a predetermined rule. Therefore, if the 3D information of the solder balls in the regions A1 to A5 is known by 3D imaging, the 3D information on the solder balls in the remaining regions B1 to B15 is estimated by geometrical calculation (for example, interpolation or extrapolation) Is possible. In inspection of 2D imaging regions B1 to B15, in addition to 2D information obtained from 2D images of regions B1 to B15, using 2D information of regions B1 to B15 estimated from 3D information of regions A1 to A5 Pseudo 3D inspection can also be performed on the imaging regions B1 to B15.

  The 2D · 3D combination inspection is suitable when the region to be inspected is larger than the field of view of the imaging system, that is, when the region to be inspected needs to be divided and imaged into two or more regions. The division method and division number of the area can be designed arbitrarily, and the allocation of the area for performing 2D imaging and the area for performing 3D imaging (hereinafter, referred to as “allocation of visual field”) can also be designed arbitrarily. For example, as shown in FIG. 1, the field of view may be automatically assigned according to a predetermined rule such as “the four corners and the center area are 3D imaging and the other areas are 2D imaging”. Assignment may be made. Besides, as shown in FIG. 2A, 2D imaging may be performed in all regions (in both corner regions, both 2D imaging and 3D imaging are performed), and as shown in FIG. 2B, at the time of 3D imaging. The field of view size of and the field of view size at the time of 2D imaging may be made different. Also, as shown in FIG. 2C, the area where the back surface component is present may be preferentially assigned to the 3D imaging area. This is because 2D imaging can not be performed if there is a back surface part.

  The 2D-3D combination inspection can be preferably applied to, for example, a solder inspection (tray inspection) of an electronic component placed on a tray, in addition to the inspection of the surface mounting substrate.

(X-ray inspection device)
Next, a specific configuration of the X-ray inspection apparatus having the 2D-3D combination inspection function will be described. FIG. 3 is a view schematically showing a hardware configuration of the X-ray inspection apparatus according to the embodiment of the present invention.

  The X-ray inspection apparatus 1 generally includes a control device 10, a stage 11, an X-ray source 12, and an X-ray detector 13. The X-ray inspection apparatus 1 is an X-ray board inspection apparatus which inspects the solder joint of the component 15 mounted on the substrate 14. In the present embodiment, as shown in FIG. 3, the orthogonal XYZ coordinate system is set so that the substrate surface and the XY plane are parallel and the substrate surface and the Z axis are perpendicular.

  The X-ray source 12 is a means for irradiating the substrate 14 with X-rays, and is constituted of, for example, a cone beam type or fan beam type X-ray generator. The X-ray detector 13 is an imaging unit that detects X-rays transmitted through the substrate 14 and outputs data of an X-ray transmission image, and includes, for example, a scintillator and a two-dimensional CMOS sensor. The stage 11 is a means for holding and transporting the substrate 14, and performs alignment of the field of view of the imaging system including the X-ray source 12 and the X-ray detector 13 with the component 15. When moving the field of view, the stage 11 may be moved, or the imaging system (the X-ray source 12 and the X-ray detector 13) may be moved, or both of the stage 11 and the imaging system may be moved. You may move.

  The X-ray inspection apparatus 1 can execute both 2D imaging that acquires a 2D image by one X-ray irradiation and 3D imaging that acquires a 3D image by multiple X-ray irradiations. When 2D imaging is performed, X-rays are irradiated from a direction perpendicular to the substrate surface (that is, the Z direction). On the other hand, in the case of performing 3D imaging, one view is imaged a plurality of times while changing the irradiation direction of X-rays. Therefore, the X-ray inspection apparatus 1 also has a moving mechanism (not shown) for changing the irradiation direction of the X-ray to the substrate 14. As a configuration of the moving mechanism, for example, a method in which the X-ray source 12 and the X-ray detector 13 rotate around the substrate 14, a method in which the X-ray source 12 and the X-ray detector 13 are fixed and the substrate 14 rotates There are various methods, such as a method in which the X-ray source 12 and the X-ray detector 13 rotate with the substrate 14 interposed therebetween, but any method may be adopted.

  The control device 10 controls and processes the X-ray inspection apparatus 1 (for example, movement of visual field, irradiation of X-rays, capture of X-ray transmission image, generation of 2D image, generation of 3D image, estimation of 3D geometric information, inspection Processing, cooperation with external devices, data transmission, etc.). The control device 10 may be configured by, for example, a general-purpose computer provided with a CPU (processor), memory, storage (hard disk etc.), input device (keyboard, mouse, touch panel etc.), display device, communication I / F etc. it can. In this case, the control device 10 may be configured by one computer, or the control device 10 may be realized by cooperation of a plurality of computers. For example, distributed computing or cloud computing technology may be used. The functions of the control device 10 described later are realized by the CPU (processor) executing necessary programs. However, part or all of the functions can be configured by circuits such as an ASIC or an FPGA.

(Control device)
FIG. 4 is a block diagram showing a functional configuration of the control device 10. As shown in FIG. The control device 10 has a visual field setting unit 100, a storage unit 101, a 2D image generation unit 102, a 3D image generation unit 103, an extraction unit 104, a 3D information estimation unit 105, an inspection unit 106, and a result output unit 107 as its functions. doing.

  The visual field setting unit 100 is a function of setting an inspection area and performing area division and visual field allocation in 2D-3D combination inspection.

  The storage unit 101 is a function of storing a setting file of the X-ray examination apparatus 1, an examination program, and the like. The setting file is, for example, data in which setting values of the X-ray source 12 and the X-ray detector 13 are described. The inspection program is data defining an X-ray inspection procedure, and is created and stored in advance for each type of object. The inspection program includes information on the object and the inspection object (for example, the size of the board, part number, position, size, etc.), visual field assignment condition for each type of part, inspection logic (for example, measurement items to be measured from an image) It is preferable that definitions such as inspection criteria (threshold, value range), processing according to the determination result, and the like for determining the quality of the inspection be included. The storage unit 101 is configured of a non-volatile storage medium.

  The 2D image generation unit 102 is a function of generating a 2D image (a transmission image or the like) of the component 15 based on data obtained by one X-ray irradiation. In the present embodiment, the X-ray source 12, the X-ray detector 13, and the 2D image generation unit 102 constitute a 2D processing unit that performs 2D imaging.

The 3D image generation unit 103 is a function of generating a 3D image (such as 3D volume data) of the component 15 based on data obtained by a plurality of X-ray irradiations. In the present embodiment, the X-ray source 12, the X-ray detector 13, and the 3D image generation unit 103 constitute a 3D processing unit that performs 3D imaging. As a 3D imaging method, CT, tomosynthesis, and other methods can be used. Moreover, as a reconstruction algorithm of 3D image, a simple back projection method, a filtered correction back projection method, a successive approximation method (SIRT (Simultaneous Reconstruction Technique) method, ART (Algebraic Reconstruction Technique) method), a search method (
(Gradient Method, conjugate gradient method, steepest descent method), etc., and any algorithm may be used.

  The extraction unit 104 is a function of extracting (measuring) information on the inspection object from the 2D image and the 3D image. The 2D information extracted from the 2D image includes, for example, geometric information such as the position, size, width, distance, area, and shape of the inspection object in the XY plane, and the 3D information extracted from the 3D image includes, for example , Geometric information such as position, size, height, width, distance, area of arbitrary cross section, volume, shape, etc. of the object in XYZ space. Of course, information other than these geometric information may be extracted. The 3D information estimation unit 105 is a function of estimating 3D information of the object in the 2D image based on the information extracted from the 3D image and the information extracted from the 2D image. The inspection unit 106 is a function of inspecting the inspection object using the information extracted by the extraction unit 104 and the information estimated by the 3D information estimation unit 105. The result output unit 107 is a function of outputting the information extracted by the extraction unit 104, the information estimated by the 3D information estimation unit 105, the inspection result of the inspection unit 106, and the like to a display device or an external device. Details of these functions will be described later.

(Operation of X-ray inspection system)
An example of the processing operation of the 2D-3D combination inspection by the X-ray inspection apparatus 1 will be described with reference to the flowchart of FIG. 5.

When inspection is started, first, the visual field setting unit 100 reads information of a part to be inspected from the storage unit 101, and sets an inspection area for the part (step S500). The inspection area is generally set to include all the inspection objects (solder balls) of the part. Next, the visual field setting unit 100 compares the size of the visual field of the inspection area and the imaging system, and if the inspection area is larger than the visual field (that is, if the whole can not be imaged in one shot) While dividing into a plurality of areas, allocation of visual fields of 2D imaging and 3D imaging is performed (step S501). At this time, the visual field setting unit 100 may perform area division and visual field assignment in accordance with an input instruction from the operator, or automatically perform area division and visual field according to a rule (visual field assignment condition) registered in advance in the storage unit 101. Assignment may be made. Here, the following description will be made on the assumption that 3D imaging areas A1 to A5 and 2D imaging areas B1 to B15 are allocated as in the example of FIG. 1. In order to distinguish the solder balls in the 3D imaging area A1 to A5 from the solder balls in the 2D imaging area B1 to B15, the former is "3D imaging ball" or "first inspection object" and the latter is "2D imaging ball" Or it calls it a "2nd to-be-tested object."

  Next, 2D imaging is performed on each of the 2D imaging regions B1 to B15, and a 2D image is generated by the 2D image generation unit 102 (step S502). Further, 3D imaging is performed on each of the 3D imaging regions A1 to A5, and the 3D image generation unit 103 generates a 3D image (step S503).

Next, the extraction unit 104 analyzes 3D images of 3D imaging regions (first regions) A1 to A5, and extracts 3D information of 3D imaging balls included in each region (step S504). In the present embodiment, as shown in FIG. 6, as the 3D information, the position (X _B , Y _B , Z _B ) of the pad side end (first end) 61 of the solder ball 60, the component side end of the solder ball 60 Position (second end) 62 (X _T , Y _T , Z _T ), height in the Z direction H of the solder ball 60 (= Z _T- Z _B ), and projection image 63 of the solder ball 60 onto the XY plane area a, the position of the center point 64 of the projected image 63 in the XY plane of the solder balls _{60 (X C, Y C)} , such as the volume information V of the solder balls 60 are extracted.

For the volume information V, an exact volume value may be calculated from the 3D shape of the solder ball 60, but in the present embodiment, the volume value is approximated by the following equation in a simplified manner. For the purpose of using for the non-wetting inspection to be described later, even with this approximate value V, it is possible to make a sufficient quality determination.

Volume information V = Z-direction height H of solder ball × area A of projected image of solder ball

  The projected image 63 of the solder ball 60 may be acquired by any method. For example, as shown in FIG. 7, a plurality of XY slice images 71a, 71b, 71c corresponding to different Z positions 70a, 70b, 70c are acquired from the 3D image of the solder ball 60 and projected by overlapping them. The shape of the image 63 may be simply estimated. Alternatively, the projected image of the solder ball 60 may be measured by performing 2D imaging also on the solder ball 60 in the 3D imaging region.

Next, the extraction unit 104 analyzes 2D images of 2D imaging regions (second regions) B1 to B15, and extracts 2D information of 2D imaging balls included in each region (step S505). For example, as shown in FIG. 8, as 2D information, the position (X _C , Y _C ) of the center point 81 in the XY plane of the solder ball 80, the area A of the solder ball 80, the solder ball 82 adjacent to the solder ball 80 An inter-solder distance L with (an adjacent inspection object), a roundness of the solder ball, a major axis angle 83, and the like are extracted. The roundness of a solder ball is a measure indicating how close the outer shape of the projected image of the solder ball is to a geometrically perfect circle, where the case of being coincident with the perfect circle is 100%, which is true. The smaller the distance from the circle, the smaller the value. For calculation of the degree of circularity, for example, it is possible to use a calculation formula in which the major axis / minor axis / area / perimeter length etc. are variables. The major axis angle 83 is an angle that the major axis of the projected image of the solder ball makes with the X axis.

  Subsequently, the 3D information estimation unit 105 estimates 3D information of the 2D imaging ball in the 2D imaging areas B1 to B15 based on the 3D information of the 3D imaging ball obtained in step S504 (step S506). In this embodiment, (1) height in the Z direction of the 2D imaging ball, (2) volume information of the 2D imaging ball, (3) theoretical position of the 2D imaging ball, and (4) pad side end and parts of the 2D imaging ball The four types of values of the side end displacement amount are calculated. Hereinafter, an example of the calculation method of each value is demonstrated.

(1) Z Direction Height of 2D Imaging Ball FIGS. 9A to 9C show an example of a method of approximately calculating the Z direction height of a 2D imaging ball based on 3D information of the 3D imaging ball. 9A to 9C are side views, in which the 3D imaging balls SB1 and SB4 are drawn by solid lines, and the 2D imaging balls SB2, SB3 and SB5 are drawn by broken lines. In addition, although the example in which the 3D imaging ball and the 2D imaging ball are arranged one-dimensionally is used here for convenience of explanation, the 3D imaging ball and the 2D imaging ball are two-dimensionally arranged as shown in FIG. Similar approximation calculations are possible in this case as well.

The 3D information estimation unit 105 uses 3D position coordinates of the pad side end and the part side end (black circle in FIG. 9A) as 3D information of the 3D imaging ball. In FIG. 9A, the XYZ positions of the pad side end and component side end of the 3D imaging ball SB1 are (X1 _B , Y1 _B , Z1 _B ) and (X1 _T , Y1 _T , Z1 _T ), respectively, and the 3D imaging ball is The XYZ positions of the pad side end and the component side end of SB 4 are (X 4 _B , Y 4 _B , Z 4 _B ) and (X 4 _T , Y 4 _T , Z 4 _T ), respectively.

On the other hand, the 3D information estimation unit 105 uses 2D position coordinates of the center of the solder ball (one-dot chain line in FIG. 9B) as 2D information of the 2D imaging ball. In Figure 9B, 2D imaging ball SB2, SB3, XY position of the center of SB5 respectively _{(X2 C, Y2 C),} an _{(X3 C, Y3 C),} (X5 C, Y5 C).

The 3D information estimation unit 105 interpolates or extrapolates the 3D position coordinates of the pad side ends of the plurality of 3D imaging balls SB1 and SB4 to obtain the pad side ends at the XY positions of the 2D imaging balls SB2, SB3 and SB5. Z position of the part _{Z2 B,} calculates the Z3 _B, Z5 B. Further, the 3D information estimation unit 105 interpolates or extrapolates the 3D position coordinates of the component side end portions of the plurality of 3D imaging balls SB1 and SB4 to parts at the XY positions of the 2D imaging balls SB2, SB3 and SB5. calculating the Z position of the side end portion _{Z2 T, Z3 T, Z5 T} . The points indicated by white circles in FIG. 9C are the pad side end and the component side end of the 2D imaging ball calculated by interpolation or extrapolation. For the calculation of interpolation or extrapolation, approximation by linear equation (linear interpolation or linear extrapolation) may be used, or approximation by polynomial of n-th order (n ≧ 2) may be used.

Then, the 3D information estimation unit 105 calculates the Z-direction height H (= Z _T −Z _B ) of each 2D imaging ball from the Z position of the pad side end and the Z position of the part side end obtained by approximation. Do. Since the Z positions of the pad side end and the part side end of each solder ball can be assumed to have only continuous fluctuation along the warp or tilt of the substrate or part, the above approximate calculation can be performed with sufficient accuracy. It is possible to estimate the Z-direction height of the 2D imaging ball.

  In this embodiment, the Z position of the pad side end and the Z position of the part side end are separately calculated, and then the Z direction height of the solder ball is calculated by taking the difference thereof. Approximation calculation of H is not limited to this method. For example, the Z-direction height corresponding to the XY position of each 2D imaging ball may be directly obtained by calculating the Z-direction height of each of the plurality of 3D imaging balls and interpolating or extrapolating them.

(2) Volume Information of 2D Imaging Ball FIG. 10 shows an example of a method of approximately calculating volume information of a 2D imaging ball based on 3D information of the 3D imaging ball. FIG. 10 is a side view, in which 3D imaging balls SB1 and SB4 are drawn by solid lines, and 2D imaging balls SB2, SB3 and SB5 are drawn by broken lines. In addition, although the example in which the 3D imaging ball and the 2D imaging ball are arranged one-dimensionally is used here for convenience of explanation, the 3D imaging ball and the 2D imaging ball are two-dimensionally arranged as shown in FIG. Similar approximation calculations are possible in this case as well.

The 3D information estimation unit 105 uses 2D position coordinates of the center of the solder ball and volume information as 3D information of the 3D imaging ball. In FIG. 10, the XY position of the center of the 3D imaging ball SB1 is (X1 _C , Y1 _C ), and the volume information is V1. Further, the XY position of the center of the 3D imaging ball SB4 is (X4 _C , Y4 _C ), and the volume information is V4. On the other hand, the 3D information estimation unit 105 uses 2D position coordinates of the center of the solder ball as 2D information of the 2D imaging ball. In Figure 10, 2D imaging ball SB2, SB3, XY position of the center of SB5 respectively _{(X2 C, Y2 C),} an _{(X3 C, Y3 C),} (X5 C, Y5 C).

  The 3D information estimation unit 105 interpolates or extrapolates the volume information V1 and V4 of the plurality of 3D imaging balls SB1 and SB4 to obtain volume information V2, V3 and V3 at the XY positions of the 2D imaging balls SB2, SB3 and SB5. Calculate V5. For the calculation of interpolation or extrapolation, approximation by linear equation (linear interpolation or linear extrapolation) may be used, or approximation by polynomial of n-th order (n ≧ 2) may be used.

(3) Theoretical Position of 2D Imaging Ball FIGS. 11A and 11B show an example of a method of approximately calculating the XY position (theoretical position) where the 2D imaging ball should exist based on the 3D information of the 3D imaging ball. . 11A is a top view showing the XY position of the pad side end, and FIG. 11B is a top view showing the XY position of the part side end. The 3D imaging ball is drawn as a solid line, and the 2D imaging ball is drawn as a dashed line.

The 3D information estimation unit 105 uses 3D position coordinates of the pad side end (black circle in FIG. 11A) and the component side end (black circle in FIG. 11B) as 3D information of the 3D imaging ball. In the example of FIGS. 11A and 11B, the XYZ positions of the pad side end and the component side end of the 3D imaging ball SB1 are (X1 _B , Y1 _B , Z1 _B ) and (X1 _T , Y1 _T , Z1 _T ), respectively. There, XYZ position of the pad end and the component-side end of the 3D imaging ball SB3 are each _{(X3 B, Y3 B, Z3} B) and _{(X3 T, Y3 T, Z3} T).

  The 3D information estimation unit 105 refers to the inspection program stored in the storage unit 101, and acquires the arrangement information of the solder balls of the component to be inspected. The placement information is information that defines the number, placement, spacing, and the like of the solder balls, and represents the design location of the solder balls. For example, in the example of FIG. 11A, arrangement information is obtained in which fifteen solder balls are arranged at equal intervals in five rows and three rows.

  Next, the 3D information estimation unit 105 calculates the relative positional relationship of the 2D imaging ball to the 3D imaging ball based on the placement information of the solder ball. For example, in the example of FIG. 11A, one 2D imaging ball SB2 exists just in the middle of the 3D imaging balls SB1 and SB3, and three 2D imaging balls are arranged at equal intervals between the 3D imaging balls SB1 and SB4. It is possible to obtain a positional relationship such as The 3D information estimation unit 105 interpolates or outputs the 3D position coordinates of the pad side end and the part side end of the 3D imaging ball based on the positional relationship between the 3D imaging ball and the 2D imaging ball calculated from the arrangement information. By inserting, 3D position coordinates of the pad side end and the part side end of each of the 2D imaging balls are calculated. For the calculation of interpolation or extrapolation, linear approximation (linear interpolation or linear extrapolation) may be used, or n-th order (n22) polynomial approximation may be used.

The points indicated by white circles in FIGS. 11A and 11B are the pad side end and the component side end of the 2D imaging ball calculated by interpolation or extrapolation. For example, in the case of using linear approximation, the 3D position coordinates of the pad side end and the component side end of the 2D imaging ball SB2 are ((X1 _B + X3 _B ) / 2, (Y1 _B + Y3 _B ) / 2, ( _{Z1 B + Z3 B) / 2} ), obtained as _{((X1 T + X3 T)} / 2, (Y1 T + Y3 T) / 2, (Z1 T + Z3 T) / 2). Further, by averaging the XY position of the XY position and the component-side end portion of the pad end, XY position of the center of the 2D imaging ball _{SB2 ((X1 B + X3 B} + X1 T + X3 T) / 4, (Y1 B + Y3 _B + Y 1 _T + Y 3 _T ) / 4) is obtained.

  Note that the position of the 2D imaging ball calculated here represents a virtual position estimated from the relative positional relationship with the 3D imaging ball, that is, a theoretical position where the 2D imaging ball should originally exist, and the 2D imaging Note that it is not the actual position of the ball. The theoretical position of the 2D imaging ball calculated here is used as a reference (correct) position of the 2D imaging ball in the subsequent pseudo 3D inspection.

(4) Deviation amount between pad side end and component side end of 2D imaging ball The 3D information estimation unit 105 calculates XYZ positions and parts of the pad side end (first end) of the 2D imaging ball calculated in (3) Using the XYZ position of the side end (the second end), calculate the amount of deviation in the XY plane between the pad side end and the part side end (hereinafter, also simply referred to as “part-pad deviation amount”) . The component-to-pad displacement amount is an index that represents the displacement of the mounting position of the component relative to the pad (or the substrate). Further, the component-to-pad displacement amount can be said to be an index that represents the inclination of the solder ball (the inclination of the central axis of the solder ball (the straight line connecting the pad side end and the part side end) to the Z axis).

  As described above, extraction of 3D information of the 3D imaging ball (step S504), extraction of 2D information of the 2D imaging ball (step S505), and estimation of 3D information of the 2D imaging ball (step S506) are completed, Proceed to the inspection process of each solder ball.

  First, in step S507, the inspection unit 106 performs 3D inspection on the 3D imaging ball. Inspection items of 3D inspection include misalignment inspection, mounting misalignment inspection, non-wetting inspection, bridge inspection, ball height inspection and the like. The misalignment inspection is an inspection to check whether each solder is not deviated from the reference position. For example, the inspection unit 106 compares the XY position of the center of the solder ball with the reference position registered in the inspection program, and determines that the product is defective if the difference exceeds the threshold and is non-defective if the difference is less than the threshold. The mounting misalignment inspection is an inspection to check whether the entire part is not offset from the pad. For example, the inspection unit 106 calculates the component-pad displacement amount for each of the solder balls included in the component, and determines that the component is defective if the maximum displacement amount exceeds the threshold and is non-defective if it is less than the threshold. The non-wetting inspection is an inspection to check whether the solder on the part side and the solder on the pad side are separated or not fused. The bridge inspection is an inspection to check whether adjacent electrodes are electrically connected by solder (bridging). The ball height inspection is an inspection to check whether the Z-direction height of the solder ball is too high or too low. For example, the inspection unit 106 determines that the solder ball is defective if the height in the Z direction exceeds the first threshold and is less than the second threshold (second threshold <first threshold), and is otherwise non-defective. Incidentally, the inspection method described here is only one example, a specific method of 3D inspection may be by any method including existing technologies.

  Next, in step S508, the inspection unit 106 performs a pseudo 3D inspection of the 2D imaging ball using the 3D information estimated in step S506. Inspection items of the pseudo 3D inspection include misalignment inspection, mounting misalignment inspection, non-wetting inspection, bridge inspection, ball height inspection and the like. A specific example of the pseudo 3D inspection process will be described later.

  Thereafter, in step S509, the result output unit 107 outputs the results of the 3D inspection in step S507 and the pseudo 3D inspection in step S508 to the display device or the external device. As the inspection result, in addition to the determination result (non-defective product / defective result) of each inspection item, information such as the measurement value, cross section position, XY coordinates, image, etc. used in each inspection item may be output.

(Pseudo 3D inspection)
An example of a process of pseudo 3D inspection of a 2D imaging ball performed in step S508 of FIG. 5 will be described. In addition, about the threshold value used by test | inspection, an appropriate value shall be previously set for every inspection item for every inspection item.

(1) Misalignment Inspection of 2D Imaging Ball FIG. 12 shows a processing flow of misalignment inspection of the 2D imaging ball. The inspection unit 106 acquires the theoretical position (XY position) of the 2D imaging ball estimated from the 3D information of the 3D imaging ball (step S120), and the actual center position (XY) of the 2D imaging ball measured from the 2D image Position) is acquired (step S121). Then, the inspection unit 106 calculates the difference between the theoretical position and the actual position, and if the difference is equal to or less than the threshold (YES in step S122), determines that the product is good (step S123). It determines with (NO of step S122) misregistration defect (step S124).

(2) Mounting displacement inspection of 2D imaging ball The inspection unit 106 acquires the component-pad displacement amount of the 2D imaging ball estimated from 3D information of the 3D imaging ball. Then, the inspection unit 106 determines that the displacement amount is equal to or less than the threshold value as a non-defective product, and determines the mounting displacement defect if the displacement amount exceeds the threshold value.

(3) Unwetting Inspection of 2D Imaging Ball FIG. 13 shows a processing flow of nonwetting inspection of the 2D imaging ball. First, the inspection unit 106 acquires the roundness (measured value) of the 2D imaging ball measured from the 2D image (step S130). Then, the inspection unit 106 compares the roundness (measured value) with the threshold (step S131).

  If the roundness of the 2D imaging ball is less than the threshold value (NO in step S131), the process proceeds to an inspection as to whether the solder ball on the component side and the cream solder on the pad side are fused or not fused (Head In Pillow). In the case where the solder ball and the cream solder are fused, it can be assumed that the roundness of the solder ball depends on the part-pad offset amount. Therefore, the inspection unit 106 estimates the roundness in the case where the 2D imaging ball is non-defective based on the part-pad displacement amount of the 2D imaging ball (step S132), and the estimated value and the roundness The actual value is compared (step S133). If the difference between the estimated value and the actual measurement value is equal to or less than the threshold (YES in step S133), the inspection unit 106 determines that the product is good (step S134). Fusion state) is determined (step S135).

  On the other hand, if the roundness of the 2D imaging ball is equal to or greater than the threshold (YES in step S131), the process proceeds to an inspection of the presence or absence of a gap between the solder ball on the component side and the cream solder on the pad side. When the solder ball is separated from the cream solder, it can be assumed that the projected area of the solder ball is larger than in the case of the non-defective product. Thus, the inspection unit 106 calculates the volume information Va (measured value) of the 2D imaging ball by multiplying the area A of the 2D imaging ball measured from the 2D image by the height H of the 2D imaging ball estimated from 3D information. At the same time (step S136), the volume information Vb (estimated value) of the 2D imaging ball estimated from the 3D information is obtained (step S137), and the measured value Va of the volume information is compared with the estimated value Vb (step S138). If the difference between the actual measurement value Va and the estimated value Vb is equal to or less than the threshold (YES in step S138), the inspection unit 106 determines that the product is good (step S139). If the difference exceeds the threshold (NO in step S138) It is determined that (the open state) (step S140).

  In the present embodiment, the unfused state is inspected by evaluating the roundness, but the major axis angle may be used for evaluation instead of or together with the roundness. This is because it can be assumed that the major diameter angle of the solder ball depends on the component-pad displacement direction.

(4) Bridge inspection of a 2D imaging ball In a bridge inspection, not only a defective part that has a bridge is detected, but also a tendency to cause a bridge (a distance between adjacent solder balls is short) Is also detected as a defect. However, there are cases where it is difficult to determine the pass / fail by merely evaluating the inter-solder distance L measured from the 2D image. This will be described with reference to FIGS. 14A to 14C. FIG. 14A is an example of a non-defective product, and FIG. 14B is an example of a bridge failure. From a 2D image, while the distance L between the solders is sufficiently wide in FIG. 14A, the distance L between the solders is clearly narrow in FIG. 14B. Therefore, by evaluating whether the inter-solder distance L is larger than the threshold value TH, FIG. 14A can be correctly determined as a non-defective product and FIG. 14B as a defect. On the other hand, FIG. 14C is also an example of a non-defective product. However, the component and the pad are offset in the XY plane, and the solder ball is inclined with respect to the Z axis. The solder ball in the 2D image is a projected image on the XY plane, so the projected image becomes larger as the solder ball is inclined to the Z axis, and the distance between the solder (apparent distance) on the 2D image is It becomes smaller. Therefore, by simply comparing the distance L between the solder and the threshold value TH, the non-defective product of FIG. 14C can not be distinguished from the bridge failure of FIG. 14B, and the state of FIG.

  So, in this embodiment, in addition to the distance L between solders, a bridge inspection is performed in consideration of the displacement amount between components and pads. Specifically, the threshold value TH is dynamically changed in accordance with the component-pad displacement amount, thereby suppressing erroneous determination in the case shown in FIG. 14C.

  FIG. 15 shows an example of a processing flow of bridge inspection of a 2D imaging ball. First, the inspection unit 106 acquires the inter-solder distance L measured from the 2D image (step S150). When the inter-solder distance L is zero (that is, when the 2D imaging ball to be inspected is in contact with the adjacent solder ball), the inspection unit 106 determines that the bridge is defective (steps S151 and S152). Further, when the distance L between the solders is larger than the threshold TH, the inspection unit 106 determines that the product is non-defective (step S153). Otherwise, the process proceeds to step S154.

  In step S154, the inspection unit 106 acquires the part-pad misalignment amount of the 2D imaging ball estimated from the 3D information. If the amount of deviation is greater than zero, the inspection unit 106 corrects the threshold TH in accordance with the amount of deviation (step S155). For example, a value obtained by subtracting the deviation amount from the threshold value TH may be set as a new threshold value TH. Here, when the displacement direction between the component and the pad and the arrangement direction of the solder balls are different (not parallel), the amount of displacement along the arrangement direction of the solder balls is calculated, and the deviation amount is subtracted from the threshold value TH. You may

  Then, the inspection unit 106 compares the inter-solder distance L with the corrected threshold TH (step S156), and determines that the inter-solder distance L is non-defective if the inter-solder distance L is larger than the threshold TH (step S153) In the case of “1”, it is determined that the bridge is defective (step S152).

(5) Ball Height Inspection of 2D Imaging Ball The inspection unit 106 acquires the Z-direction height H of the 2D imaging ball estimated from 3D information of the 3D imaging ball. Then, the inspection unit 106 determines that the height H in the Z direction exceeds the first threshold TH1 and the case where the height H is smaller than the second threshold TH2 (provided that TH2 <TH1), otherwise (TH2). In the case of ≦ H ≦ TH1, it is judged as a non-defective product.

(Advantages of the present embodiment)
According to the 2D / 3D combination inspection of the present embodiment described above, the area to be subjected to 3D imaging is limited to a part of the inspection area (only 3D imaging areas A1 to A5). As compared with 3D imaging, the number of imaging times and the number of X-ray irradiations can be reduced, and the time required for imaging and inspection can be shortened, and the exposure dose can be reduced. In addition, 3D information of 2D imaging balls in 2D imaging areas B1 to B15 is estimated based on 3D information of 3D imaging balls extracted from 3D images of 3D imaging areas, and the 3D information is used for inspection of 2D imaging balls By doing this, it is possible to conduct a pseudo 3D inspection even for solder balls that are only subjected to 2D imaging. As a result, inspection can be performed even for defects that are difficult to detect in the conventional 2D inspection (for example, non-wetting state such as unfused or open, bridge failure, etc.).

  Further, in the present embodiment, the theoretical position of the 2D imaging ball is determined based on the 3D information of the 3D imaging ball, and the theoretical position is used for the inspection of the 2D imaging ball. As a result, it is possible to obtain a reasonable inspection result in which the actual state of the part (for example, manufacturing error, positional deviation of the whole part, etc.) is taken into consideration, and inspection accuracy and reliability can be improved.

  Further, in the present embodiment, when the bridge inspection is performed, the displacement amount between the component and the pad and the displacement direction are considered. Therefore, the appropriate inspection result in which the inclination of the solder ball is considered can be obtained, and the inspection accuracy and the reliability can be improved.

  The configuration of the embodiment described above is merely one example of the present invention. The scope of the present invention is not limited to the above-described embodiment, and various modifications are possible within the scope of the technical idea thereof. For example, 2D information measured from a 2D image or 3D information measured from a 3D image may be any image information that can be extracted from an image. Further, the type of 3D information of the 2D imaging ball estimated from 3D information of the 3D imaging ball and the estimation method thereof can be arbitrarily designed.

A1 to A5: 3D imaging area (first area), B1 to B15: 2D imaging area (second area)
1: X-ray inspection apparatus 10: controller 11: stage 12: X-ray source 13: X-ray detector 14: substrate 15: component 100: field of view setting unit 101: storage unit 102: 2D Image generation unit 103: 3D image generation unit 104: extraction unit 105: 3D information estimation unit 106: inspection unit 107: result output unit 60: solder ball (inspection object) 61: pad side end ( First end), 62: component side end (second end), 63: projection image, 64: center point 70a, 70b, 70c: slice position, 71a, 71b, 71c: slice image 80: solder ball, 81: Center point, 82: adjacent solder ball, 83: major diameter angle 160: component, 161: substrate, 162: solder ball, 163: cream solder

Claims (28)

A processing apparatus for X-ray examination for examining an object including a plurality of examination objects by X-ray,
A 3D image acquisition unit configured to acquire a 3D image based on a plurality of X-ray transmission images having different projection directions with respect to a first area that is a partial area of an inspection area set for the object;
A 2D image acquisition unit configured to acquire a 2D image, which is an X-ray transmission image, of a second area that is at least partially different from the first area of the inspection area;
The 3D information of the first inspection object in the first region is extracted from the 3D image of the first region, and the inspection in the second region is extracted from the 2D image of the second region An extraction unit for extracting 2D information of a second test object which is a test object different from the first test object among objects;
A 3D information estimation unit configured to estimate 3D information of the second inspection object based on 3D information of the first inspection object extracted by the extraction unit;
The second inspection object is inspected using the 2D information of the second inspection object extracted by the extraction unit and the 3D information of the second inspection object estimated by the 3D information estimation unit. Inspection department,
An X-ray examination processing apparatus characterized by having: A plurality of first inspection objects are included in the first region,
The 3D information estimation unit according to claim 1, wherein the 3D information of the second inspection object is calculated by interpolating or extrapolating 3D information of the plurality of first inspection objects. X-ray examination processor. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
The extraction unit extracts an XY position and a Z-direction height of each of the plurality of first inspection objects from the 3D image of the first region, and the second inspection object from the 2D image of the second region. Extract the XY position,
The 3D information estimation unit is characterized in that the Z-direction height at the XY position of the second test object is calculated by interpolating or extrapolating the Z-direction heights of the plurality of first test objects. The processing apparatus for X-ray examination according to claim 2. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
The extraction unit extracts the XY position and volume information of each of the plurality of first inspection objects from the 3D image of the first region, and the XY position of the second inspection object from the 2D image of the second region To extract
The 3D information estimation unit calculates volume information at the XY position of the second inspection object by interpolating or extrapolating the volume information of the plurality of first inspection objects. The processing apparatus of the X-ray examination as described in 2 or 3. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
The extraction unit extracts an XY position of each of the plurality of first inspection objects from the 3D image of the first area,
The 3D information estimation unit calculates an XY position where the second test object should exist by interpolating or extrapolating the XY positions of the plurality of first test objects. The processing apparatus of the X-ray examination of any one of 2-4. The XYZ coordinate system is placed so that the 2D image is an X-ray transmission image projected onto the XY plane, and both ends of the inspection object in the Z direction are referred to as a first end and a second end,
The extraction unit extracts an XYZ position of the first end and an XYZ position of the second end of each of the plurality of first inspection objects from the 3D image of the first area,
The 3D information estimation unit calculates an XYZ position at which the first end of the second inspection object should be present by interpolating or extrapolating the XYZ positions of the first end of the plurality of first inspection objects. And calculating the XYZ position where the second end of the second inspection object should be present by interpolating or extrapolating the XYZ positions of the second end of the plurality of first inspection objects. The processing apparatus of the X-ray inspection of any one of Claims 2-4. The 3D information estimation unit determines the first end and the second end of the second inspection object from the estimated XYZ position where the first end should exist and the XYZ position where the second end should exist. 7. The processing apparatus for X-ray examination according to claim 6, wherein the deviation amount in the XY plane is calculated. The 3D information estimation unit averages the estimated XYZ position where the first end should exist and the XYZ position where the second end should exist, thereby determining the XY position where the second inspection object should exist. The processor for X-ray examination according to claim 6 or 7, wherein it calculates. The extraction unit extracts an actual XY position of the second inspection object from the 2D image of the second area,
The inspection unit compares an actual XY position of the second inspection object extracted by the extraction unit with an XY position where the second inspection object estimated by the 3D information estimation unit should be present. The processing apparatus for X-ray examination according to claim 5 or 8, characterized in that the quality of the second inspection object is determined. The extraction unit extracts the distance between the second inspection object and the adjacent inspection object adjacent to the second inspection object from the 2D image of the second area,
The inspection unit is based on the distance extracted by the extraction unit and an amount of deviation of the first end and the second end of the second inspection object in the XY plane estimated by the 3D information estimation unit. The processing apparatus for X-ray examination according to claim 7, wherein the quality of the second inspection object is determined. The extraction unit extracts an actual roundness of the second inspection object from the 2D image of the second area,
The inspection unit estimates roundness of the second test object based on an amount of deviation of the second test object in the XY plane between the first end and the second end, and estimates the roundness The quality of the said 2nd to-be-tested object is determined by comparing with the actual roundness of the said 2nd to-be-tested object extracted by the said extraction part, It is characterized by the above-mentioned. X-ray examination processor. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
The extraction unit extracts the XY position, the Z direction height, and the volume information of each of the plurality of first inspection objects from the 3D image of the first area, and the second information from the 2D image of the second area. Extract the XY position and area of the inspection object,
The 3D information estimation unit interpolates or extrapolates Z-direction height and volume information of the plurality of first inspection objects to obtain Z-direction height and volume information at the XY position of the second inspection object Calculate
The inspection unit calculates the second inspection object calculated from the area of the second inspection object extracted by the extraction unit and the Z-direction height of the second inspection object estimated by the 3D information estimation unit. 4. The quality of the second test object is determined by comparing the volume information of the second test object with the volume information of the second test object estimated by the 3D information estimation unit. The processing apparatus of the X-ray examination of any one of 11. The X-ray examination according to any one of claims 1 to 12, wherein the object is an electronic component, and the inspection object is a solder for joining the electronic component and a substrate. Processing unit. An X-ray source for irradiating the object with X-rays;
An X-ray detector which detects X-rays transmitted through the object and outputs an X-ray transmission image;
The processing apparatus for X-ray examination according to any one of claims 1 to 13, wherein X-ray examination is performed using an X-ray transmission image obtained by the X-ray detector.
An X-ray inspection apparatus characterized by having: An X-ray inspection method for inspecting an object including a plurality of inspection objects by X-rays, comprising:
Acquiring a 3D image based on a plurality of X-ray transmission images having different projection directions for a first area which is a partial area of an inspection area set in the object;
Acquiring a 2D image, which is an X-ray transmission image, of a second area that is at least partially different from the first area of the inspection area;
The 3D information of the first object to be inspected in the first area is extracted from the acquired 3D image of the first area, and the 2D image of the second area obtained is An extraction step of extracting 2D information of a second test object which is a test object different from the first test object among test objects in two regions;
Estimating 3D information of the second inspection object based on the extracted 3D information of the first inspection object;
An inspection step of inspecting the second inspection object using the extracted 2D information of the second inspection object and the estimated 3D information of the second inspection object;
X-ray examination method characterized by including. A plurality of first inspection objects are included in the first region,
The X direction according to claim 15, wherein, in the estimation step, 3D information of the second inspection object is calculated by interpolating or extrapolating 3D information of the plurality of first inspection objects. Line inspection method. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
In the extraction step, the XY position and the Z direction height of each of the plurality of first inspection objects are extracted from the 3D image of the first region, and the second inspection object is extracted from the 2D image of the second region. Extract the XY position,
In the estimation step, the Z-direction height at the XY position of the second inspection object is calculated by interpolating or extrapolating the Z-direction heights of the plurality of first inspection objects. The X-ray inspection method according to claim 16. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
In the extraction step, the XY position and volume information of each of the plurality of first inspection objects are extracted from the 3D image of the first region, and the XY position of the second inspection object from the 2D image of the second region To extract
The volume information at the XY position of the second inspection object is calculated in the estimation step by interpolating or extrapolating the volume information of the plurality of first inspection objects. The X-ray inspection method as described in 17. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
In the extraction step, the XY position of each of the plurality of first inspection objects is extracted from the 3D image of the first region,
In the estimation step, the XY position of the second inspection object is calculated by interpolating or extrapolating the XY positions of the plurality of first inspection objects. The X-ray inspection method according to any one of 18. The XYZ coordinate system is placed so that the 2D image is an X-ray transmission image projected onto the XY plane, and both ends of the inspection object in the Z direction are referred to as a first end and a second end,
In the extraction step, the XYZ positions of the first end and the XYZ positions of the second end of each of the plurality of first inspection objects are extracted from the 3D image of the first area,
In the estimation step, an XYZ position at which the first end of the second inspection object should be calculated is calculated by interpolating or extrapolating the XYZ positions of the first ends of the plurality of first inspection objects, and The XYZ position where the second end of the second inspection object should be calculated is calculated by interpolating or extrapolating the XYZ positions of the second end of the plurality of first inspection objects. The X-ray inspection method according to any one of 16 to 18. In the estimation step, an XY plane of the first end and the second end of the second inspection object from the estimated XYZ position where the first end should exist and the XYZ position where the second end should exist. 21. The X-ray examination method according to claim 20, wherein the amount of deviation is calculated. In the estimation step, the XY position where the second inspection object should exist is calculated by averaging the XYZ position where the first end should exist and the XYZ position where the second end should exist. 22. The X-ray examination method according to claim 20 or 21, characterized in that: In the extraction step, the actual XY position of the second inspection object is extracted from the 2D image of the second area,
In the inspection step, the actual XY position of the second inspection object extracted in the extraction step is compared with the XY position where the second inspection object should be present estimated in the estimation step. The X-ray inspection method according to claim 19, wherein the quality of the second inspection object is determined. In the extraction step, the distance between the second inspection object and the adjacent inspection object adjacent to the second inspection object is extracted from the 2D image of the second area,
In the inspection step, based on the distance extracted in the extraction step and the deviation amount in the XY plane of the first end and the second end of the second inspection object estimated in the estimation step, 22. The X-ray inspection method according to claim 21, wherein the quality of the second inspection object is determined. In the extraction step, the actual roundness of the second inspection object is extracted from the 2D image of the second area,
In the inspection step, the roundness of the second test object is estimated based on the amount of deviation of the second test object in the XY plane between the first end and the second end, and the circularity is estimated The quality of the second object to be inspected is determined by comparing the actual roundness of the second object to be inspected extracted in the extraction step with the above-mentioned extraction step. X-ray inspection method. When the XYZ coordinate system is set such that the 2D image is an X-ray transmission image projected onto the XY plane,
In the extraction step, the XY position, the Z direction height, and the volume information of each of the plurality of first inspection objects are extracted from the 3D image of the first area, and the second information is extracted from the 2D image of the second area. Extract the XY position and area of the inspection object,
In the estimation step, Z-direction height and volume information at the XY position of the second inspection object is calculated by interpolating or extrapolating Z-direction height and volume information of the plurality of first inspection objects. And
In the inspection step, the volume of the second inspection object calculated from the area of the second inspection object extracted in the extraction step and the height in the Z direction of the second inspection object estimated in the estimation step The quality of the said 2nd to-be-tested object is determined by comparing information and the volume information of the said 2nd to-be-tested object estimated by the said estimation step, The any one of Claims 16-25 characterized by the above-mentioned. The X-ray inspection method according to item 1 or 2. The X-ray inspection method according to any one of claims 15 to 26, wherein the object is an electronic component, and the inspection object is a solder for joining the electronic component and a substrate. .   A program for causing a computer to execute each step of the X-ray examination method according to any one of claims 15 to 27.

Images