JP2007163266A - Inspection device and inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect accurately the vertex of a projection part even when the vertex part of the projection part is not formed flatly but the vertex side caves in. <P>SOLUTION: This inspection device 100 is provided with the first camera 102 for imaging a ball 204 from the vertical direction with respect to the plane of a substrate 200, the second camera 104 for imaging the ball 204 from the oblique direction with respect to the plane of the substrate 200, an edge detection part 128 for detecting the edge part of a caving-in part of the ball 204 in the first image imaged by the first camera 102 and in the second image imaged by the second camera 104, and a vertex height calculation part 132 for determining the height to the plane of the edge part by comparing the position of the edge part in the first image with the position of the edge part in the second image when the edge part of the caving-in part is detected in the first image and in the second image by the edge detection part 128, and calculating the height of the highest part at the edge part as the height of the vertex of the ball 204. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an inspection apparatus and an inspection method for obtaining the height of a vertex of a substantially spherical protrusion formed on a plane of an inspection object.

  In recent years, with the development of IC ball devices, it is necessary to provide a low-cost appearance inspection function. Since it is difficult to increase the accuracy of the inspection technique for imaging the ball on the substrate three-dimensionally, the height of the vertex of the ball is generally required from a two-dimensional image.

  As this type of inspection apparatus, there is known an apparatus provided with illumination for irradiating a substrate and a camera for imaging each ball on the substrate. The illumination irradiates each ball with light in a substantially horizontal direction, and the camera images the illuminated ball installed vertically upward and obliquely from above and obliquely upward. Then, the vertex position is calculated from the shape of the donut-shaped bright spot in each captured image. However, since this inspection device is based on the premise that the ball is a perfect sphere, there is a problem that the actual vertex position does not match the calculated vertex position if the ball is not a perfect sphere. there were.

As an inspection apparatus in which this problem does not occur, an apparatus has been proposed in which illumination is installed vertically upward and obliquely upward and the reflected light at the apex portion is detected by a camera (see, for example, Patent Document 1). According to this inspection apparatus, the position of the bright spot at the apex portion of the ball by irradiation is calculated as the apex position.
Japanese Patent Laid-Open No. 10-239025

  By the way, the inspection method by the inspection apparatus described in Patent Document 1 is based on the premise that a sufficient amount of reflected light with respect to the irradiation light is secured at the apex of the ball. Therefore, there is a problem that when the apex portion of the ball is not formed in a planar shape, the apex of the ball cannot be accurately detected. When the contact on the substrate side occurs, the apex of the ball is easily depressed, and in this case, the accuracy of detecting the coplanarity of the entire BGA package is deteriorated.

  According to the present invention, there is provided an inspection apparatus for obtaining a height of a vertex of a substantially spherical protrusion formed on a plane of an inspection object, wherein the protrusion is from a direction perpendicular to the plane of the inspection object. A first imaging unit that captures the projection, a second imaging unit that captures the protrusion from a direction oblique to the plane of the inspection object, a first image captured by the first imaging unit, and the Edge detecting means for detecting a depressed edge portion of the protruding portion in the second image picked up by the second image pickup portion; a position of the edge portion in the first image detected by the edge detecting means; A vertex height calculating means for calculating the height of the vertex of the protruding portion based on the position of the edge portion in the second image detected by the edge detecting means; An apparatus is provided.

  In this inspection apparatus, the first imaging unit and the second imaging unit that capture an image of the protrusion can image the protrusion from different angles. And the edge part of the depression in the 1st image of a 1st imaging part and the 2nd image of a 2nd imaging part is detected by an edge detection means. When an edge portion is detected, there is a high probability that the apex side of the ball is depressed and the depressed edge portion is apex. In this case, the vertex height calculating means calculates the height of the vertex of the protruding portion based on the position of the edge portion in the first image and the position of the edge portion in the second image. When it is determined that the apex side of the projecting portion is not depressed, for example, a method of detecting the apex by detecting the reflected light of the irradiation light using the first imaging unit, the second imaging unit, etc. Detect the apex of the protrusion.

  Further, according to the present invention, there is provided an inspection method for obtaining a height of a vertex of a substantially spherical protrusion formed on a plane of the inspection object, wherein the height of the inspection object is perpendicular to the plane. A first image acquisition step of acquiring a first image by imaging the protrusion, and a second image acquisition of acquiring a second image by imaging the protrusion from a direction oblique to the plane of the inspection object And an edge detection step for detecting a depressed edge portion of the protruding portion in the first image acquired in the first image acquisition step and the second image acquired in the second image acquisition step; When the depressed edge portion is detected in the first image and the second image in the edge detection step, the position of the edge portion in the first image and the position of the edge portion in the second image Based on the height of the apex of the protrusion Inspection method characterized by including a vertex height calculation step of calculating is provided.

  As described above, according to the present invention, when the apex portion of the projecting portion is not formed in a planar shape and the apex side is depressed, the apex of the projecting portion can be accurately detected.

  A preferred embodiment of an inspection apparatus and an inspection method according to the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

(First embodiment)
1 to 4 show a first embodiment of the present invention. FIG. 1 is an explanatory diagram of a schematic configuration of an inspection apparatus, FIG. 2 is an explanatory diagram of a first image taken by a first camera, and FIG. Explanatory drawing of the 2nd image imaged with the 2nd camera, FIG. 4 is a flowchart which shows operation | movement at the time of implementing the inspection method of an inspection apparatus.

  As shown in FIG. 1, the inspection apparatus 100 obtains the apex of the ball 204 of the BGA 202 formed on the plane of the substrate 200. In the first embodiment, all the balls 204 are imaged with respect to the BGA 202 of the substrate 200 as the inspection object, the XYZ positions of all the balls 204 are calculated from the captured camera images, and the vertex positions of all the balls 204 are calculated. The coplanarity value of the entire package of the BGA 202 is calculated from the data. Here, the direction parallel to the upper surface of the substrate 200 is the XY direction, and the direction perpendicular to the upper surface is the Z direction.

  The inspection apparatus 100 includes a first camera 102 that images the ball 204 from a direction perpendicular to the plane of the substrate 200, and a second camera 104 that images the ball 204 from a direction oblique to the plane of the substrate 200. I have. The inspection apparatus 100 includes a first illumination 110 that irradiates light substantially perpendicular to the plane of the substrate 200, a second illumination 112 that irradiates light at a predetermined incident angle with respect to the plane of the substrate 200, and a substrate. And a third illumination 114 that irradiates light substantially horizontally with respect to the plane of 200.

  In the present embodiment, since the substrate 200 is fixed substantially horizontally, the BGA 202 appears on the upper surface of the substrate 200. The first camera 102 is disposed above the substrate 200 and points toward the substrate 200 vertically below. The second camera 104 is disposed obliquely above the substrate 200 and is directed toward the substrate 200 obliquely below. The second camera 104 is installed at a position where the elevation angle is approximately 45 degrees when viewed from the substrate 200 side.

  The first illumination 110 is disposed above the substrate 200 and irradiates light toward the substrate 200 vertically below. That is, the first camera 102 is arranged so that light reflected from the flat portion 206 (see FIG. 2) parallel to the upper surface of the substrate 200 in the ball 204 is incident on the light emitted from the first illumination 110. It will be.

  Further, the second illumination 112 is disposed obliquely above the substrate 200 on the side opposite to the second camera 104 and is directed to the substrate 200 obliquely below. The second illumination 112 is installed at a position where the elevation angle is approximately 45 degrees when viewed from the substrate 200 side. That is, the second camera 104 is arranged so that the reflected light of the flat portion 206 (see FIG. 3) of the ball 204 is incident among the light emitted from the second illumination 112.

  The third illumination 114 is configured in a ring shape surrounding the substrate 200. In the present embodiment, the third illumination 114 is installed at a position where the elevation angle is about 5 degrees when viewed from the substrate 200 side. By irradiating light from the third illumination 114, the outline of the entire ball 204 is raised in a top view.

  The inspection apparatus 100 also detects a position of the ball XY position detection unit 120 that detects the position of the ball 204 on the XY plane based on the first image A captured by the first camera 102 in a state where light is emitted from the third illumination 114. A first flat position detector 124 for detecting the position of the flat portion 206 on the ball 204 from the first image A of the first camera 102, and a flat portion 206 on the ball 204 from the second image B of the second camera 104. And a second flat position detecting unit 126 for detecting the position of.

  Furthermore, the inspection apparatus 100 includes the edge portion 210 (see FIGS. 2 and 3) of the depression 208 (see FIGS. 2 and 3) of the ball 204 in the first image A of the first camera 102 and the second image B of the second camera 104. 3), and the edge detection unit 128 detects the edge part 210 of the depression 208 in the first image A and the second image B, the edge part 210 of the first image A is detected. A ball apex calculation unit 132 that calculates the height of the apex of each ball 204 based on the position and the position of the edge 210 in the second image B; and the apex height of each ball 204 calculated by the ball apex calculation unit 132 In addition, a coplanarity calculation unit 134 that calculates the coplanarity with respect to the plane of each ball is provided.

  The ball vertex calculation unit 132 compares the position of the edge part 210 in the first image A and the position of the edge part 210 in the second image B to obtain the height of the edge part 210 with respect to the plane, and is the highest in the edge part 210 The height of the part is calculated as the height of the apex of the ball 204. In the present embodiment, the ball vertex calculation unit 132 calculates the difference between the position coordinates of the predetermined feature point CP and the position coordinates of the edge part 210 for each of the first image A and the second image B, and calculates the difference. Is compared with the position of the edge portion 210 in the first image A and the position of the edge portion 210 in the second image B. The ball vertex calculation unit 132 sets the predetermined feature point CP as the center of gravity of the reflection portion of the flat portion 206 in the state where the first illumination 110 is irradiated with light for the first image A, and for the second image B. The center of gravity of the reflection portion of the flat portion 206 in a state where light is emitted from the second illumination 112 is used. In addition, the ball vertex calculation unit 132 performs the first irradiation with light from the first illumination 110 when the edge detection unit 128 does not detect the edge part 210 of the depression 208 in the first image A and the second image B. The height of the flat portion 206 obtained by comparing the reflection portion of the flat portion 206 in one image A and the reflection portion of the flat portion 206 in the second image B in a state where light is irradiated from the second illumination 112 is set as the height of the ball 204. Calculated as the height of the vertex of.

  Each of these components is realized by an arbitrary combination of hardware and software, centering on a CPU, a memory, a storage unit for storing a program, and various interfaces. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus. FIG. 1 shows functional unit blocks instead of hardware units.

  The shape, position, etc. of the object in the image can be specified by acquiring the angles of the cameras 102 and 104 with respect to the substrate 200 and the offset values of the XYZ axes in advance and combining the images. In the present embodiment, specifically, the first image A of the first camera 102 and the second image B of the second camera 104 are measured in height within the range where the BGA 202 is located, and a reference point is set. Which pixel in the images A and B in the two cameras 102 and 104 is specified by, for example, calibration work using a jig or the like. Then, in which pixel of the first image A of the first camera 102 the target point is located, and in which pixel of the second image B of the second camera 104 the target point is located, what is the height? Can be identified by comparing with a reference point.

  An inspection method using the inspection apparatus 100 configured as described above will be described with reference to the flowchart of FIG.

  First, the third illumination 114 is turned on to highlight the contours of all the balls 204, and the ball XY position detection unit 120 determines the XY position of the ball center of all the balls 204 from the first image A acquired by the first camera 102. Specify (step S10). Specifically, the ball XY position detection unit 120 detects the center of gravity at the position on the XY plane of all the balls 204, and separates the area of the substrate 200 and the area of the ball 204 in the first image A. FIG. 2A shows an imaging state of the ball 204 in the first image A, and FIG. 3A shows an imaging state of the ball 204 in the second image B. FIGS. 2 and 3 show a state in which the apex 212 side of the ball 204 is depressed.

  Next, the third illumination 114 is turned off and the first illumination 110 is turned on, and the flat portion 206 of the ball 204 is detected from the first image A acquired by the first camera 102 by the first flat position detector 124 ( Step S20). As shown in FIG. 2B, in the first image A, the flat portion 206 that is parallel to the plane of the substrate 200 in the shape of the ball 204 has the highest reflected light intensity with respect to irradiation from above. Get higher. The flat portion 206 of one ball 204 obtained by the first flat position detection unit 124 becomes a portion that emits reflected light with high luminance by the light of the first illumination 110 in the divided region of the ball 204.

  Subsequently, the edge detection unit 128 detects the edge part 210 of the depression 208 of the ball 204 from the first image A acquired by the first camera 102 (step S30). Specifically, the edge portion 210 is a portion having a large luminance difference as seen when the depressed portion around the high luminance portion of the flat portion 206 detected by the first flat position detection unit 124 is formed. .

  Thereafter, the first illumination 110 is turned off and the second illumination 112 is turned on, and the flat portion 206 of the ball 204 is detected from the second image B acquired by the second camera 104 by the second flat position detection unit 126 ( Step S40). In the second image B, the flat portion 206 has the highest brightness of reflected light having a reflection angle of 45 degrees with respect to irradiation with an incident angle of 45 degrees with respect to the plane of the substrate 200. As shown in FIG. 3B, the flat portion 206 of one ball 204 obtained by the second flat position detecting unit 126 is increased by the light of the second illumination 112 in the area of the cut ball 204. It becomes a part that emits reflected light of luminance. Thereby, on the two images of the first image A and the second image B, the position of the same flat portion 206 can be specified by extracting a portion with high luminance.

  Subsequently, the edge detection unit 128 detects the edge part 210 of the depression 208 of the ball 204 from the second image B acquired by the second camera 104 (step S50). Specifically, the edge portion 210 is a portion having a large luminance difference as seen when the depressed portion around the high luminance portion of the flat portion 206 detected by the second flat position detection unit 126 is formed. .

  After step S50, it is determined whether or not the apex side of the ball 204 is depressed based on the detection state of the edge portion 210 (step S60). In the present embodiment, it is determined that the ball 204 is depressed when the edge portion 210 is detected in steps S30 and S50, and it is determined that the ball 204 is not depressed when the edge portion 210 is not detected. Is done.

  When the edge part 210 is detected in step S60 and it is determined that the ball 204 is depressed, the ball vertex calculation part 132 compares the edge part 210 of the first image A and the second image B, and The vertex 212 is calculated (step S70). Specifically, the ball vertex calculation unit 132 detects and compares the contour points of the edge portion 210 of the depression 208 in both the first image A and the second image B in dot units.

  More specifically, as shown in FIGS. 2B and 3B, in the ball apex calculation unit 132, a flat portion that emits reflected light with high luminance detected by the first flat position detection unit 124. The barycentric position of 206 is grasped as a feature point CP on the first image A, and the barycentric position of the flat part 206 that emits high-intensity reflected light detected by the second flat position detecting unit 126 is displayed on the second image B. It is grasped as a feature point CP. The two screens are corrected for differences in screen position, tilt, etc. by the calibration function, and the ball apex calculation unit 132 knows which position on the screen appears when one object has a predetermined height. ing. That is, it is possible to specify the height of the target shape depending on the position in each of the first image A and the second image B.

  Then, after the flat portion 206 is rubbed together, the edge portion 210 around the flat portion 206 portion is rubbed. These alignments may be performed simultaneously in parallel. The calculation of the vertex 212 based on the edge portion 210 is that the probability that the vertex 212 of the ball 204 appears on the ridge line caused by the collapse when the ball 204 is collapsed by contact with the outside and has a shape different from the spherical shape. It is effective because it is high (see FIGS. 2 and 3). In addition, when the edge part 210 is not detected, the vertex 212 can be detected by each flat position detection part 124,126.

  As shown in FIG. 2C and FIG. 3C, the ball apex calculation unit 132 uses the barycentric position of the flat portion 206, which is a high-luminance lump, as the feature point CP, and the brightness and darkness located around the feature point CP. The edge portion 210 is detected. The detection of the edge part 210 is performed by, for example, differentiating the luminance to detect the polar region. In the present embodiment, in addition to the edge portion 210 serving as a turning point from a relatively bright portion to a dark portion around the feature point CP, an edge portion serving as a turning point from a relatively dark portion to a bright portion outside the feature point CP. 210 is also detected. The turning point from the bright part to the dark part around the feature point CP is a detection target because it may be an edge of the surface forming the vertex 212. In addition, the turning point from the dark part to the bright part on the outside may be a shadow due to the difference in height between the high part outside the depression 208 on the ball 204 and the angle of illumination. Since there is a possibility that the edge portion 210 includes a ridge line forming the apex 212 of the ball 204, it is a detection target.

  The detection of the edge portion 210 is detected as a difference in position with respect to the center of gravity of the flat portion 206, and an edge common to the periphery of the target flat portion 206 in the two screens of the first image A and the second image B. When the part 210 exists, the edge part 210 is made effective. FIG. 2C and FIG. 3C show a vector in which one point is clarified with a center point as a base point and a pixel portion corresponding to the edge portion 210 as an end point. When the flat portion 206 does not exist, a vector is created by selecting another feature point CP. The detected edge portion 210 is invalidated when there is a discrepancy in the position coordinates between the two screens, or when the rate of change in luminance fluctuation of the edge portion 210 is low. Even when the edge portions 210 are aligned, since the positions of the balls 204 are detected in substantially the same scale in the first image A and the second image B, it is determined whether the positions of the edge portions 210 have substantially the same positional relationship. Thus, the shape can be aligned.

  The ball apex calculation unit 132 includes the barycentric position of the flat portion 206 detected by the first flat position detection unit 124 and the second flat position detection unit 126, and the first image A and the second image detected by the edge detection unit 128. When the position of the edge portion 210 of B is transferred as data, the XY coordinates of the center of gravity of the flat portion 206 when viewed from above and the HY coordinates of the center of gravity of the flat portion 206 when viewed from obliquely above are targeted. The XYZ position of the shape is calculated. Then, the edge portion 210 is extracted from the first image A and the second image B with the center of gravity of the flat portion 206 as a base point. In the present embodiment, the position of the Y-axis direction is set to be parallel between the first image A and the second image B so that the axis of the second camera 104 faces to be parallel to the X-axis. Also good. Thereby, with respect to the height position on the Z-axis of the edge portion 210, the position on the Y-axis on the XY-axis plane of the first image A is on the Y-axis on the HY-axis plane of the corresponding second image B. It can be calculated from the value on the H-axis at the position. In the second image B, the H axis is an axis extending in the vertical direction in the second image B, and the Y axis is an axis extending in the horizontal direction in agreement with the Y axis in the first image A. It should be noted that the coordinate system of the second image B is arbitrary and can be changed according to the installation position of the second camera 104 and the like.

  If it is determined in step S60 that the ball 204 is not depressed, the ball vertex calculation unit 132 calculates the position of the flat portion 206 as the vertex 212 of the ball 204 (step S80). Specifically, the ball apex calculation unit 132 associates the positions of the light spots (flat portions 206) detected by the first flat position detection unit 124 and the second flat position detection unit 126, and the reference obtained in the calibration operation. The actual height of each ball 204 is calculated in comparison with the point. In the first image A and the second image B, the position of the ball 204 is detected at substantially the same scale. Therefore, when a plurality of flat portions 206 are detected in two images, the positions of the flat portions 206 are substantially the same. It is possible to perform shape matching by determining whether or not the positional relationship is satisfied.

  After step S70 and step S80, the coplanarity calculation unit 134 calculates the coplanarity value of the entire package of the BGA 202 (step S90). Specifically, the coplanarity calculation unit 134 calculates the least square plane from the heights of all the balls 204, shifts the calculated plane so that the most protruding ball 204 has a height of zero, and uses this as a reference A surface. The distance at which the vertex 212 of each ball 204 is separated from the reference plane in the normal direction with respect to the created reference plane is defined as the coplanarity value of each ball 204, and the maximum coplanarity value of all balls 204 is the target coplanarity of the BGA 202. Define as value and output.

  As described above, according to the inspection apparatus 100 of the present embodiment, the position of the edge portion 210 is three-dimensionally grasped by the first image A and the second image B having different imaging angles with respect to the plane of the substrate 200. The height of the portion 210 with respect to the plane can be obtained without hindrance. Therefore, when the apex 212 of the ball 204 is not formed in a planar shape and the apex side is depressed, the apex of the ball 204 can be accurately detected. And the detection accuracy of the coplanarity of the whole package of BGA202 can be improved greatly.

  In the above-described embodiment, the relative positions of the cameras 102 and 104 and the lights 110, 112, and 114 with respect to the substrate 200 are not changed. However, these relative positions change. It may be. For example, as shown in FIG. 5, an XY stage 140 as a moving unit that moves the substrate 200 in a direction parallel to the plane, and a stage control unit 136 that controls the movement of the substrate 200 by controlling the XY stage 140. You may make it prepare.

  Also in this inspection apparatus 300, the vertex 212 of the ball 204 is detected in the same procedure as in the above embodiment. In the inspection apparatus 300, the package of the BGA 202 can be detected by dividing it into multiple fields of view, and the coplanarity of the package of the BGA 202 can be calculated from the positional information of the apex 212 of the ball 204 detected in a plurality of fields of view.

  In this inspection apparatus 300, the position of the ball 204 detected in each field of view on the package of one BGA 202, including the information of the stage control unit 136 of the XY stage 140, is included in the position of the ball 204. Manage location information. As for the height information of the ball 204 when straddling each field of view, it is necessary to correct the inclination of the plane of the XY stage 140 and connect the height information of a plurality of fields of view. This correction information is measured and recorded in advance at the time of calibration. Then, the heights of the vertices 212 of the balls of a plurality of fields of view are totaled in consideration of correction information of the position of the substrate 200 on the XY stage 140 and the tilt of the XY stage 140 at the time when the images A and B are acquired. From this, the coplanarity is calculated.

  In this inspection apparatus 300, since the coplanarity of one package can be inspected by dividing it into a plurality of fields of view, the resolution per pixel of each of the cameras 102 and 104 can be increased, and more accurate inspection is possible. Further, for example, since there is no restriction for imaging one package on one screen, it is possible to handle packages of various sizes of BGA 202 while maintaining the same resolution. Therefore, it is possible to deal with many types of products while maintaining high accuracy.

  In the above-described embodiment, the cameras 102 and 104 are arranged so as to face the substrate 200. For example, an image sensor corresponding to the second camera 104 is directed downward, and the substrate 200 is formed by a lens. Needless to say, the present invention can also be applied to an optical system in which the upper image is bent and imaged. In short, the first imaging unit is arranged so that the light reflected from the flat part parallel to the plane of the projecting part of the light emitted from the first irradiation part is incident, and the second imaging part is irradiated from the second irradiation part. It is only necessary that the reflected light from the flat portion of the protruding portion is incident on the projected light.

  In the above-described embodiment, the feature point CP using the center of gravity of the flat portion 206 is shown, but another position may be used as the feature point CP. The feature point CP is a reference point for performing the shape matching of the edge portion 210 in the first image A and the second image B, and can be arbitrarily set.

  In the above-described embodiment, the cameras 110, 112, and 114 are used to detect the edge portion 210. However, the edge portion 210 is detected from images taken from different angles, and the edge is detected. What is necessary is just to detect the height of the part 210. Further, the elevation angle of the second camera 104 and the second illumination 112 when viewed from the substrate 200 may not be 45 degrees. However, the range of 45 degrees to 60 degrees is preferable. In addition, the angle of the third illumination 114 is arbitrary, and it is needless to say that other specific detailed structures can be appropriately changed.

BRIEF DESCRIPTION OF THE DRAWINGS It is schematic structure explanatory drawing of the test | inspection apparatus which shows the 1st Embodiment of this invention. It is explanatory drawing of the 1st image A imaged with the 1st camera, (a) shows only the outline part of a ball | bowl, (b) shows the state lighted up with 1st illumination, (c) is 1st illumination. Shows a state in which the edge portion is specified by being lit up. It is explanatory drawing of the 2nd image B imaged with the 2nd camera, (a) shows only the outline part of a ball | bowl, (b) shows the state lit up with 2nd illumination, (c) is 2nd illumination. Shows a state in which the edge portion is specified by being lit up. It is a flowchart which shows operation | movement at the time of implementing the inspection method of an inspection apparatus. It is schematic structure explanatory drawing of the inspection apparatus which shows a modification.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Inspection apparatus 102 1st camera 104 2nd camera 110 1st illumination 112 2nd illumination 114 3rd illumination 120 XY position detection part 124 1st flat position detection part 126 2nd flat position detection part 128 Edge detection part 132 Ball vertex calculation Unit 134 coplanarity calculation unit 136 stage control unit 140 stage 200 substrate 202 BGA
204 Ball 206 Flat part 208 Depression 210 Edge part 212 Vertex 300 Inspection device A 1st image B 2nd image CP Feature point

Claims (15)

An inspection device for obtaining the height of the apex of a substantially spherical protrusion formed on the plane of an inspection object,
A first imaging unit that images the protrusion from a direction perpendicular to the plane of the inspection object;
A second imaging unit that images the protrusion from an oblique direction with respect to the plane of the inspection object;
Edge detecting means for detecting a depressed edge portion of the protruding portion in the first image imaged by the first imaging unit and the second image imaged by the second imaging unit;
Based on the position of the edge portion in the first image detected by the edge detection means and the position of the edge portion in the second image detected by the edge detection means, An inspection apparatus comprising: apex height calculating means for calculating a height.   The vertex height calculation means calculates the height of the edge portion relative to the plane by comparing the position of the edge portion in the first image and the position of the edge portion in the second image, and in the edge portion The inspection apparatus according to claim 1, wherein the height of the highest portion is calculated as the height of the apex of the protruding portion.   The vertex height calculating means calculates a difference between a position coordinate of a predetermined feature point and a position coordinate of the edge portion for each of the first image and the second image, and compares the difference. The inspection apparatus according to claim 2, wherein the position of the edge portion in the first image is compared with the position of the edge portion in the second image. A first irradiation unit that irradiates light substantially perpendicularly to the plane of the inspection object;
A second irradiation unit that irradiates light at a predetermined incident angle with respect to the plane of the inspection object;
The first imaging unit is arranged so that light reflected from a flat part parallel to the flat surface of the protruding part is incident on the light emitted from the first irradiation part,
The second imaging unit is arranged so that the reflected light of the flat portion in the protruding portion is incident among the light irradiated from the second irradiation unit,
The vertex height calculation means sets the predetermined feature point as the center of gravity of the reflection portion of the flat portion in a state where the first image is irradiated with light from the first irradiation unit, and the second image. The inspection apparatus according to claim 3, wherein the center of gravity of the reflection portion of the flat portion in a state where light is irradiated from the second irradiation portion.   The vertex height calculating unit is configured to irradiate light from the first irradiation unit when the edge detection unit does not detect the edge portion of the depression in the first image and the second image. The height of the flat part obtained by comparing the reflection part of the flat part in the first image and the reflection part of the flat part in the second image in a state where light is irradiated from the second irradiation part. The inspection apparatus according to claim 4, wherein the inspection apparatus calculates the height of the top of the protrusion.   The inspection apparatus according to claim 1, further comprising a moving unit that moves the inspection object in a direction parallel to the plane. A plurality of the protrusions are formed on the inspection object,
The coplanarity calculating means for calculating the coplanarity of each protrusion with respect to the plane from the height of the vertex of each protrusion calculated by the vertex height calculating means. The inspection apparatus as described in any one. A third irradiation unit that irradiates light substantially horizontally with respect to the plane of the inspection object;
The position detection part which detects the position of the said protrusion part based on the image imaged by the said 1st imaging part in the state which irradiated the light from the said 3rd irradiation part, It is further provided, It is characterized by the above-mentioned. The inspection device described in 1. An inspection method for obtaining the height of the apex of a substantially spherical protrusion formed on a plane of an inspection object,
A first image acquisition step of acquiring the first image by imaging the protrusion from a direction perpendicular to the plane of the inspection object;
A second image acquisition step of acquiring the second image by imaging the protrusion from a direction oblique to the plane of the inspection object;
An edge detection step of detecting a depressed edge portion of the protruding portion in the first image acquired in the first image acquisition step and the second image acquired in the second image acquisition step;
When the depressed edge portion is detected in the first image and the second image in the edge detection step, the position of the edge portion in the first image and the position of the edge portion in the second image are determined. And a vertex height calculating step for calculating a height of the vertex of the protruding portion based on the inspection method.   In the vertex height calculating step, the position of the edge portion in the first image is compared with the position of the edge portion in the second image to obtain the height of the edge portion with respect to the plane, and the edge portion The inspection method according to claim 9, wherein the height of the highest portion of the projection is calculated as the height of the apex of the protruding portion.   In the vertex height calculation step, for each of the first image and the second image, by calculating a difference between a position coordinate of a predetermined feature point and a position coordinate of the edge portion, and comparing the difference The inspection method according to claim 10, wherein the position of the edge portion in the first image is compared with the position of the edge portion in the second image. In the first image acquisition step, light is irradiated substantially perpendicularly to the plane of the inspection object, and light reflected by a flat portion parallel to the plane in the protruding portion of the irradiated light is reflected. Appear in the first image,
In the second image acquisition step, the plane of the inspection object is irradiated with light at a predetermined incident angle, and the irradiated light is reflected by a flat portion parallel to the plane of the protruding portion. So that light is reflected in the second image,
The inspection according to claim 11, wherein, in the vertex height calculation step, the predetermined feature point is a center of gravity of a reflection portion of the flat portion in the image with respect to the first image and the second image. Method.   In the vertex height calculation step, when the edge portion of the depression is not detected in the first image and the second image in the edge detection step, the reflection portion of the flat portion and the first image in the first image are detected. The inspection method according to claim 12, wherein the height of the flat portion obtained by comparing the reflection portion of the flat portion in two images is calculated as the height of the apex of the protruding portion. A plurality of the protrusions are formed on the inspection object,
The coplanarity calculation process of calculating the coplanarity with respect to the said plane of each said projection part from the height of the vertex of each said projection part calculated in the said vertex height calculation process is characterized by the above-mentioned. The inspection method according to any one of the above.   A position for detecting the position of the protrusion by imaging the protrusion from a direction perpendicular to the plane of the inspection object in a state where light is irradiated substantially horizontally with respect to the plane of the inspection object. The inspection method according to claim 14, further comprising a detection step.

Images