JP2009145072A - Measuring method and inspecting method, and measuring device and inspecting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for accurately measuring the position of a mirror-finished spherical body even if its surface is made rough. <P>SOLUTION: An intersection position of reflection light of two line illuminations 21, 22, which are coaxial with a vertical camera 11 and orthogonally intersect with each other, is found, and an intersection position of reflection light of two line illuminations 21, 23, which are coaxial with a tilted camera 12 and orthogonally intersect with each other, is found so as to acquire a three-dimensional center coordinate of a solder ball on the basis of the parallax. An edge section of the solder ball is illuminated by a profile illumination 24, and a radius is found as a distance from the center position to the edge section from an image obtained by the tilted camera 12. Then, the height of the solder ball is found by adding the height of the center to the radius of the solder ball. The X (Y) coordinates of the intersection of the reflection light are preferably found as X (Y) coordinates where luminance value found by integration in the Y (X) direction is the largest. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a technique for measuring the center position and height of a specular sphere.

With the progress of semiconductor devices, there is a strong demand for smaller and lighter packages. In order to satisfy such a requirement, a BGA (Ball Grid Array) package in which terminals of solder balls (solder bumps) are arranged in a grid pattern on the back surface of the package has become the mainstream of IC packages. FIG. 24 shows the appearance of the BGA.

  In order to improve the yield (number of non-defective products / number of inputs) in mounting using these packages, it is important that there are no missing bumps and there is no variation in bump height. If these are not satisfied, some of the terminals will not be bonded to the land of the substrate, resulting in a mounting failure.

  Conventionally, a method using a stereo camera method is known for inspecting variation in bump height (Patent Documents 1 to 3). In the stereo camera method, spot light is irradiated onto the solder bumps, the reflected light from the solder bump vertices is photographed with multiple cameras, their positions are extracted, and the coordinates of the bump vertices are obtained from the difference in their coordinate positions. Yes.

As a technique for inspecting the bump height, there are methods such as a confocal method and a light cutting method in addition to the stereo camera method.
Japanese Patent Laid-Open No. 9-304030 Japanese Patent Laid-Open No. 10-239025 Japanese Patent Laid-Open No. 11-287628

  By the way, in recent years, in consideration of environmental problems, lead-free solder is used for solder bumps. Lead-free solder has a complicated surface state and has a rough surface or a smooth surface, and its characteristics are not stable. When the surface of the solder bump is smooth, only light from the bump apex can be received as regular reflection when spot light is irradiated, and the apex position of the bump can be obtained with high accuracy (FIG. 25 ( a)). However, when the surface of the solder bump is rough, the illumination is scattered and it is difficult to accurately obtain the apex position of the bump (see FIG. 25B).

  As described above, when the surface of the solder bump is rough (there is unevenness), an error occurs when measuring the vertex position of the solder bump, and the height of the solder bump measured as a result also includes an error. Problem arises.

  The present invention has been made in view of the above circumstances, and the object of the present invention is not only a smooth sphere with a surface but also a technique that can accurately measure the position of a rough mirror sphere. Is to provide.

  In order to achieve the above object, in the present invention, the position of the specular sphere is measured by the following means or processing.

  That is, in the measuring method according to the present invention, for each of the plurality of imaging means, the mirror sphere is irradiated with two line illuminations that are coaxial with the imaging means and are not parallel to each other. A step of obtaining an intersection of the two line illuminations, and a step of obtaining a center position of the specular sphere from the positions of the intersections obtained for a plurality of imaging means in the step.

  Line illumination refers to light that looks like a line, but in the present invention, it may not be a line, and may be a plurality of point light sources arranged in a straight line, for example.

  In this way, the center of the specular sphere is obtained from the images captured by the plurality of imaging means, and the three-dimensional coordinates of the center position of the specular sphere are obtained based on the parallax. Since the center is obtained as an intersection of two line illuminations, the influence of surface roughness can be absorbed. Therefore, even if the surface of the mirror sphere is rough, the center position can be obtained with high accuracy. The sphere need not be a complete sphere, and may be, for example, a hemisphere.

  The measurement method according to the present invention includes a step of obtaining a center position of a specular sphere, a step of irradiating a contour detection illumination around the specular sphere, and two line illuminations that are coaxial with the imaging means and not parallel to each other. Irradiating the specular sphere; and determining the radius of the specular sphere from the position of the contour detection illumination and the position of the intersection of the two line illumination lines in the image captured by the imaging means; And obtaining the height of the specular sphere from the center and radius of the specular sphere.

  The height (vertex position) of the specular sphere can be obtained from the center position and radius of the specular sphere. Therefore, the measuring method according to the present invention obtains the height of the specular sphere by obtaining the radius in addition to the center position. Here, the measurement of the radius is performed by irradiating the periphery (edge) of the specular sphere with illumination for contour detection so that the edge of the specular sphere can be detected. The distance between the detected edge of the specular sphere and the intersection of line illumination is the radius of the specular sphere. According to this method, even if the surface of the specular sphere is rough, the height of the specular sphere can be obtained with high accuracy.

  In the measurement method according to the present invention, the step of obtaining the intersection of the two line illuminations is performed by integrating the luminance value in a direction perpendicular to the line direction of one of the line illuminations in the image captured by the imaging unit, and the integrated value is maximum. The luminance value is integrated in the direction perpendicular to the line direction of the other line illumination in the image obtained by the imaging means in the step of obtaining the position in the line direction of the one line illumination, and the integrated value is maximized. It is preferable to comprise the step of obtaining the position of the other line illumination in the line direction.

  In this way, the influence of the surface roughness can be absorbed by integrating the luminance values. In addition, since the integrated value of the line sensor can be calculated for each scan, processing can be performed at high speed.

  Here, it is preferable that the two line illuminations are arranged at an angle so that reflected light from other than the specular sphere is not received by the imaging means.

  When the imaging unit captures a specular sphere from the vertical direction, the light emitted from the line illumination may be reflected by an object other than the specular sphere (for example, a substrate) and received by the imaging unit. . If this occurs, it may adversely affect the accuracy of derivation of the center by the imaging means, so it is desired to prevent this occurrence. Therefore, as described above, it is preferable that the two line illuminations are angled so that regular reflection from an object other than the specular sphere is not received by the imaging means.

  Moreover, it is preferable that the same line illumination is coaxial with respect to a plurality of imaging means.

  Each imaging means obtains the center coordinates from the intersection of two line illuminations. The line illumination needs to be coaxial with the imaging means, but the number of line illuminations can be reduced by making one line illumination coaxial with the plurality of imaging means. When there are two imaging means, the above method can be performed with only three line illuminations instead of four line illuminations. Thus, by reducing the number of line illuminations, it is possible to reduce the size and cost of the apparatus.

  Moreover, it is preferable that a certain imaging unit and coaxial name line illumination irradiate light having a wavelength different from that of line illumination coaxial with other imaging units.

  The reflected light from the line illumination for one imaging means may be received by the other imaging means. Therefore, by varying the wavelength (color) of the light used in the line illumination corresponding to each imaging means, each imaging means can measure only the light of the corresponding wavelength, and light from unnecessary line illumination. The influence by can be eliminated.

  The inspection method according to the present invention is an inspection method for inspecting the heights of a plurality of bumps arranged two-dimensionally in a package. The height of each bump is calculated by the measurement method described above, and the average is calculated from the average. It is characterized in that it is determined whether the package is a non-defective product or a defective product on the basis of the variation in the above.

  According to such an inspection method, the height of the bump can be measured with high accuracy even when the surface of the bump is rough, and therefore, a non-defective product / defective product can be determined with high accuracy.

  In addition, this invention can be grasped | ascertained as the measuring method or inspection method containing at least one part of the said process. Moreover, this invention can also be grasped | ascertained as a measuring apparatus or a test | inspection apparatus for performing at least one part of the said process. Each of the above means and processes can be combined with each other as much as possible to constitute the present invention.

  For example, a measuring apparatus according to an aspect of the present invention is a measuring apparatus that measures the position of a specular sphere, and is a plurality of imaging units, coaxial with the imaging units installed in each imaging unit, and mutually. Center position calculation means for obtaining the intersection of two line illumination lines and the center position of the specular sphere from the position of the intersection of the two line illuminations intersecting with each of the images captured by the plurality of imaging means. And.

  According to the present invention, it is possible to accurately measure the position of a mirror sphere having a rough surface.

  Exemplary embodiments of the present invention will be described in detail below with reference to the drawings.

<Measurement principle>
The purpose of this embodiment is to measure the center position and height of a BGA solder bump as shown in FIG. First, the measurement principle will be described.

  The solder bump to be measured is a spherical solder ball. Therefore, as shown in FIG. 1, the height can be obtained as (the height of the center position of the sphere) + (radius). In this embodiment, the center position and the radius of the sphere are obtained in order to measure the bump height using this.

(Measurement of center position)
When a spherical solder ball is photographed with a camera, as shown in FIG. The center position of the circle projected in the captured image is equal to the center position of the solder ball sphere viewed from the camera. Therefore, if the center position of a circle obtained by photographing the solder ball with two cameras is obtained, the center position (three-dimensional coordinates) of the sphere can be obtained from the parallax of the two center positions by the stereo camera method.

  Next, a method for obtaining the center from a circular image of the solder ball photographed by the camera will be described.

  As shown in FIG. 3, when a solder ball is photographed with a camera, light is irradiated to the solder ball with line illumination coaxial with the camera. Note that the fact that the camera and the line illumination are coaxial means that the imaging axis of the camera is included on the plane created by the light of the line illumination. As described above, when the solder ball is irradiated with the coaxial line illumination, the reflected light photographed by the camera becomes a straight line passing through the center of the circle image of the solder ball photographed as shown in FIG.

  FIG. 5 is a view when photographing is performed by irradiating light onto a solder ball using two coaxial line illuminations. The two coaxial line illuminations are orthogonal to each other. In this case, the reflected lights of the two line illuminations photographed by the camera intersect at right angles at the center of the solder ball circle as shown in FIG. However, as long as the reflected lights of the line illuminations intersect, it is not necessary that the two line illuminations themselves intersect.

  When the solder ball is irradiated with light by line illumination that is not coaxial with the camera as shown in FIG. 7, the reflected light to be photographed becomes a curve that does not pass through the center of the circle as shown in FIG.

  Next, a process for obtaining the intersection of reflected light of two line illuminations (that is, the center of the circle of the photographed solder ball) from the photographed image as shown in FIG. 6 will be described with reference to FIG. Here, as shown in FIG. 9, the direction of one line illumination is the X direction, and the direction of the other line illumination is the Y direction. First, in order to obtain the X coordinate of the intersection position, a value obtained by integrating the luminance value in the Y direction is obtained for each X coordinate. The X coordinate at which the accumulated luminance value in the Y direction is the largest is taken as the X coordinate of the intersection. Similarly, in order to obtain the Y coordinate of the intersection position, a value obtained by integrating the luminance value in the X direction is obtained for each Y coordinate. The Y coordinate at which the accumulated luminance value in the X direction is the largest is taken as the Y coordinate of the intersection. In this way, the intersection position can be obtained.

  In FIG. 6 and FIG. 9, the reflected light of the line illumination is drawn as a clean straight line passing through the center. However, when the surface of the solder ball is rough, the illumination is scattered and may not be a clean straight line. However, in the present embodiment, as described above, the position where the value obtained by integrating the luminance values in the X and Y directions is maximized is obtained as the intersection position, and thus the surface of the solder ball is rough. However, the position of the intersection can be obtained with high accuracy without being greatly affected by scattering.

  Here, the method of obtaining the center of the photographed solder ball image for one camera has been described. However, the center of the solder ball image is similarly obtained for the other camera. Then, the three-dimensional position of the solder ball center is obtained from the parallax in the two images. Since the calculation of the three-dimensional position by the stereo camera method is a known technique, the description thereof is omitted here.

(Radius measurement)
Next, how to determine the radius of the solder ball will be described.

In order to obtain the radius, as shown in FIG. 10, light is applied to the solder ball from a direction substantially perpendicular to the shooting axis of the camera so that illumination is applied to the edge of the circle of the solder ball shot by the camera. When the camera is shooting the solder ball from diagonally above, low-angle illumination that illuminates from substantially the horizontal direction is used. This low-angle illumination may be anything as long as it is installed so that the reflected light is photographed from the edge (or a part thereof) of the solder ball image photographed by the camera.

  Although omitted in the figure, line illumination coaxial with the camera is also used to obtain the center of the solder ball.

  FIG. 11 shows a photographed image when the light is applied to the solder ball using the low-angle illumination and the photograph is taken with the camera. As shown in the figure, the reflected light from the low angle illumination hits the edge of the circular image of the solder ball in the X direction. The X coordinate position of the edge is obtained from the luminance value integrated in the Y direction.

  Since the X coordinate of the center of the solder ball can be obtained as described above, the radius of the solder ball can be obtained from the difference in position between the center and the edge.

(Derivation of bump height)
As described above, when the three-dimensional position of the center of the solder ball and the radius of the solder ball are obtained, by adding the radius to the height of the center of the solder ball, the height of the solder ball (bump height), that is, The height of the top of the solder ball can be obtained.

  According to this method, the height from the reference to the vertex can be calculated not only for a perfect sphere but also for a part of the sphere.

<Example 1>
FIG. 12 is a schematic configuration diagram of the optical system 20 and the imaging system 10 of the bump inspection apparatus. The optical system 20 and the imaging system 10 are disposed above a conveyance line that conveys the BGA package along the horizontal direction. The BGA package has an array with a plurality of solder bumps arranged in the vertical and horizontal directions on the back surface thereof (see FIG. 24).

  The optical system 20 has the following configuration.

  A vertical camera 11 for photographing from the vertical direction with respect to the BGA solder bumps is arranged. In addition, two line illuminations 21 and 22 for irradiating light from a direction perpendicular to the solder bumps are arranged. The line illumination 21 is a linear light source parallel to the transport direction A and extending along the horizontal direction. The line illumination 22 is a linear light source perpendicular to the transport direction A and extending along the horizontal direction. These line lights 21 and 22 are coaxial with the vertical camera 11.

  An oblique camera 12 is further installed that photographs the BGA solder bumps obliquely from the upstream side in the transport direction A. Further, a line illumination 23 that irradiates light to the solder bumps obliquely is provided coaxially with the oblique camera 12. The line illumination 23 is a linear light source perpendicular to the transport direction A and extending along the horizontal direction. The oblique camera 12 is installed so that its photographing axis is included in the light irradiation surface of the line illumination 21. That is, the oblique camera 12 and the line illumination 21 are coaxial.

  On the downstream side of the BGA solder bump in the conveyance direction A, an outline illumination 24 for irradiating light from a substantially horizontal direction is provided. In this embodiment, a linear light source is used as the contour illumination 24.

Here, the color (wavelength) of light emitted by each illumination will be described. The line illumination 22, the line illumination 23, and the contour illumination 24 irradiate light of different colors. In this embodiment, the line illumination 22 emits light of green (G), the line illumination 23 emits red (R), and the contour illumination 24 emits light of blue (B). Note that these light colors are examples, and any different colors can be adopted as long as they can be separated from each other. Moreover, although the color of the line illumination 21 may adopt a color different from other illuminations, the same color light as the line illumination 22 or the line illumination 23 orthogonal to the line illumination 21 can be adopted. In the present embodiment, the color of the line illumination 21 employs the same green (G) light as that of the line illumination 22.

  In this embodiment in which the illumination and the camera are arranged in this way, an image of a bump photographed by each camera is shown in FIG. In the image P1 photographed by the vertical camera 11, the reflected lights L21 and L22 by the line illuminations 21 and 22 which are line illuminations coaxial with the camera appear as straight lines orthogonal to the center of the photographed bump circle. The reflected light L23 from the line illumination 23 that is not coaxial with the vertical camera 11 is a curve that does not pass through the center of the circle. Further, in the image P2 photographed by the oblique camera 12, the reflected lights L21 and L23 by the line illuminations 21 and 23 which are line illuminations coaxial with the camera are straight lines orthogonal to each other at the center of the photographed bump circle. Reflect. The reflected light L22 from the line illumination 22 that is not coaxial with the oblique camera 12 is a curve that does not pass through the center of the circle. The reflected light L24 from the contour illumination 24 illuminates the end of the imaged bump circle on the conveyance direction A side.

  FIG. 14 is a block diagram showing an overall schematic configuration of the bump inspection apparatus. The bump inspection apparatus generally includes an imaging system 10, an optical system 20, an analysis / drive control system 30, and a transport system 40. The analysis / drive control system 30 includes a controller 31, a capture board 32, and a measurement / inspection unit 33.

  The controller 31 controls the servo motor 41 of the transport system 40 to drive the transport stage 42 and transport the BGA package to the measurement position. Further, when there is a bump abnormality, a discharge mechanism (not shown) for discharging the abnormal package from the transport stage 42 is driven.

  The capture board 32 acquires image data from the vertical camera 11 and the oblique camera 12 and stores them in the memory.

  The measurement / inspection unit 33 acquires image data, measures the center position and height of the bump, and performs a determination process as to whether or not there is an abnormality in the arrangement position or height of the bump. Note that the measurement / inspection unit 33 includes various threshold values serving as determination criteria in addition to information on height / size / pitch and distribution serving as a reference for the BGA package to be inspected in advance.

  The analysis / drive control system 30 is configured by a computer (information processing apparatus), and the function is realized by the CPU executing a program stored in the memory.

Next, the flow of processing performed by the measurement inspection unit 33 will be described with reference to the flowchart of FIG. First, the measurement / inspection unit 33 calculates the three-dimensional position of the center of the solder ball from the parallax of the center position between the vertical camera 11 and the oblique camera 12 (S10). Then, it is determined whether the center position is within the tolerance (S11). When the deviation of the center position is larger than the tolerance (S11-NO), it is determined that the position deviation is defective (S17). If the deviation of the center position is within the tolerance (S11-YES), the ball radius is calculated from the image of the oblique camera 12 (S12). Then, it is determined whether the deviation of the ball radius is within the tolerance (S13). When the deviation of the ball radius is larger than the tolerance (S13-NO), it is determined that the radius is defective (S18). When the deviation of the ball radius is within the tolerance (S13-YES), the height of the solder ball (the coordinates of the vertex) is calculated as a value obtained by adding the ball radius to the height of the center position of the ball (S14). . Then, it is determined whether the flatness of the ball height is within the tolerance (S15). If the deviation in flatness is larger than the tolerance, it is determined that the height is defective (S19). When the flatness of the ball height is within the tolerance (S15-YES), it is determined to be a non-defective product because it is normal in all inspections (S16).

  Next, the ball center position calculation process in S10 will be described with reference to FIGS. FIG. 16 is a flowchart showing the flow of the ball center position calculation process, and FIG. 17 conceptually illustrates the ball center position calculation process.

  In the following description, as shown in FIG. 17, in the image taken by the vertical camera 11 and the oblique camera 12, the direction parallel to the line illumination 21 (also parallel to the BGA transport direction) is defined as the X direction. The direction orthogonal to the X direction (also parallel to the line illuminations 22 and 23) is defined as the Y direction.

  In the ball center position calculation process, first, at each X coordinate, the luminance value of the color of the coaxial line illumination is integrated in the Y direction (S10-1). As shown in FIG. 17A, in the image P <b> 1 of the vertical camera 11, a green luminance value that is the color of the line illumination 22 parallel to the Y direction of the vertical camera 11 is obtained. As shown in FIG. 17B, in the image P2 of the oblique camera 12, a red luminance value that is the color of the line illumination 23 parallel to the Y direction of the oblique camera 12 is obtained. Then, the X coordinate having the maximum integrated value is set as the X coordinate of the center position (S10-2).

  Next, in each Y coordinate, the luminance value of the color of the coaxial line illumination is integrated in the X direction (S10-3). As shown in FIGS. 17A and 17B, the luminance value of green, which is the color of the line illumination 21 parallel to the X direction, is obtained in any of the images P1 and P2 of the vertical camera 11 and the oblique camera 12. Then, the Y coordinate having the maximum integrated value is set as the Y coordinate of the center position (S10-2).

  Finally, the three-dimensional position of the solder ball center is obtained from the parallax of the center of the solder ball photographed as a circle in the image P1 of the vertical camera 11 and the image P2 of the oblique camera 12 (S10-5).

  Note that when green is integrated in the Y direction in the image P1 of the vertical camera 11, the reflected light L21 from the line illumination 21 is also integrated. However, since the contribution by the reflected light L21 hardly depends on the X coordinate, the maximum integrated value is maximum. This does not affect the process for obtaining. Similarly, when integrating in the X direction in the image P1 of the vertical camera 11, the reflected light L22 from the line illumination 22 is also integrated. In addition, when green is integrated in the X direction in the image P2 of the oblique camera 12, the reflected light L22 from the line illumination 22 is also integrated. Similarly, since the contribution amount hardly depends on the coordinates, the processing for obtaining the maximum integrated value is not affected. However, if different colors of light are used for all lighting, this effect can be completely eliminated.

  Next, the ball radius calculation process in S12 will be described with reference to FIGS. FIG. 18 is a flowchart showing the flow of the ball radius calculation process, and FIG. 19 is a diagram conceptually illustrating the ball radius calculation process.

  In the ball radius calculation process, an edge calculation process in the X direction is performed on the image P2 of the oblique camera 12 (S12-1). Here, a second-order differential filter such as a Laplacian filter is used as the edge calculation process. The accumulated luminance value in the X direction and the graph of the edge intensity are as shown in FIGS. 19 (a) and 19 (b), respectively.

Next, a zero cross point is searched from the ball center point toward the positive direction of the X axis (S12-2). The position where the reflected light L24 of the contour illumination 24 disappears due to the search for the zero cross point,
That is, the position of the end of the solder ball to be photographed is obtained.

  Then, the X coordinate of the zero cross point is subtracted from the X coordinate of the center position (S12-3), and the radius of the ball can be obtained from the obtained difference in the number of pixels and the resolution of the camera (S12-4).

<Example 2>
When the solder bumps are photographed using the vertical camera 11 by irradiating light from the vertical directions 21 and 22 as in the first embodiment, light reflected from other than the bumps such as the substrate is photographed by the vertical camera 11. May end up. If the reflected light from other than the solder bumps is photographed, the position of the reflected light may be detected as the center position of the solder bump, which adversely affects the accuracy of derivation of the center.

  Therefore, in this embodiment, the arrangement of the line illumination is devised so as not to receive the reflected light from other than the solder bumps. The two line illuminations 21 and 22 that emit light from the vertical direction are angled so as not to be horizontal. For example, as shown in FIG. 20, the line illumination 21 is formed in a square shape by providing different angles on one side 21 a and the other side 21 b of the vertical camera 11. Similarly, the line illumination 22 is also provided with an angle so as not to be horizontal.

  By doing so, the light that is irradiated from the line illumination and regularly reflected from other than the solder bumps is not received by the vertical camera 11. Therefore, the erroneous detection as described above is eliminated, and the accuracy of center derivation is not adversely affected.

<Example 3>
In the third embodiment, as in the second embodiment, in order to eliminate the influence of the reflected light from other than the solder bumps, the arrangement of the illumination and the camera is devised. 21 and 22 show the arrangement of the optical system of the bump inspection apparatus in this embodiment. FIG. 21 is a perspective view of the optical system, and FIG. 22 is a three-view diagram of the optical system.

  First, cameras for photographing solder bumps are two oblique cameras 311 and 312 each having an angle of 45 ° from the vertical direction. Each camera is provided with two line illuminations coaxial with the imaging axis of the camera. Of the two line illuminations coaxial with the oblique camera 311, the line illumination 321 is parallel to the transport direction A, and the line illumination 322 is orthogonal to the transport direction A and is horizontal. Of the two line illuminations coaxial with the oblique camera 312, the line illumination 323 is parallel to the transport direction A, and the line illumination 324 is orthogonal to the transport direction A and is horizontal. In addition, a ring-shaped illumination 325 that irradiates light in the circumferential direction from substantially the same height as the solder bump is provided.

  The line illuminations 321 and 322 emit green light, the line illuminations 323 and 324 emit blue light, and the ring illumination 325 emits blue light. Thereby, by separating only the light of interest and using it for measurement, it is possible to prevent erroneous measurement due to unnecessary reflected light of illumination.

<Modification 1>
In the above embodiment, the color (wavelength) of light emitted from each illumination is changed. However, the timing at which each illumination emits light may be changed. Even in this way, it is possible to detect only the reflected light of the illumination necessary for the measurement.

<Modification 2>
A method of obtaining the luminance integrated value in the X (Y) direction as described above, which is determined by obtaining the intersection of the reflected lights of the line illumination perpendicular to the center of the solder ball photographed as a circle by the camera You may obtain | require an intersection by methods other than.

  For example, as shown in FIG. 23A, the intersection may be obtained as the center of gravity of the area where the reflected light is detected.

  Further, as shown in FIG. 23B, for each Y coordinate, the midpoint of the reflected light is obtained for one line in the X direction. This can be done by extracting the edge by quadratic differentiation and determining the midpoint of the two zero cross points. After obtaining the midpoint for each Y coordinate, this is linearly approximated to obtain the Y-direction center line of the reflected light. Similarly, obtain the X-direction center line of the reflected light. Then, the center of the reflected light can be obtained as the intersection of the X-direction center line and the Y-direction center line.

  Although the center line is obtained by linear approximation here, the average of the X (Y) coordinates of the midpoints for each Y (X) may be obtained and used as the center X (Y) coordinates.

<Modification 3>
In the above description, a solder bump is described as an example of an object to be measured. However, it is apparent that the measurement method and the inspection method of the present invention can be used for mirror spheres other than the solder bump.

It is a figure explaining that the height of a solder ball is calculated | required from the height and radius of the center position. It is a figure explaining the image when a solder ball is photoed with a camera. It is a figure at the time of imaging | photography by irradiating light to a solder ball by coaxial line illumination. It is a figure which shows the example of an image when a solder ball is image | photographed using coaxial line illumination. It is a figure at the time of image | photographing by irradiating light to a solder ball by two coaxial line illuminations orthogonal to each other. It is a figure which shows the example of an image when a solder ball is image | photographed using two coaxial line illuminations orthogonal to each other. It is a figure at the time of imaging | photography by irradiating light to a solder ball by non-coaxial line illumination. It is a figure which shows the example of an image when a solder ball is image | photographed using non-coaxial line illumination. It is a figure explaining the process which calculates | requires the intersection position (namely, center of the image | photographed solder ball circle) of the reflected light of two coaxial line illuminations. It is a figure at the time of imaging | photography by irradiating light to the periphery of a solder ball by low angle illumination. It is a figure which shows the example of an image when a solder ball is image | photographed using low angle illumination. 1 is a diagram illustrating a schematic configuration of an optical system of a bump inspection apparatus according to Embodiment 1. FIG. It is a figure which shows the image of the bump image | photographed with the bump inspection apparatus which concerns on Example 1. FIG. 1 is a block diagram illustrating a schematic configuration of an entire bump inspection apparatus according to a first embodiment. 6 is a flowchart illustrating a flow of measurement inspection processing in the bump inspection apparatus according to the first embodiment. It is a flowchart which shows the flow of a ball center position calculation process. It is a figure which illustrates a ball center position calculation process notionally. It is a flowchart which shows the flow of a ball radius calculation process. It is a figure which illustrates ball radius calculation processing notionally. FIG. 6 is a diagram illustrating a configuration of an optical system of a bump inspection apparatus according to a second embodiment. It is a figure (perspective view) which shows the structure of the optical system of the bump inspection apparatus which concerns on Example 3. FIG. FIG. 10 is a diagram (three views) illustrating a configuration of an optical system of a bump inspection apparatus according to a third embodiment. It is a figure explaining the modification of the process which calculates | requires the intersection of reflected light. It is a figure explaining BGA. It is a figure which shows the state of the surface of a solder bump (solder ball).

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Vertical camera 12 Oblique camera 21, 22, 23 Line illumination 24 Contour illumination 31 Controller 32 Capture board 33 Measurement inspection part 41 Conveyance stage 42 Servo motor

Claims (16)

A measurement method for measuring the position of a specular sphere from images by a plurality of imaging means,
For each imaging means, the mirror sphere is irradiated with two line illuminations that are coaxial with the imaging means and are not parallel to each other to perform imaging, and in the captured image, the intersection of the straight lines of the two line illuminations is obtained. The desired process;
Obtaining the center position of the specular sphere from the position of the intersection obtained for the plurality of imaging means in the step;
Measuring method including Obtaining the center position of the specular sphere;
Irradiating contour detection illumination around the specular sphere;
Irradiating the specular sphere with two line illuminations coaxial with the imaging means and not parallel to each other;
A step of obtaining a radius of the specular sphere from the position of the contour detection illumination and the position of the intersection of the straight lines of the two line illuminations in the image captured by the imaging means;
Obtaining the height of the specular sphere from the center and radius of the specular sphere;
Measuring method including   The measurement method according to claim 2, wherein the step of obtaining the center position of the specular sphere is performed by the measurement method according to claim 1. The process of finding the intersection of two line lighting lines is:
In the image captured by the imaging means, integrating the luminance value in a direction perpendicular to the line direction of the one line illumination, obtaining a position in the line direction of the one line illumination that maximizes the integrated value;
In the image captured by the imaging means, integrating a luminance value in a direction perpendicular to the line direction of the other line illumination, obtaining a position in the line direction of the other line illumination that maximizes the integrated value;
The measuring method according to any one of claims 1 to 3, wherein The measurement method according to claim 1, wherein the line illumination is arranged at an angle so that reflected light from other than the specular sphere is not received by the imaging unit. The measuring method according to claim 1, wherein there is line illumination coaxial with the plurality of imaging means. The measuring method according to claim 1, wherein the line illumination coaxial with a certain imaging unit irradiates light having a wavelength different from that of the line illumination coaxial with another imaging unit. An inspection method for inspecting the height of a plurality of bumps arranged two-dimensionally in a package,
The height of each bump is calculated by the method according to claim 2,
An inspection method characterized by determining whether a package is a non-defective product or a defective product based on variation from the average. A measuring device for measuring the position of a specular sphere,
A plurality of imaging means;
Two line illuminations installed in each imaging means, coaxial with the imaging means and not parallel to each other;
In each image picked up by the plurality of image pickup means, a center position calculation means for obtaining an intersection of two line illumination lines and obtaining a center position of the specular sphere from the position of the intersection;
A measuring apparatus comprising: Center position measuring means for measuring the center position of the specular sphere,
Imaging means;
Illumination for contour detection that irradiates light around the specular sphere,
Two line illuminations coaxial with the imaging means and not parallel to each other;
Radius calculation means for obtaining a radius of the specular sphere from the position of the contour detection illumination and the position of the intersection of the two line illuminations in the image captured by the imaging means;
Height calculating means for calculating the height of the specular sphere from the center position and radius of the specular sphere,
A measuring apparatus comprising:   The measuring apparatus according to claim 10, wherein the center position measuring unit is the measuring apparatus according to claim 9. The calculation of the intersection of the two line illumination lines is
In the image captured by the imaging means, the luminance value is integrated in the direction perpendicular to the line direction of one line illumination, and the position in the line direction of the one line illumination that maximizes the integrated value is obtained.
In the image captured by the imaging means, the luminance value is integrated in a direction perpendicular to the line direction of the other line illumination, and the position in the line direction of the other line illumination that maximizes the integrated value is obtained.
It is performed by this, The measuring apparatus in any one of Claims 9-11 characterized by the above-mentioned. The measuring apparatus according to claim 9, wherein the line illumination is arranged at an angle so that reflected light from other than the specular sphere is not received by the imaging unit. The measuring apparatus according to claim 9, wherein there is line illumination coaxial with the plurality of imaging units. The measuring apparatus according to claim 9, wherein the line illumination coaxial with a certain imaging unit irradiates light having a wavelength different from that of the line illumination coaxial with another imaging unit. An inspection apparatus for inspecting the height of a plurality of bumps arranged two-dimensionally in a package,
A measuring device according to any of claims 10 to 15,
A quality determination means for determining an average of the height of each bump measured by the measuring device, and determining whether the bump is a non-defective product or a defective product based on a variation from the average;
An inspection apparatus comprising:

Images