JP2006337033A - Measuring apparatus having confocal optical system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a measuring apparatus having a confocal optical system which can perform the measurement of a desired part under measurement within an object under measurement even when it is covered with a semitransparent material or the like. <P>SOLUTION: The measuring apparatus having a confocal optical system successively acquires and stores the luminous quantity data of the object under measurement for different height positions while relatively moving the focal plane of the confocal optical system with respect to the object under measurement in its height direction. By specifying the part under measurement of the object under measurement and arbitrarily specifying a desired measurement range on a light amount data distribution graph for the height direction of the specified part under measurement, the height or the like of the part under measurement can be measured on the basis of the stored light amount data. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a measuring apparatus that measures a test site with high accuracy using a confocal optical system.

2. Description of the Related Art Conventionally, a shape measuring device using a confocal optical system is known as a device that can measure the shape of a minute test object with high accuracy. The confocal optical system is designed to obtain a high-contrast image by removing the light emitted from other than the focal plane (the focused surface) by placing a pinhole at the back focal position of the objective lens. Yes, if a scanning image (image data) of the test object is obtained by moving the focal plane of the confocal optical system relative to the test object by a certain amount in the height direction, Since an image in which only the portion that coincides with the focal plane becomes bright is obtained at each height, a highly accurate shape measurement result can be obtained by performing three-dimensional image processing on the image.
JP 2002-13917 A

  However, when the height of the test object is measured, that is, when the outer shape of the test object is obtained, all of the scanned image data at predetermined intervals on each Z-axis is captured. For example, in the case where the test object is covered with a protective resin (semi-transparent member) such as a semiconductor chip, a plurality of focal planes (peak light quantity images) are detected in the Z-axis direction. become. Then, the shape measuring device can measure the outer shape of the semiconductor chip, but shows the maximum value among the plurality of light intensity peaks even when trying to measure the height of the test site hidden in the protective resin of the semiconductor chip. Only image data (contour image) is acquired as a measurement result, and the height of a desired test site may not be measured.

  The present invention has been made in view of such problems, and has a confocal optical system capable of measuring a desired test site inside a test object covered with a translucent substance or the like. The object is to provide a measuring device.

  In order to achieve such an object, the measuring apparatus of the present invention includes a confocal optical system, and moves the focal plane of the confocal optical system relative to the test object in the height direction while different height positions. A device that sequentially obtains and stores light quantity data of the test object for each time, specifies a test site of the test object, and on the light quantity data distribution graph in the height direction of the specified test site In this case, by arbitrarily designating a desired measurement range, it is possible to measure the height of the region to be examined based on the stored light quantity data.

According to the measuring apparatus according to the present invention, by designating the test site of the test object, the measurement range in the height direction of the test site can be freely set. Can do. In addition, by storing the measurement range as teaching data, a scan range can be set without waste, and shape measurement can be performed based on the light amount data acquired within the range, thereby speeding up data processing such as shape measurement. .

(Overall configuration of measuring apparatus 1)
Hereinafter, a preferred embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows a configuration diagram of a confocal optical system of a height measuring apparatus 1 according to an embodiment of the present invention. FIG. 2 shows a block diagram of the processing device 18 of the measuring device 1.
FIG. 1 shows a measuring apparatus 1 according to the present invention, which is a height measuring optical system 2 that measures a wide field of view with a single scan, and performs two-dimensional measurement of a position to be obtained by zooming with high accuracy. And a dimension measuring optical system 3.

  The height measurement optical system 2 forms a confocal optical system, and is arranged on the optical axis from the test object 40 so that the ¼λ plate 6, the first objective lens 7, the second objective lens 8, and the confocal point. It has a disk 9, a first polarizing beam splitter 11, a first relay lens 12, a half prism 13, and a high sensitivity CF camera 14, and has a low sensitivity CF camera 15 on the side of the half prism 13, A first condenser lens 17 and a first illumination light source 16 are provided on the side of the one-polarization beam splitter 11.

  In the height measuring optical system 2, the light beam emitted from the first illumination light source 16 is converted into parallel light by the first condenser lens 17, and enters the first polarization beam splitter 11. The first polarization beam splitter 11 has a property of transmitting P-polarized light and reflecting S-polarized light among the incident light beams, and the reflected light beam (S-polarized light) is irradiated on the upper surface of the confocal disk 9. .

  The confocal disc 9 is formed in a thin disc shape, and a large number of confocal pinholes 9a penetrating in the thickness direction are formed in a spiral shape (the confocal disc having such a configuration is also called a nippo disc), It arrange | positions so that it may orthogonally cross with respect to the optical axis of the height measurement optical system 2 so that the light ray (illumination light) reflected by the 1st polarizing beam splitter 11 may be interrupted | blocked. The confocal disc 9 is rotated at a predetermined rotational speed by a motor 10, and the illumination light that has passed through the confocal pinhole 9 a is placed on the stage by the second objective lens 8 and the first objective lens 7. The object 40 is condensed and irradiated as an image of the confocal pinhole 9a. In FIG. 1, two confocal pinholes 9a are shown.

  The image of the confocal pinhole 9a focused and irradiated on the test object 40 is reflected by the surface of the test object 40 (hereinafter referred to as “object plane O”), and is incident on the first objective lens 7 again. Then, the light is condensed by the first objective lens 7 and the second objective lens 8 to form an image of the surface of the test object 40 on the lower surface of the confocal disc 9, and further passes through the confocal pinhole 9a. At this time, the illumination light (S-polarized light) emitted from the first illumination light source 16 and passed through the confocal pinhole 9a is 1 / 4λ disposed between the first objective lens 7 and the test object 40. Since it passes through the plate 6 twice, it is converted to P-polarized light. Therefore, it can pass through the first polarizing beam splitter 11, and after passing through the first polarizing beam splitter 11, it is condensed by the first relay lens 12, and a part of the light passes through the half prism 13. The image is formed on the imaging surface of the sensitivity CF camera 14 as a reflected image of the confocal pinhole 9a. Further, the remaining light beam is reflected by the half prism 13 and formed on the imaging surface of the low-sensitivity CF camera 15 as a reflected image of the confocal pinhole 9a. The half prism 13 is configured to reflect about 70% of light and transmit about 30% of light.

  The object 10 to be inspected in the scan measurement apparatus 1 according to the present embodiment may have a difference of about 200 times in the reflectance of the illumination light on the object plane O due to the difference in material. It is difficult to efficiently measure the reflected image of the confocal pinhole 9a with a single camera. Therefore, two CCD cameras (a high-sensitivity CF camera 14 and a low-sensitivity CF camera 15) having different detection sensitivities according to the material (reflectance) of the test object 40 are divided into transmitted light and reflected light by the half prism 13. ) To detect. The reflected image signals detected by the cameras 14 and 15 are passed to the processing device 18.

  As shown in FIG. 2, the processing device 18 includes basic components such as an arithmetic control device 31, a storage device 32, a display device 33, a keyboard 34, and a pointing device 35, and details thereof will be described later.

  Further, the surface of the optical member closer to the test object than the first polarizing beam splitter 11 is configured by converting the S-polarized light into P-polarized light by the ¼λ plate 6 and transmitting it through the first polarizing beam splitter 11. Is blocked by the first polarization beam splitter 11 so that it does not enter the cameras 14 and 15 (this flare light is S-polarized light because it does not pass through the quarter-λ plate 6). And the accuracy of height measurement can be improved.

  Thus, by rotating the confocal disk 9 on which the confocal pinhole 9a is formed at a predetermined speed by the motor 10, the illumination light passing through the confocal pinhole 9a is spot light and the test object 40 is rotated. The object plane O is scanned, and the reflected image of the confocal pinhole 9a (spot light) is detected by the high-sensitivity CF camera 14 or the low-sensitivity CF camera 15 and processed by the processing device 18, thereby The height information of the object 40 can be obtained.

  The reflected image of the confocal pinhole 9a (spot light) formed on the imaging surfaces of the cameras 14 and 15 appears bright only when the focal plane of the first objective lens 7 coincides with the object plane O. When O deviates from the focal plane in the optical axis direction, it appears dark. Therefore, the scan measurement apparatus 1 according to the present embodiment is configured to obtain the height information by moving the first objective lens 7 in the optical axis direction and acquiring the light intensity change of the reflected image with the cameras 14 and 15. Has been. Hereinafter, a specific method will be described.

  The first objective lens 7 is configured to be movable up and down along the optical axis with respect to the test object 40 by a drive unit 19 and a position sensor (such as an optical encoder) 19a that detects a position in the optical axis direction. ing. Therefore, when the first objective lens 7 is moved up and down so that the object plane O of the test object 40 is positioned in the vicinity of the focal plane of the first objective lens 7, the optical axis direction position Z of the first objective lens 7 is reached. 4 and the light intensity I of the reflected images detected by the cameras 14 and 15, the light intensity I at the position Z where the focal plane of the first objective lens 7 coincides with the object plane O as shown in FIG. The distribution is such that the intensity I is the largest. Therefore, the processing device 18 controls the drive unit 19 to move the first objective lens 7 in the optical axis direction, and obtains the light intensity peak value of the signals detected by the cameras 14 and 15, thereby obtaining the first value at that time. The height information of the test object 40 can be obtained from the position Z (information obtained from the position sensor 19a) of the objective lens 7 in the optical axis direction. At this time, by forming the imaging optical system including the first objective lens 7 and the second objective lens 8 so that the object 40 side is telecentric, the first objective lens 7 moves and defocuses. Since there is no magnification variation in this imaging optical system, a reflected image can be measured at the same magnification, and measurement accuracy is not affected.

  As described above, since the object 40 is scanned by rotating the confocal disk 9 at a predetermined speed at a high speed, the height information for a large number of points on the object 40 (object plane O) is scanned. Can be measured simultaneously. At this time, the measurement range on the object plane O corresponds to the imaging field of view of the imaging surfaces of the cameras 14 and 15, and the measurement range of the height information of the test object 40 is the movement of the first objective lens 7 in the optical axis direction. Corresponds to the range. Further, when the object plane O is inclined with respect to the optical axis of the height measuring optical system 2, reflected light (reflected image of the confocal pinhole 9 a) reflected by the object plane O enters the first objective lens 7. Cannot measure without. Therefore, the range of the inclination of the object plane O in which the height information can be measured is determined by the numerical aperture of the first objective lens 7.

  Next, the two-dimensional measurement optical system 3 will be described. The two-dimensional measurement optical system 3 forms a bright field optical system, and the first mirror 20 is inserted and removed between the second objective lens 8 and the confocal disk 9 on the optical axis of the height measurement optical system 2. By doing so, the two-dimensional measuring optical system 3 is configured so that the test object 40 can be measured. Therefore, the two-dimensional measurement optical system 3 is arranged in order from the test object 40 to the first objective lens 7, the second objective lens 8, the first mirror 20, and the side of the first mirror 20. The second mirror 21 that is opposed to the second mirror 21, the second polarizing beam splitter 22 disposed above the second mirror 21, the second relay lens 23, and the two relay lenses 23 are disposed between the two lenses. The zoom optical system 24 and the bright field camera 25 are configured. A second condenser lens 27 and a second illumination light source 26 are provided on the side of the second polarizing beam splitter 22. The first objective lens 7 and the second objective lens 8 are shared with the height measuring optical system 2.

  In the two-dimensional measurement optical system 3, the light beam emitted from the second illumination light source 26 is converted into parallel light by the second condenser lens 27 and enters the second polarization beam splitter 22. The second polarizing beam splitter 22 also has a property of transmitting P-polarized light among the incident light rays and reflecting S-polarized light, and the reflected light rays (S-polarized light) enter the second mirror 21. Then, the light beam reflected laterally by the second mirror 21 is further reflected downward by the first mirror 20, condensed by the second objective lens 8 and the first objective lens 7, and irradiated on the test object 40. . The light beam reflected by the test object 40 is collected by the first objective lens 7 and the second objective lens 8, reflected by the first mirror 20 and the second mirror 21, and emitted upward to the second polarizing beam splitter 22. Incident.

  Also in the two-dimensional measurement optical system 3, the S-polarized light beam reflected by the second polarizing beam splitter 22 is transmitted twice through the ¼λ plate 6 and converted into a P-polarized light beam. The illumination light reflected by the object 40 can pass through the second polarizing beam splitter 22. Therefore, since the flare light generated on the surface of the optical member closer to the test surface than the second polarization beam splitter 22 is blocked by the second polarization beam splitter 22, a clear image of the test object 40 can be acquired. it can. The light beam that has passed through the second polarizing beam splitter 22 is scaled by the second relay lens 23 and the zoom optical system 24 and is formed on the imaging surface of the bright field camera 25 as a reflected image of the test object 40. The signal detected by the bright field camera 25 is passed to the processing device 18.

  Image data obtained by these cameras 14, 15 and 25 is input to the arithmetic and control unit 31 as shown in FIG. The position of the drive unit 19 in the Z-axis direction is detected by the position sensor 19a, and the detection information is input to the arithmetic control device 31 (microprocessor). The arithmetic control device 31 includes a pinhole disk drive unit 10, a drive unit 19, a storage device 32 (semiconductor memory, removable memory, etc.), a display device (liquid crystal display) 33, a keyboard 34, and a pointing device (for example, a mouse). 35 is connected to output control signals to the pinhole disk drive unit 10 and the drive unit 19 to operate them, and to store the image data sent from the cameras 14 and 15 in the storage device 32 (save) And can be displayed on the display device 33. The storage device 32 also stores image data of the bright field camera 25.

  Therefore, in the present height measurement apparatus 1, if the pinhole disc 9 is rotated and the objective lens 7 is moved every predetermined step (every interval) in the Z-axis direction, the image is taken each time. An image of the cross-sectional shape of the test object 40 obtained when the test object 40 is cut by the focal plane FS of the focus optical system 2 (an image in which only the outline of the cross-section is bright, that is, a confocal image of the test object 40) ) Is obtained in a state associated with information on the position in the height direction (z-axis direction) of the focal plane FS where the image is taken. Since the height position (height direction position) of the focal plane FS has a one-to-one correspondence with the Z-axis direction position of the objective lens 7, the arithmetic and control unit 31 outputs the Z of the objective lens 7 output from the position sensor 19a. The height position of the focal plane FS can be grasped based on the axial position information. Here, the height position of the focal plane FS is the position of the focal plane FS in the height direction with respect to the stage 30 (the height direction of the test object 40), and the focal plane FS is on the upper surface 30a of the stage 30. If the height position of the focal plane FS when matching is set to zero, the focal plane FS coincides with the test site in the test object 40 (one site in the test object whose height is to be measured). In the state, the value of the height position of the focal plane FS detected at that time is the value of the height of the region to be examined (the height from the upper surface 30a of the stage 30) as it is. Hereinafter, the height position (distance) of the focal plane FS relative to the upper surface 30a of the stage 30 is referred to as a focal plane position z.

  In the above description, the relative movement between the test object 40 and the confocal optical system 2 is realized by moving the objective lens 7 in the Z-axis direction. However, the present invention is not limited to this, and the stage 30 is moved in the Z-axis direction. Alternatively, the objective lens 7 and the stage 30 may be relatively moved.

(Description of the structure of the test object 40 for measuring the shape and height)
The structure of the test object 40 (here, IC chip) according to the present embodiment will be described. FIG. 3 is a cross-sectional view of an IC package that is the test object 40. FIG. 4 shows an example of the height measurement of a part of the wire 144 of the IC package, where the vertical axis shows the displacement in the Z-axis direction and the horizontal axis shows the change in the received light quantity I.

  As shown in FIGS. 3 and 4, the test object 40 that is an IC chip covers the substrate 141 made of a non-transparent material, the semiconductor chip 142 placed on the upper surface 141 a of the substrate 141, and the semiconductor chip 142. And a protective resin 143 made of a non-transparent material. A plurality of wires 144 for conducting the substrate 141 and the semiconductor chip 142 are connected to the plurality of electrodes 141 b on the substrate and the plurality of electrodes 142 a on the semiconductor chip 142. The plurality of wires 144 are protected by a protective resin 143.

  Here, an example in which the present measuring apparatus is used for determining a defect of an IC chip will be described. For example, in the case of a normal IC chip, all the wires 144 are covered with the protective resin 143 by the protective resin 143. However, if the wire bonder is defective, the wires 144 jump out of the protective resin 143. Sometimes. In order to determine whether or not such a wire 144 exists, high-speed measurement by this measuring apparatus is necessary. For this purpose, the maximum height of each wire 141 connected to the semiconductor chip 142 and the substrate 141 may be measured to determine whether or not each maximum height is within the height of the protective resin 143.

  The test object 40 which is an IC chip shows an example in which a plurality of wires 144 exist and the height of each wire 144 is measured at the maximum height. For convenience of explanation, FIG. 4 shows an example of measuring the height of a part of the wire 144, the vertical axis indicates the displacement in the Z-axis direction of the focal plane, and the horizontal axis indicates the longitudinal position of the wire 144 and the received light amount I. Shows changes.

  Note that the scanning direction (moving direction) of the focal plane FS in the scanning image acquisition step is a positive direction (a direction from bottom to top) with respect to the z axis. The start point of the height direction scanning region is an arbitrary position (focal plane position z = zs) lower than the upper surface 141a of the substrate 141, and the end point is an arbitrary position higher than the upper surface 143a of the protective resin 143 (focal plane position z = ze).

  The amount of light received by one pixel constituting the CCD cameras 14 and 15 changes during the scanning process in the height direction (z-axis direction), and the pixel is focused on a portion on the test object 40 as an imaging region. In the process of approaching the surface FS, the received light quantity (light quantity data) increases, and reaches a maximum (light quantity peak) when the two coincide. In the process in which the focal plane FS moves away from the site on the test object 40, the amount of received light decreases. FIG. 4 shows the relationship between the focal plane position (scanning direction) z and the amount of received light I in the pixel. From this, when measuring the height of the test site of the wire 144 in the test object 40, for example, pay attention to the three pixels (A, B, C) including the test site in the imaging region. To explain, the scanning image (collection of image data for each height corresponding to the cross-sectional contour shape of the test object 40) in the height direction region is read out with a predetermined height direction resolution (z1 to z15) and the amount of light. If the peak (I12, I22, I32) is detected and the focal plane position (z4, z8, z12) corresponding to the light intensity peak is obtained, the focal plane position z (= zi) becomes the height of the region to be examined as it is. .

  In order to measure the height with higher accuracy, an interpolation calculation process may be used as described below. In the present height measuring apparatus 1, the focal plane FS of the confocal optical system 2 has a predetermined resolution in the height direction (z-axis direction) with respect to the test object 40 (predetermined steps zi, i = 1 to 15). While relatively moving (changing the focal plane position z in the height direction), imaging is performed by the CCD cameras 14 and 15 at different height positions. Therefore, when the CCD cameras 14 and 15 obtain a scanned image (image data) of the test object 40, the data of the change in the amount of received light with respect to the change in the height position (focal plane position zi) of the focal plane FS is discrete. Since it can only be obtained, it is not possible to directly detect the focal plane position z below the resolution corresponding to the light intensity peak. Therefore, in order to obtain the focal plane position (Zpi, i = 1 to 3) of the light quantity peak, the focal plane positions before and after the light quantity peak (for example, in the case of the pixel A, z3, z4 and z5) are used to determine the focal plane position below a predetermined resolution.

(Description of the shape measurement process of the test object)
Hereinafter, a specific procedure for measuring the shape of the test object 40, which is an IC package, by the height measuring apparatus according to the present embodiment will be described with reference to FIGS. FIG. 5 shows a conceptual diagram of the shape measuring process (image processing) of the measuring apparatus. FIG. 6 is a flowchart of the shape measurement process (image process) of the measurement apparatus. 7 and 10 show display examples in which the display device 33 displays the test object 40 at a predetermined photographing magnification of the measuring device 1. FIG. 8 is a diagram for schematically explaining a cross section of the test object 40 displayed on the display device 33. 9 and 11 show graphs of light quantity distribution data of image data of the test object 40 displayed on the display device 33. FIG.

  As shown in FIG. 5, the characteristic configuration of the shape measurement process includes a peak detection unit 31a, a three-dimensional (3D) image construction unit 31b, an integration processing unit 31c, and an error processing unit 31d. The received scanned image data is in acquisition and storage. The scanned image data here refers to image data in units of pixels of the imaging device 19 (two-dimensional CCD element). That is, the imaging device 19 stores scanned image data while moving the focal plane of the confocal optical system in the z-axis direction at predetermined resolutions (predetermined intervals) in order to acquire a three-dimensional image of the test object 40. However, the storage device 32 does not store all of the scanned image data sent from the imaging device 19. Unnecessary scanned image data (image data in units of pixels) is removed by the peak detector 31a of the arithmetic and control unit 31 interposed between the imaging device 19 and the storage device 32, and the contour of the test object 40 is stored in the storage device 32. Only image data corresponding to an image (xyz coordinate information and light amount data information corresponding to image data having a peak value of light amount data) and only defective pixel information are stored. The principle is that only the optical image (confocal image) of the test site at the confocal position of the test object 40 formed by the confocal optical system is extracted from the image data of the imaging device 19. is there. Only the image data corresponding to the confocal image is stored in the storage device 32 (32a to 32d) as the image data associated with the xyz coordinate information.

  The imaging device 19 sequentially examines the object at different height positions (predetermined resolution in the z-axis direction) while moving the focal plane of the confocal optical system 2 relative to the object 40 in the height direction. 40 pieces of image data are acquired for each pixel of the imaging device 19, and the image data acquired at different height positions are sequentially output to the peak detector 31a.

  The peak detector 31a compares the light quantity data of the pixel unit image data acquired for each height position, and finally extracts the image data indicating the peak value of the light quantity data in each pixel (for the champion data). Extraction). At that time, when the height position is obtained with higher accuracy than the resolution for each height position, the image data indicating the peak value and the image data before and after the peak value are extracted for the interpolation processing calculation. The image data for each unit pixel is composed of light amount data information (Ik-1, Ik, Ik + 1) and xyz coordinate information (Zk-1, Zk, Zk + 1), and is stored in the image data storage unit 32d. Remembered. The xyz coordinate information is sent to the three-dimensional (3D) image construction unit 31b, and the light amount data information is sent to the integration processing unit 31c.

  The three-dimensional (3D) image construction unit 31b obtains the height of each part of the test object based on the xyz coordinate information (Zk-1, Zk, Zk + 1) of the image data indicating the peak value at each pixel. The height data of each part of the test object is output and stored in the 3D image storage unit 32b. This 3D image is displayed on the display device 33 by, for example, pseudo color information based on the height data of the test object 40. Thereby, when the monitor of the display device 33 is viewed, the height of each test part of the test object 40 can be identified by the color.

  Based on the light amount data information (Ik-1, Ik, Ik + 1) of image data at different heights, the integration processing unit 31c generates an integrated image (synthesized image) by integrating them. This integrated image data is obtained by integrating the light amount data information of the image data indicating the peak value of the test object 40 and the light amount data information of the image data before and after that (based on the reflected light from the object 40). Image data). The accumulated image data is output to and stored in the accumulated image storage unit 32c.

  The error processing unit 31d specifies a defective pixel from the image data for each pixel acquired by the imaging device 19. By monitoring the light amount data information of the image data acquired for each pixel, for example, the light amount data is compared with a predetermined upper limit or lower limit light amount data, and from the appropriate luminance range of the upper limit or lower limit light amount data. A deviating pixel can be identified as a defective pixel. As a result, the address information corresponding to the defective pixel on the imaging surface of the imaging device 19 is output to and stored in the error image storage unit 32d.

The data stored in the storage device 32 can be transferred to an external device such as a PC through a transfer path such as Ethernet.
Next, the processing procedure of the arithmetic and control unit 31 of this measuring apparatus will be described with reference to FIGS.

  FIG. 6 is a flowchart showing the contents of an execution program stored in the storage device 32. The arithmetic and control unit 31 loads the execution program from an execution program storage device (not shown) into an internal memory (not shown) and executes it.

  When measuring the shape of the finished product that is the test object 40 of the IC chip, the height of the protective resin 143 and the height of the wire 144 of the test object 40 are measured. For example, as shown in the display device 33 of FIG. 7, the test object 40 is enlarged and displayed at a predetermined photographing magnification of the measuring device 1, and the location indicated by the measurement mark 50 is measured. FIG. 8 is a diagram for schematically explaining a cross section of the test object 40 displayed in FIG. 7 for convenience of explanation.

  By measuring the maximum height of each of a large number of wires 144 formed on the test object 40 shown in FIG. 7, it is inspected whether there is a defect in which each wire 144 jumps out from the protective resin 143 ( The distance between the protective resin 143 and the wire 144 can be obtained). Therefore, the shape measurement of the wire 144 will be specifically described below.

  First, in step 1 to step 8, image data of an image acquisition area (rectangular area) of the test object 40 displayed on the display device 33 of FIG. 7 is acquired, that is, basic image data of the test object 40. To get.

  Steps 1 and 2: First, as shown in FIG. 6, the arithmetic and control unit 31 operates the drive unit 19 to set the focal point F of the confocal optical system 2 to the start point (z1 in FIG. 4) of the height direction scanning region. After matching (step S1), the pinhole disk drive unit 24 is operated to start rotation of the pinhole disk 14 (step S2). As a result, an image (image of the surface 141a of the substrate 141) corresponding to the current focal plane position zi = z1 (starting point of the height direction scanning region) is displayed on the display device 33 on the CCD cameras 14 and 15.

  Step 3: The arithmetic and control unit 31 acquires the scanned image data (light amount data information and xyz coordinate information) at the focal plane position zi from the CCD cameras 14 and 15 and temporarily stores it in the image data file. This image data file stores a predetermined number of pieces of image data, and luminance data information and xyz coordinate information are stored for each pixel unit as each image data.

This image data is sequentially sent to the peak detection processing unit 31a (step 4) and the error processing unit 31d.
The error processing unit 31d identifies a defective pixel from the image data that has been sequentially sent as already described. The address data of the identified defective pixel is stored in the error image storage unit 32d.

  Step 4: The peak detection process is executed. The arithmetic and control unit 31 opens the image data file and sequentially reads the image data. From the read image data, light amount data information about a region in the image corresponding to the test region of the test object 40 is detected, and the light amount data information in the image data read this time and the light amount data information in the image data read last time, Compare Then, from the state where the light amount data information of the image data read this time is larger than the light amount data information of the image data read last time, the light amount data information of the image data read this time is larger than that of the image data read last time. When switching to a state smaller than the light amount data information, it is determined that there is a light amount peak for the area in the image in the meantime (detection of the light amount peak). For example, when N is a certain integer, the light quantity data information IN of the Nth image data (focal plane position is zN) and the light quantity data information IN-1 of the N−1th image data (focal plane position is In this case, the light intensity data information IN + 1 of the (N + 1) th image data (the focal plane position is assumed to be zN + 1) and the Nth image data When IN + 1 <IN in relation to the light quantity data information IN, there is a light quantity peak for the above-mentioned region in the image between the light quantity data information IN + 1 and the light quantity data information IN-1. I understand that.

  For example, the peak value I12 is detected based on the light amount data distribution IA at the portion A of the test site A of the wire 144 in FIG. Similarly, peak values I32 and I22 are detected based on the light amount data distributions IB and IC at the test sites B and C on the wire 144. In order to perform this peak detection, when interpolation processing is performed, light amount data information of image data before and after the image data indicating the peak value (I11 and I13 are detected in the pixel A corresponding to the point A, and the point B is detected. I11 and I13 are detected in the pixel B corresponding to, and I21 and I23 are detected in the pixel C corresponding to the point C).

  Step 5: 3D image construction processing is executed. The xyz coordinate data related to the image data having the light quantity peak detected in step 4 (light quantity data I12, I22, I32 in FIG. 4) is acquired. In FIG. 4, at point A of the wire 144, xyz coordinate information (z3 to z5) related to the light amount data information (I11 to I13) is acquired. At point B, xyz coordinate information (z11 to z13) related to the light amount data information (I31 to I33) is acquired. At point C, xyz coordinate information (z7 to z9) related to the light amount data information (I21 to I23) is acquired. Based on the xyz coordinate information, the shape of the wire 144 is three-dimensionally constructed from the height information (z information) of the wire 144. For example, the three-dimensional image of the wire 144 is constructed by color by pseudo-coloring the height information. Alternatively, a three-dimensional image of the wire 144 is constructed with brightness and darkness by displaying the height information in brightness.

Then, only the xyz coordinate information of the image data having the light amount peak of the wire 144 is stored in the 3D image storage unit 32 b of the storage device 32.
That is, when focusing attention on the pixel of the imaging device 19 corresponding to the point A portion of the wire 144 shown in FIG. 4, there are 15 pieces of image data Z1 to Z15 as the point A image data. . However, in the apparatus of the present embodiment, not all 15 image data of the pixels of the imaging device 19 corresponding to the part of the point A are stored in the storage device 32, and only 3 image data of Z3 to Z5 are stored. Will be. Thus, in this embodiment, the storage capacity of image data can be suppressed, and the shape measurement process of the wire 144 can be executed at high speed. Since the same processing as described above is performed in relation to each pixel of the imaging device 19, the storage capacity of the image data can be significantly reduced.

  Step 6: Perform integrated image processing. Only the light amount data information is acquired from the image data having the light amount peak detected in step 4. In FIG. 4, light amount data information (I11 to I13) is acquired at point A of the wire 144. At point B, light quantity data information (I31 to I33) is acquired. At point C, light quantity data information (I21 to I23) is acquired. An integrated image of point A of the wire 144 is created by integrating light amount data information (I11 to I13). Similarly, the points B and C are created by integrating three pieces of light quantity data information. As a result, the wire 144 is shaped according to the brightness information.

Only the light amount data information of the image data having the light amount peak of the wire 144 is stored in the integrated image storage unit 32 c of the storage device 32. Other image data is not stored.
That is, when focusing attention on the pixel of the imaging device 19 corresponding to the point A portion of the wire 144 shown in FIG. 4, there are 15 pieces of image data Z1 to Z15 as the point A image data. . However, in the apparatus of the present embodiment, not all 15 image data of the pixels of the imaging device 19 corresponding to the part of the point A are stored in the storage device 32, and only 3 image data of Z3 to Z5 are stored. Will be. Thus, in this embodiment, the storage capacity of image data can be suppressed, and the shape measurement process of the wire 144 can be executed at high speed. Since the same processing as described above is performed in relation to each pixel of the imaging device 19, the storage capacity of the image data can be significantly reduced.

  Step 7: Move the focal plane position of the confocal optical system by a fixed amount zi (i = 1 to 15) in the Z-axis direction. For every fixed amount of movement, image data having a light amount peak is searched as in the above-described step.

  Step 8: Steps 3 to 7 are repeated until the focal plane position of the confocal optical system is scanned in the Z-axis direction (until z15 is reached). Thereby, the image data of the test object 40 can be efficiently acquired, the storage capacity of the storage device 32 can be suppressed, and the shape measurement of the test object 40 can be speeded up.

  Step 9: In the image display routine of the test object 40, a 3D image and an integrated image of the test object 40 can be displayed on the monitor screen of the display device 33. At that time, the user can select whether or not to read out defective pixel data using the keyboard 34. The user can read error information from the address data of the defective pixel and display that the image data obtained from the defective pixel has low reliability. For example, by attaching a specific color to the image data corresponding to the defective pixel, it can be identified as a normal pixel on the monitor display. Thus, if a defective pixel can be specified, the reliability of the height calculation result can be obtained as information at the same time.

Further, in the height / step calculation subroutine of step 9, as will be described later, the height / step information of the test part of the test object desired by the operator is efficiently obtained.
In the teaching process in step 10, as will be described later, the test site of the test object desired by the operator set in step 9 is stored, and the method of acquiring the height / step information is stored in the storage device 32. To do.

Next, the processing in step 9 and step 10 will be described in detail with reference to FIGS.
As shown in FIG. 7, the operator can measure a desired height of the test object 40 based on a partially enlarged image of the test object 40 displayed on the display device 33 (an image based on the imaging magnification of the measurement apparatus 1). Is pointed on the display device 33 (specified by the designation mark 50). Height measurement is performed at a location designated by the designation mark 50 (a desired location on the test object 40). As described above, this height measurement processing is performed by the arithmetic and control unit 31 by interpolating or non-interpolating the coordinate Z based on the integrated image, that is, the light quantity peak data, Z can be obtained by interpolation or non-interpolation processing.

  The image of the test object 40 displayed on the display device 33 shown in FIG. 7 may be a bright field image obtained by the CCD camera 25 or an accumulated image (or 3D image) obtained by the CCD cameras 14 and 15.

  When the image displayed on the display device 33 is a bright field image by the CCD camera 25, only the portion (pixel) designated by the designation mark 50 is subjected to the image processing already described with reference to FIG. Then, the image data of the family account image and the 3D image is stored in the storage device 32.

If the operator has selected to perform layer thickness calculation (level difference calculation), the thickness of the detected part of the test object is calculated from the XYZ coordinate data, as in the height measurement.
Image data corresponding to the pixel B on the point (designation mark 50) pointed by the operator (by the mouse) is read from the storage device 32. The light amount data distribution related to the luminance of the read image data is displayed on the display device 33 as shown in FIG. On the display device 33, a display using an integrated image (or a 3D image or a bright field image acquired by the CCD camera 25) of the test object 40 in FIG. 7 and a graph display of the light amount data distribution on the designation mark 50 in FIG. And may be displayed separately.

  With reference to FIGS. 9A and 9B, two methods for manually setting a location where the operator measures the height will be described. As conceptually shown in FIG. 8, in the light amount data distribution in the Z-axis direction of the designation mark 50, a light amount peak P1 occurs on the surface of the protective resin 143 at the Z1 position, and on the surface of the wire 144 at the Z2 position. A light amount peak P2 occurs, and a light amount peak P3 occurs on the surface of the substrate surface 141a at the Z3 position.

First, the first manual setting method will be described.
As shown in FIG. 7, the operator designates a desired measurement location (test site of the test object) with a designation mark 50 on the image displayed on the display device 33. A light amount data distribution graph at a location designated by the designation mark 50 is displayed as shown in FIG. The operator designates a measurement position in the height direction to be measured, such as a desired height and a step, using the measurement line 51A while looking at the light quantity data distribution graph of FIG.

  The measurement line 51A can be moved left and right in the height direction (Z direction) on the display screen using the pointing device 35, and the operator can check the measurement position in the height direction while checking the measurement value display section 33a on the display screen. To decide. The measurement value display unit 33a displays numerically the position in the Z direction of the intersection between the measurement line 51A and the light quantity data distribution graph.

  Specifically, when the operator wants to measure the height position of the wire 144 designated by the designation mark 50 and the step from the surface of the protective resin 143 to the wire 144, the following setting operation is performed. It becomes.

  First, the operator sets the designation mark 50 at a desired position of the wire 144 on the display screen of FIG. On the display device 33, a light amount data distribution graph at a location designated by the designation mark 50 is displayed as shown in FIG. The operator clicks the measurement line 51A on the display screen of FIG. 9A with the mouse and moves it to the vicinity of the second light quantity peak P2. As the measurement line 51A moves, a numerical value display is displayed on the measurement value display unit 33a, so that the operator can search for the apex of the wire 144 while observing the numerical value display, and can determine the position in the Z direction.

  After that, when the operator turns on a teaching switch (not shown) to execute the teaching data storage process of step 10, the operator sets the XY coordinate position data of the designated mark 50 and the Z coordinate position of the measurement line 51A in the height direction. The data is stored in the storage device 32. From the next measurement, the operator can read the teaching data from the storage device 32 and execute the automatic measurement operation.

Next, a second manual setting method will be described.
As shown in FIG. 7, the operator designates a desired measurement location (test site of the test object) with a designation mark 50 on the image displayed on the display device 33. A light amount data distribution graph at a location designated by the designation mark 50 is displayed as shown in FIG. As shown in FIG. 9B, the operator can set a measurement calculation range to be measured, such as a desired height and level difference, using the measurement line 51A and the measurement line 51B while looking at the light amount data distribution graph. The pair of measurement lines 51A and 51B can freely move in the Z direction on the display screen of FIG. 9B, and the area sandwiched between the pair of measurement lines is the range in which the height / step measurement is calculated. By setting this measurement line as teaching data, unnecessary measurement calculation processing is not performed.

Next, the teaching data storing operation will be described (step 10).
The designation mark 50 and the measurement lines 51A and 51B set by the operator are stored as teaching data in the storage device 32 as described above by operating a storage switch (not shown), and subsequent automatic measurement operations are stored in the teaching data. It becomes possible to process based on.

  When the shape measurement process is performed based on the teaching data, the measurement range of the measurement lines 51A and 51B is the focal plane of the confocal optical system 2 by the drive unit (10, 19 in FIG. 1) and the test object. 40 and a calculation range for height measurement based on the light amount data by the calculation control device 31. That is, in this case, the scan range and the calculation range coincide with the measurement range set by the pair of measurement lines.

Next, based on FIG.10 and FIG.11, the shape measurement process advanced from the above-mentioned shape measurement process is demonstrated.
As shown in FIG. 10, three desired positions on the test object 40 are specified by pointing (designating with designation marks 60 </ b> A to 60 </ b> C) the height measurement positions of the test object 40 desired by the operator on the display device 33. Is measured. If the operator has selected to perform layer thickness calculation (level difference calculation), the thickness of the detected part of the test object is calculated from the XYZ coordinate data, as in the height measurement.

  Image data (image data in the Z direction) corresponding to the pixel on the location (designation marks 60A to 60C) pointed by the operator (with the mouse) is read from the storage device 32. The light amount data distribution related to the luminance of the read image data is synthesized and displayed on the display device 33 as shown in FIG.

  On the display device 33, the display of the integrated image (or 3D image or bright field image) of the test object 40 in FIG. 7 and the graph display of the light amount data distribution on the designation marks 60A to 60C in FIG. May be. Referring to FIG. 11D conceptually, FIGS. 11A to 11C correspond to the light amount data distribution in the Z-axis direction of the designation marks 60A to 60C, respectively. FIG. 11A is a graph showing a state in which light intensity peaks are generated between the surface of the protective resin 143 and the surface of the wire 144 in the pixel of the designation mark 60A. FIG. 11B is a graph showing a state in which a light amount peak is generated between the surface of the protective resin 143 and the surface of the semiconductor chip 142 in the pixel of the designation mark 60B. FIG. 11C is a graph showing a state in which light intensity peaks are generated between the surface of the protective resin 143 and the surface of the substrate surface 141a in the pixel of the designation mark 60C. When these FIGS. 11A to 11C are shown in one graph, the distribution graph of the light amount data as shown in FIG. 11D is obtained, and FIG. 11D is finally displayed on the display device 33. . Depending on the operator's convenience, the display device 33 may be configured to simultaneously display the drawings (A) to (C).

  The operator can set the measurement calculation range to be measured, such as a desired height and level difference, using the measurement line 61A and the measurement line 61B while looking at FIG. The pair of measurement lines 61A and 61B can freely move in the Z direction on the display screen of FIG. 11D, and an area sandwiched between the pair of measurement lines is a range in which measurement and calculation are performed. In this case, the height of the semiconductor chip 142 from the surface of the substrate 141 or the height of the wire 144 is obtained.

  When the light amount data actually used for the measurement calculation is set to the minimum necessary within the measurement range sandwiched between the measurement lines 61A and 61B, the calculation ranges 71A, 71B, and 71C are obtained. The calculation ranges 71A to 71C are automatically set from the three light quantity peak values of the light quantity data distribution graph. That is, the arithmetic and control unit 31 automatically extracts a light intensity peak value greater than or equal to a predetermined value from the distribution graph of the light intensity data, and sets a predetermined range as calculation ranges 71A to 71C in the front-rear direction around the extracted light intensity peak. . Accordingly, in this case, the range 61 sandwiched between the measurement lines 61A and 61B is set as the scan range, and the calculation ranges 71A to 71C are set as the ranges in which the measurement calculation processing is performed within the scan range. By storing these ranges in the storage device 32 as teaching data, the subsequent measurement calculation processing is efficiently executed.

  As another method for setting the scan range, there is the following method. That is, in the case shown in FIG. 11D, the interval between the adjacent calculation ranges 71A and 71B is more than a predetermined distance, and the interval between the calculation ranges 71B and 71C is within the predetermined distance. The device 31 sets a minimum scan range including the calculation range 71A as a scan range, and automatically sets a minimum scan range including both the calculation ranges 71B and 71C separately. In this way, the scanning operation can be made more efficient.

FIG. 1 is a diagram showing a confocal optical system of a shape measuring apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram of the shape measuring apparatus according to the first embodiment. FIG. 3 is a side view showing an IC package which is an example of a test object. FIG. 4 shows an example of the height measurement of a part of the wire 144 of the IC package, where the vertical axis shows the displacement in the Z-axis direction and the horizontal axis shows the change in the received light quantity I. FIG. 5 shows a conceptual diagram of the shape measuring process (image processing) of the measuring apparatus. FIG. 6 is a flowchart of the shape measurement process (image process) of the measurement apparatus. FIG. 7 shows a display example in which the display device 33 displays the test object 40 at a predetermined photographing magnification of the shape measuring device. FIG. 8 is a diagram for schematically explaining the relationship between the cross section of the test object 40 displayed on the display device 33 and the light amount data distribution. FIG. 9 shows a graph of the light amount data distribution on the designation mark of the test object 40 displayed on the display device 33. FIG. 10 shows a display example in which the display device 33 displays the test object 40 at a predetermined photographing magnification of the shape measuring device. FIG. 11 shows a graph of the light amount data distribution on the three designation marks of the test object 40 displayed on the display device 33.

Explanation of symbols

1 Height measuring device
2 Confocal optics
7 Objective lens
16 Light source
14, 15, 25 CCD camera
19 Drive unit
30 stages
31 arithmetic control unit
32 storage devices
34 Keyboard
40 specimen

Claims (7)

Confocal optics,
Driving means for relatively moving the focal plane of the confocal optical system and the test object in the height direction;
The light amount data of the test object is sequentially received for each height position of the focal plane of the confocal optical system that receives the image of the test object via the confocal optical system and is relatively moved by the driving unit. In a measuring apparatus having a confocal optical system comprising a photoelectric conversion means for obtaining
Display means for displaying the shape of the test object;
On the specimen image displayed by the display means, first designation means for designating a position of a desired test site to be measured;
Graph display means for displaying a distribution graph of the light quantity data in the height direction obtained by the photoelectric conversion means at the test site designated by the first designation means;
On the distribution graph of the light amount data displayed by the graph display means, second specifying means for specifying a desired position in the height direction to be measured;
A measurement operation means for measuring the shape of the test object based on the light amount data at the position in the height direction designated by the second designation means; and a confocal optical system comprising: measuring device. The measuring apparatus having the confocal optical system according to claim 1,
The second designation means displays one measurement line that moves in the height direction on the distribution graph of the graph display means, and designates a desired location of the distribution graph by the measurement line,
The measurement apparatus having a confocal optical system, wherein the measurement calculation means calculates a height in the height direction at an intersection of the measurement line and the distribution graph. The measuring apparatus having the confocal optical system according to claim 1,
The second specifying means displays a pair of measurement lines for setting a desired measurement range on the distribution graph displayed by the graph display means, and sandwiches the distribution graph between the pair of measurement lines. Specify
The measurement calculation means calculates a height in the height direction based on the light amount data corresponding to the distribution graph sandwiched between the pair of measurement lines, and has a confocal optical system . In the measuring apparatus having the confocal optical system according to any one of claims 1 to 3,
A measuring apparatus having a confocal optical system, comprising teaching data storage means for storing designation information designated by the first and second designation means as teaching data. In the measuring apparatus having the confocal optical system according to claim 4,
The confocal optical system characterized in that the measurement range set by the second designation means stored by the teaching data storage means is a scan range by the drive means and a calculation range by the measurement calculation means Measuring device. In the measuring apparatus having the confocal optical system according to claim 4,
The measurement calculation means detects a peak light amount position from the distribution graph of the light amount data, resets a predetermined range including the peak light amount position in the measurement range as a calculation range,
The drive means sets the measurement range to a scan range,
The measuring apparatus having a confocal optical system, wherein the teaching data storage unit stores the calculation range of the measurement calculation unit and the scan range of the driving unit as teaching data. The measuring apparatus having the confocal optical system according to claim 6,
When there are a plurality of calculation ranges, if the calculation ranges adjacent to each other are within a predetermined distance, one scan range including all the calculation ranges is set, and conversely, the calculation range is equal to or greater than a predetermined distance. A measuring apparatus having a confocal optical system, characterized in that the scan range is divided and set when they are separated from each other.

Images