JP2012078330A - Method for adjusting movement of camera unit in lens inspection apparatus and focus check tool - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve automatic movement correction for a camera unit in a lens inspection apparatus so that a test lens can be measured accurately and easily regardless of position accuracy of a three-axis stage which moves the camera unit in three-axis directions.SOLUTION: Before mounting a test lens, a check plate 65 is mounted as a tool for focus checking on a lens mount 6. On a rear side of a display plate 67 is printed a design pattern, and a center portion and a peripheral portion of the design pattern are imaged by a camera unit 30 to find a best focus position. Based on a difference between a position of the camera unit 30 when best focus is obtained in the center portion and a position of the camera unit 30 when best focus is obtained in the peripheral portion, a correction factor is found for moving the camera unit 30 with a three-axis stage.

Description

  The present invention relates to a lens inspection machine that uses a camera unit to image the imaging state of a lens to be evaluated in order to evaluate imaging performance. Specifically, the camera unit is moved to an appropriate imaging position using a triaxial stage. In doing so, the present invention relates to a camera unit movement adjustment method and a focus check jig of a lens inspection machine that can obtain high positional accuracy.

  In order to inspect the performance of a lens built in or used in a camera, for example, a lens inspection machine capable of measuring digitized data based on MTF (Modulation Transfer Function) is used. Such a lens inspection machine, for example, as known from Patent Document 1, makes measurement light from a test chart facing a predetermined distance from a test lens enter the test lens and form an image on the test lens. The aerial image of the test chart is taken with a camera unit. The camera unit incorporates a relay optical system, an imaging optical system, and a CCD or CMOS type image sensor. The aerial image of the test chart is enlarged by the relay optical system and then imaged on the image sensor by the imaging optical system. I am letting. Then, the imaging data obtained from the image sensor is analyzed to obtain numerical data, and MTF measurement is performed based on the numerical data.

  In the lens inspection machine, both the test chart and the camera unit are configured to be movable in a plane perpendicular to the optical axis of the test lens, and it is possible to measure the imaging performance of the test lens with respect to off-axis light. It has become. Therefore, when the test lens is a wide-angle lens, the measurement light from the test chart is incident from the off-axis direction inclined from the optical axis of the test lens, and the field curvature is changed based on the imaging state. Measurements can also be made.

  In addition, a lens inspection machine is also known in which a light beam from a test chart is collimated by a collimator lens and is incident on a test lens. In the case of this type of lens inspection machine, a test chart image is basically obtained on the focal plane of the test lens regardless of the angle at which the measurement light from the test chart is incident on the test lens. Therefore, it is only necessary to move the camera unit that captures the chart image in parallel with the focal plane of the lens to be examined, and efficient measurement can be performed.

JP 2004-163207 A

  In a lens inspection machine that obtains measurement data by imaging a test chart image formed in the air with a test lens with a camera unit, the chart image formed in the air is enlarged with a relay optical system to obtain sufficient measurement accuracy. After that, an image is taken with an image sensor. In addition, the camera unit is supported by the three-axis stage so that the chart image can be imaged at or near the best focus position, and further, even when measurement light is incident from the off-axis, the camera unit can be imaged at an appropriate position. In addition to the optical axis direction (Z-axis direction) of the lens to be measured, the lens can be moved in two directions (X-axis and Y-axis directions) perpendicular to each other in a plane perpendicular to the optical axis.

  By the way, when assembling the above-described lens inspection machine, first, the collimator and the lens mount, which are the reference of the measurement optical system, are fixed to the surface plate with high accuracy and the reference optical axis is set, and then the measurement system is configured. The triaxial stage and camera unit are assembled. The collimator and lens mount can be attached to the surface plate with high positional accuracy using an optical measuring instrument such as transit or a laser length measuring instrument, but a triaxial stage that holds the camera unit movably. It is very difficult to position the collimator and the lens mount with high accuracy and fix them to the surface plate.

  In particular, when the focal length of the lens to be tested is reduced to, for example, several millimeters, the tip of the camera unit incorporating the relay optical system must be positioned close to the lens mount, and the camera unit is connected. If an attempt is made to position the three-axis stage on the surface plate with this, the tip of the camera unit may come into contact with the lens mount and damage the relay optical system. For this reason, the triaxial stage must be fixed to the surface plate while performing fine adjustment while mechanically monitoring its position before mounting the camera unit, and its positioning accuracy is limited. For example, in most cases, the Z-axis of the three-axis stage has an inclination error with respect to the reference optical axis, although the reference optical axis from the collimator and the Z-axis of the three-axis stage are desired to be completely parallel to each other. Have.

  A camera unit is attached to the three-axis stage thus positioned with high precision so that the Z axis and the imaging optical axis are parallel to each other, but in a lens inspection machine, based on the coordinate system determined by the reference optical axis from the collimator. However, since the camera unit is moved in three axial directions orthogonal to each other on the three-axis stage, the tilt error described above directly affects the measurement accuracy. For example, when the camera unit is moved in the X-axis direction or Y-axis direction perpendicular to the Z-axis using a three-axis stage, the amount of movement relative to the plane perpendicular to the reference optical axis from the collimator does not become zero, and the above tilt Movement according to the error occurs. The positioning accuracy of the three-axis stage required for the lens inspection machine is ± 2 μm or less with respect to the plane perpendicular to the reference optical axis when the camera unit is moved 5 mm in the X-axis direction or Y-axis direction on the three-axis stage. However, it is actually very difficult to position and fix the three-axis stage on the surface plate with such high accuracy.

  The present invention has been made in consideration of such circumstances, and its purpose is to position the triaxial stage relative to the surface plate even if the accuracy of the mounting position of the triaxial stage cannot be driven to the high accuracy required for lens inspection. It is an object of the present invention to provide a camera unit movement adjustment method for a lens inspection machine that can move and adjust a camera unit with high precision using the three-axis stage, and to provide a focus check jig.

  In order to achieve the above object, the present invention has a lens mount that aligns the optical axis of a mounted test lens with a reference optical axis and positions and holds the test lens in the reference optical axis direction. The measurement light from the test chart is incident on the test lens held by the lens mount, and the image of the test chart imaged on the focal plane of the test lens is supported by the camera unit supported by the three-axis stage. Lens inspection for inspecting the lens to be inspected based on the obtained imaging signal while imaging in a plane perpendicular to the imaging optical axis having an inclination error with respect to the reference optical axis and moving in the imaging optical axis direction In the camera unit movement adjustment method of the machine, prior to mounting the lens to be examined, a display plate on which a pattern pattern for focus check is written is attached to the reference optical axis. A first step of positioning the lens mount so as to be perpendicular to the lens mount, and imaging the central portion of the symbol pattern while moving the camera unit on the imaging optical axis by the three-axis stage, A second step of capturing the position of the camera unit imaged in an optimally focused state as the first position data from the triaxial stage; and in the plane perpendicular to the imaging optical axis of the camera unit on the triaxial stage. Then, the camera unit is moved in parallel with the imaging optical axis, and the periphery of the symbol pattern is imaged. The position of the camera unit where the periphery of the symbol pattern is imaged in an optimally focused state is determined. A third step of capturing as second position data from the axis stage, and the imaging optical axis based on the first position data and the second position data. A fourth step of obtaining a correction coefficient representing a ratio between the movement amount of the camera unit in the vertical in-plane direction and the movement amount of the camera unit in the reference optical axis direction and storing the correction coefficient in a memory; and a focus check When the camera unit is moved to capture the test chart image formed at different positions on the focal plane after the lens to be tested is mounted on the lens mount instead of the display plate Is configured to control the operation of the three-axis stage based on the correction coefficient read from the memory.

  It is desirable to provide a cross line pattern at the central part and the peripheral part of the symbol pattern written on the display plate for focus check. Then, the three-axis stage is operated while monitoring the image captured by the camera unit, and the centering process of the camera unit is performed so that the center of each cross-hair pattern matches the center of the imaging screen of the camera unit. Thus, more accurate first position data and second position data can be obtained, and the camera unit can be moved with high accuracy when measuring the lens to be examined.

  According to the present invention, the camera unit is movably supported by the three-axis stage, and is parallel to the reference optical axis from the collimator or along the imaging plane of the lens to be measured that is orthogonal to the reference optical axis. When acquiring the measurement data by imaging the imaging state of the lens under test while moving, the camera unit is specified even if the triaxial stage is not positioned with extremely high accuracy with respect to the reference optical axis. It is possible to obtain a measurement result having no problem in terms of accuracy in practice by accurately moving to a certain position.

It is a schematic plan view of a lens inspection machine. It is principal part sectional drawing of the lens mount periphery of a lens inspection machine. It is a block diagram which shows the outline of the electrical constitution of a lens inspection machine. It is explanatory drawing which shows the chart image and imaging range used for a lens measurement. It is explanatory drawing which shows the evaluation pattern of a chart image. It is explanatory drawing of the calculation process of a MTF value. It is explanatory drawing when interpolating the MTF value which becomes a peak from the MTF value of five points. It is explanatory drawing at the time of mounting | wearing a lens mount with a check plate. It is a front view of the symbol pattern provided in the check plate. It is a flowchart which shows the process sequence at the time of calculating | requiring a correction coefficient. It is a flowchart which shows the measurement procedure of a MTF value.

  1 and 2 showing an example of a lens inspection machine in which the present invention is used, a mount base 3 is fixed to the upper surface of the surface plate 2, and a bracket 4 fixed to the mount base 3 serves as a cylindrical mount holder 5. keeping. A cylindrical lens mount 6 is rotatably incorporated in the mount holder 5 via a bearing incorporated in its inner wall. A lens holder 8 holding the test lens 7 is attached to the front end portion of the lens mount 6, and the test lens 7 rotates integrally around the reference optical axis K as the lens mount 6 rotates. The reference optical axis K is on a horizontal plane parallel to the upper surface of the surface plate 2, and is set with high accuracy by the relative position between the lens mount 6 and an axial collimator 14 described later.

  In order to fix the lens holder 8 to the lens mount 6, screws are cut on the inner peripheral wall of the front end portion of the lens mount 6 and the outer peripheral wall of the rear end of the lens holder 8, and these are screwed together to fix the lens holder 8. Done. The lens holder 8 is formed with a lens holding portion that is precisely processed according to the shape and size of the test lens 7 so that the test lens 7 can always be fixed to the lens holder 8 in a fixed state. When the lens holder 8 is tightly screwed to the lens mount 6 and fixed, the rear end surface of the lens holder 8 comes into contact with the mount reference surface which is the front end surface of the lens mount 5, and the optical axis of the lens 7 to be tested. Coincides with the reference optical axis K, and the test lens 7 is positioned at a fixed position on the reference optical axis K.

  A motor 10 and a potentiometer 11 are fixed to the bracket 4, and gears 10 a and 11 a fixed to the respective shafts are engaged with a mount gear 6 a formed on the outer periphery of the rear end of the lens mount 6. The lens mount 6 can be rotated around the reference optical axis K by driving the motor 10, so that the measurement around the optical axis of the test lens 7 can be performed even when the measurement light is incident on the test lens 7 from a certain direction. It becomes possible. The rotation signal obtained from the potentiometer 11 is used for feedback control when the lens mount 6 is rotated so that the lens mount 6 can be rotated quickly and accurately.

  The surface plate 2 has an on-axis collimator 14 that allows measurement light to enter the test lens 7 along the reference optical axis K, and a pair of axes that allow the measurement light to enter from an off-axis direction inclined with respect to the reference optical axis K. Outer collimators 15 and 16 are provided. The off-axis collimators 15 and 16 are supported by swinging stages 20 and 21 that are movable on linear guides 18 and 19 that are inclined about ± 30 ° with respect to the reference optical axis K, respectively. The oscillating stages 20 and 21 are configured to move on the linear guides 18 and 19 by driving individual motors, and the oscillating stages 20 and 21 are incorporated with motors and are respectively supported off-axis. The collimators 15 and 16 can be swung around an axis perpendicular to the surface plate 2.

The rectilinear guides 18 and 19 are tilted away from the reference optical axis K as they approach the lens mount 6, and the collimators 15 and 16 can be rotated to arbitrary angles on the rectilinear guides 18 and 19. Therefore, even if the off-axis collimators 15 and 16 are brought close to each other, the measurement light can be incident on the test lens 7 at a large angle without bringing the tips of the collimators 15 and 16 into contact with the test lens 7. As a result, even when the lens 7 to be examined is of a super wide angle system or a fisheye system, it is possible to perform a wide range measurement in consideration of the maximum angle of view.

  As exemplified by the on-axis collimator 14, each collimator 14, 15, 16 includes a white LED 23 serving as an illumination light source, an illumination optical system 24 including a diffusion plate and a condenser lens, and a chart plate provided in a turret manner. 25 is illuminated. The chart plate 25 is provided with a different chart design depending on its rotational position. For example, an appropriate one selected according to the magnification of the lens 7 to be examined and the measurement conditions is set in the optical path of the illumination optical system 24. Is done. Since the chart plate 25 is located on the focal plane of the collimating lens 26, the measurement light from the chart pattern is emitted from the collimating lens 26 as a parallel light beam in any of the collimators 14, 15, and 16.

  The movement and swing of the off-axis collimators 15 and 16 are basically automatically executed by a dedicated actuator in accordance with the initial setting of measurement conditions. That is, when the imaging performance of the lens 7 to be measured is measured, the measurement conditions are initially set, and a measurement sequence is automatically created under the measurement conditions thus initialized. The operations of the respective collimators 14, 15, 16 are automatically controlled based on the sequence. Further, after the measurement is started, the test lens 7 can be rotated around the optical axis together with the lens mount 6 as necessary, and the rotation angle, the timing of the rotation, and the like are automatically performed according to the measurement sequence.

  A triaxial stage 28 is provided behind the mount base 3, and a camera unit 30 is held by a bracket 29 fixed to the triaxial stage 28. The triaxial stage 28 is connected to a drive unit 32 that includes a stepping motor as an actuator. Then, by driving the triaxial stage 28 with the driving pulse from the drive unit 32, the in-plane perpendicular to the reference optical axis K is assumed on the assumption that the triaxial stage 28 is accurately positioned and fixed to the surface plate 2. Thus, the camera unit 30 can be arbitrarily moved in three directions: a horizontal X-axis direction, a Y-axis direction perpendicular to the same plane, and a Z-axis direction parallel to the reference optical axis K. Each of the X-axis, Y-axis, and Z-axis has a positive arrow direction, and the X-axis has a positive direction in the direction from the front surface to the back surface.

  The camera unit 30 includes a first lens barrel 33 including a relay optical system having an enlargement function, a second lens barrel 34 containing an image forming optical system that forms an image of a light beam emitted from the relay optical system, and the image forming unit. The imaging unit 35 captures an image by an optical system with an image sensor, and the imaging optical axis thereof is parallel to the Z axis of the triaxial stage 28. A part of the first lens barrel 33 and the second lens barrel 34 enter the hollow portion of the lens mount 6, and the relay optical system in the first lens barrel 33 is an image of a chart symbol (imaged in the air by the test lens 7). Hereinafter, the chart image) is enlarged and transmitted to the imaging optical system in the second lens barrel 34, and the enlarged chart image is formed on the image sensor in the imaging unit 35. Since both the on-axis collimator 14 and the off-axis collimators 15 and 16 receive parallel light beams from the chart design, the image plane of the chart image coincides with the focal plane of the lens 7 to be examined.

  Image processing representing the contrast of the chart image can be obtained by performing image processing on the imaging signal obtained by imaging the chart image. Further, the image data is analyzed to evaluate the test lens 7 based on the MTF value. Can be done. The test lens 7 is not limited to a single lens as shown in the figure, and may be a compound lens in which a plurality of lenses are combined by joint or interval ring use.

  The operation of this lens inspection machine is controlled by each functional block under the control of the system controller 40 shown in FIG. 3, and basically a measurement process is performed according to a measurement sequence prepared in advance in the sequence memory 41. The setting input device 42 is used together with a display panel 43 such as a liquid crystal monitor for inputting various measurement conditions prior to the start of measurement and optical specification data of the lens 7 to be examined. For example, whether or not the off-axis collimators 15 and 16 are used, and when measuring while rotating the test lens 7, the measurement light is incident from the unit step angle of rotation, and further from the off-axis of the test lens 7. The incident angle variable range can be interactively input while referring to the guidance items displayed on the display panel 43.

  As the measurement sequence progresses, the system controller 40 sends a lighting command to the on-axis or off-axis collimators 14 to 16, whereby the white LED 23 is lit and the measurement light is incident on the lens 7 mounted on the lens mount 6. Will come to be. When measurement by the off-axis collimators 15 and 16 is performed, a movement command is sent from the system controller 40 to the actuator drive control ICs 45 and 46 at an appropriate timing according to the progress of the measurement sequence, and the off-axis collimators 15 and 16 receive the linear guide 18. , 19 are automatically slid, and are automatically oscillated by the oscillating stages 20, 21 to change the incident angle of the measuring light to the lens 7 to be measured. When it is necessary to rotate the lens 7 to be measured around the reference optical axis K during the measurement, a command is sent to the rotation control IC 47 and the feedback control of the motor 10 is performed while monitoring the signal from the potentiometer 11. As a result, the lens mount 6 is stopped at a target rotational position with high accuracy.

  In this lens measuring machine, regardless of whether the on-axis collimator 14 or the off-axis collimators 15 and 16 are used for measurement, the camera unit 30 is stepped at a constant pitch P in the direction of the reference optical axis K (Z-axis direction). While trying to get measurement data. Therefore, as described above, the camera unit 30 is supported by the three-axis stage 28, and the destination data is sent from the system controller 40 to the actuator drive control IC 48, and a predetermined number of drive pulses are supplied to the drive unit 32 correspondingly. As a result, the camera unit 30 moves via the three-axis stage 28. Drive pulses in the Z-axis direction are sent to the drive unit 32 in accordance with the movement destination data, and the number of each drive pulse is recorded in the movement data memory 49. Note that the amounts of movement of the triaxial stage 28 per drive pulse in the X-axis, Y-axis, and Z-axis directions are equal to each other.

  An imaging signal obtained from an image sensor built in the camera unit 30 is converted into digitized image data through a signal processing circuit 50 and stored in an image data memory 51. The image data processing circuit 52 performs image processing calculation based on the stored image data, acquires the contrast distribution data of the obtained image, and inputs it to the MTF calculation unit 54. The MTF calculation unit 54 performs a fast Fourier transform (FFT) process to calculate the MTF value, and stores the obtained MTF value data in the measurement data memory 55 together with the input contrast distribution data.

  The correction coefficient calculation unit 56 and the correction coefficient memory 57 cannot strictly maintain the positioning accuracy of the triaxial stage 28 with respect to the surface plate 2 when the camera unit 30 is moved to a predetermined measurement position by the operation of the triaxial stage 28. Even in this case, the camera unit 30 is provided so that it can be moved to the target destination as much as possible. These are used when obtaining correction coefficients for movement correction processing before the measurement of the lens 7 to be examined is started. The correction coefficient calculator 56 reads drive pulse data when the triaxial stage 28 is actually operated through the system controller 40 from the movement data memory 49, and calculates a correction coefficient based on the measurement data before and after the movement. The calculated correction coefficient is recorded in the correction coefficient memory 57, and is read and used by the system controller 40 when operating the triaxial stage 28 in measuring the lens 7 to be tested.

  The basic operation of the lens inspection machine will be described. During standard MTF measurement, the on-axis collimator 14 is used first, and the camera unit 30 is set at the first target position on the reference optical axis K. The measurement light from the on-axis collimator 14 enters the test lens 7 along the reference optical axis K, and the chart image from the chart plate 25 forms an aerial image on the focal plane of the test lens 7. This chart image is magnified about 10 times by the relay lens system incorporated in the first lens barrel 33 of the camera unit 30, and is imaged by the imaging unit 35 through the imaging optical system incorporated in the second lens barrel 34. . The camera unit 30 is set at the first target position on the reference optical axis K by the operation of the three-axis stage 28. With respect to the direction of the reference optical axis K, the camera lens 30 is input from the setting input device 42. The Z-axis coordinate value of the three-axis stage 28 is calculated based on the optical specification data.

  When measuring the imaging state on the reference optical axis K, it is difficult to strictly focus the camera unit 30 on the focal plane of the lens 7 to be measured. Chart images are taken at a plurality of measurement positions while moving in the K direction. For this reason, a defocus range including a front pin area and a rear pin area having a certain width before and after the image forming plane position (focal plane position) of the lens 7 to be measured is set as the center. Although the front end of the defocus range corresponds to the front end of the front pin area, there is no extreme out-of-focus condition even at this position, and the image outline can be captured with a sharpness that can be sufficiently distinguished. The same applies to the rear end of the focus range. Then, the camera unit 30 is stepped, for example, four times at a constant pitch P from the front end to the rear end of the defocus range, and measurement data is obtained by sequentially imaging at the five stopped positions.

When stepping the camera unit 30 within the defocus range, instead of moving the camera unit 30 from the front end to the rear end of the defocus range as described above, the camera unit 30 is moved from the rear end to the front end. You may do it. In addition, after performing the first imaging at the calculated best focus position, the second imaging by moving one step to the front pin side, the second imaging by moving two steps to the rear pin side, and so on. It is also possible to sequentially capture images while stepping the camera unit 30 by a repetitive movement method centered on the predicted best focus position. In this repetitive movement method, the total amount of movement of the camera unit 30 within the defocus range is large, but since the position where the image is first captured should be closest to the actual best focus position, the camera unit 30 is first selected. The probability that the chart image at the time of moving to the target position can be clearly captured is the highest, and the time required for the centering process (described later) performed immediately after that can be shortened.

  When performing imaging five times in the defocus range, always capture the center of the chart image at the center of the imaging screen so that the influence of the relay optical system and the imaging optical system does not affect the measurement result. Thus, the camera unit 30 is subjected to centering processing. For example, the camera unit 30 is accurately set to the target position due to factors such as the individual difference of the lens 7 itself, the displacement of the chart plate 25 incorporated in the on-axis collimator 14 in a turret switching manner, and the positioning accuracy of the triaxial stage 28. Even if the camera unit 30 is moved, the camera unit 30 is not always stationary at the optimal imaging position where the center of the imaging screen of the image sensor built in the imaging unit 35 coincides with the center of the chart image.

  A centering (tracking) program for automatically moving the camera unit 30 to the optimum imaging position is prepared in the sequence memory 41 so as to cope with such a case. As shown in FIG. 4, when a chart image 62 of a predetermined pattern (cross pattern in the illustrated example) is captured on the imaging screen 60 of the image sensor indicated by a two-dot chain line, the imaging signal output from the image sensor is subjected to signal processing. It is converted to digitized image data through the circuit 50 and stored in the image data memory 51. The image data processing circuit 52 performs image processing calculation based on the stored image data, and inputs information representing the center position of the chart image 62 in the imaging screen 60 to the system controller 40.

  In accordance with the centering program stored in the sequence memory 41, the system controller 40 sends a drive signal to the actuator drive control IC 48 so that the center of the chart image (crosshair) 62 coincides with the center of the imaging screen 60, and the three-axis stage 28 is controlled. As a result, the camera unit 30 is moved and adjusted in a plane parallel to the focal plane of the lens 7 to be tested while keeping the imaging optical axis constant, and the imaging range 60 is moved accordingly, and finally a solid line. The center of the chart image 62 can be captured at the center of the imaging range 60 shown in FIG.

When the chart image 62 captured in the imaging range 60 at the initial stage of moving the camera unit 30 to the target position is only one of the vertical line or horizontal line of the cross line, or the chart image 62 is in an extremely out of focus state. When the image is taken, the direction in which the camera unit 30 is tracked may be unclear. In this regard, the initial target position of the camera unit 30 is determined on the reference optical axis K corresponding to the origin with respect to the XY coordinate plane, and the optical axis direction (Z-axis direction) is within the defocus range described above. Therefore, in most cases, the center of the chart image 62 can be included in the imaging range 60, and the centering process is performed because the chart image 62 is within the focus range that can be identified with respect to the reference optical axis K direction. There will be no problems.

  When the center of the chart image 62 is captured at the center of the imaging screen 60 as shown by a solid line in FIG. 4, the camera unit 30 captures the chart image 62 as the first measurement image, and the obtained imaging signal is digitized. The measured image data is stored in the image data memory 51. Subsequently, the camera unit 30 is stepped by a fixed pitch P in the Z-axis direction by the three-axis stage 28, and the next centering process is started. When the center of the chart image 62 is captured at the center of the imaging screen 60 by the centering process, the measurement image is captured, and the measurement image data is stored in the image data memory 51. In this way, by moving the camera unit 30 sequentially to five points within the defocus range, and performing centering and imaging, respectively, five types of measurement image data of the chart image 62 in which the focus is slightly shifted can be obtained.

  The five types of measurement image data stored in the image data memory 51 are sequentially analyzed by the image data processing circuit 52. For example, as shown in FIG. 5, the image contrast C is such that the vertical line and the horizontal line of the chart image 62 are respectively crossed along measurement lines 63a and 63b that are equidistant from the center of the imaging screen 60 in the vertical and horizontal directions. Measured. Although it is desirable to perform measurement on both of the measurement lines 63a and 63b, the measurement image data obtained from each of the measurement lines 63a and 63b are handled in exactly the same way. Henceforth, based on the measurement image data obtained from the measurement line 63a. explain.

  As shown in a graph of an example of the measurement data in FIG. 5, the pattern of the measurement data for contrast C differs depending on the focus state of the chart image 62. In an ideal focus match state, a rectangular pattern as shown by a solid line is obtained, and the peak value of contrast C is maximized, and the peak value decreases and becomes a mountain-shaped pattern with a spread at the bottom as the degree of defocus increases.

  FIG. 6 shows an example of the obtained measurement data. FIG. 6A shows measurement image data representing the contrast C when the camera unit 30 is positioned and imaged at the front end of the defocus range, and the measurement data. The graphs showing the frequency characteristics of the MTF values obtained by performing FFT transform (Fast Fourier Transform) from the image data are shown side by side. FIG. 5B shows measurement data when the camera unit 30 is moved backward by a constant pitch P, and FIG. 6C shows measurement data when the camera unit 30 is further moved backward by a constant pitch P. FIG. Since the front end of the defocus range is always determined as a position where the camera unit 30 captures the chart image 62 in the front pin state, in most cases, the direction of FIG. However, the focus is good. In the illustrated example, a clearer imaging state is obtained when the camera unit 30 is further moved backward by one pitch P.

  As described above, the camera unit 30 is stepped four times at a constant pitch P from the front end to the rear end of the defocus range set in the Z-axis direction, and each of the five measurement positions after performing the centering process. By imaging, five types of measurement image data with different imaging states are obtained as shown on the left side of FIGS. 6 (A), (B), and (C). The obtained measurement image data is sequentially stored in the image data memory 51, the contrast C is measured by the image data processing circuit 52 for each measurement image data, and the calculation process by the MTF calculation unit 54 is performed, and the constant frequency F1 is obtained. The MTF values M1 to M5 are calculated. In this case, an appropriate value for the frequency F1 is selected according to the optical specifications of the lens 7 to be examined.

  The MTF value evaluation data graphed as shown in FIG. 7 can be obtained by plotting the respective MTF values at the constant frequency F1 thus obtained for each of the measurement positions Z1 to Z5 represented by the Z-axis coordinates. it can. As can be seen from this graph, even if the camera unit 30 is not moved to the best focus position ZP at which the MTF value reaches its peak, interpolation is performed based on the data obtained at five discrete points, and the best focus position is obtained. Can be identified as the coordinate value of the Z axis, and the peak MTF value MP can also be obtained. In the case of the illustrated example, the coordinate value of the best focus position ZP is the value of the drive pulse supplied from the actuator drive control IC to the drive unit 32 when the camera unit 30 is moved in the Z-axis direction by a constant pitch P within the defocus range. Assuming that the number is T and the MTF peak value MP is obtained when U (<T) drive pulses are supplied from the Z-axis coordinate Z3, the coordinate value ZP is obtained as “Z3 + (U / T) P”. .

  Note that if the set defocus range is too wide or the number of step movements of the camera unit 30 within the range is increased too much, the time required for measurement becomes longer and the measurement efficiency decreases. Conversely, if the defocus range is too narrow or the number of movement steps is reduced too much, the best focus state cannot be obtained within the defocus range, or the reliability when obtaining the best focus position by interpolation calculation is low. Become. Therefore, the moving range (defocusing range) of the camera unit 30 at the time of MTF measurement and the number of step movements or the moving pitch P within the moving range are set in advance in consideration of individual differences of the lens 7 to be tested. It can be set or changed as appropriate.

  In measurement using the on-axis collimator 14, in parallel with the centering process, the camera unit 30 is moved in parallel with the reference optical axis K within a preset defocus range, and chart images are captured. The digitized data at the position is captured. Further, when the measurement light is incident from the direction inclined from the reference optical axis K using the off-axis collimators 15 and 16, the incident optical axis of the measurement light is in a horizontal plane passing through the center of the lens 7 to be measured. Therefore, in order to evaluate the axial symmetry of the test lens 7, the test lens 7 is measured while being rotated together with the lens mount 6. In this lens inspection machine, since the pair of off-axis collimators 15 and 16 are provided at symmetrical positions with respect to the reference optical axis K, the lens mount 6 can be rotated within a range of 180 ° by using them in combination. Thus, it is possible to make the measurement light incident on the lens 7 to be measured from an angular range of 360 ° around the optical axis.

  When the off-axis measurement is continued after the on-axis measurement is finished, the imaging optical axis is maintained parallel to the reference optical axis K after the camera unit 30 is moved to the front end of the defocus range. In this state, the camera unit 30 is moved in the X-axis direction by the three-axis stage 28 to be removed from the reference optical axis K, and the imaging state of the chart image off the axis is measured. The measurement sequence in this case is the same as the measurement on the reference optical axis K, and after the camera unit 30 is moved to the off-axis target position, the camera unit is within the same defocus range as the centering process and on-axis measurement. Five types of measurement image data are sequentially obtained while 30 is sent at a constant pitch P in the rear pin direction parallel to the reference optical axis K. Based on the measurement image data thus obtained, the MTF value is calculated in the same manner as shown in FIGS.

  By the way, when the measurement is performed while moving the camera unit 30 on the reference optical axis K using the on-axis collimator 14, the camera unit 30 moves off from the reference optical axis K after moving to the first target position. There is hardly anything. This is because the camera unit 30 is centered in a feedback manner while tracking the center of the chart image 62 imaged at a high magnification by the camera unit 30 in the imaging screen, so that the XY coordinates orthogonal to the reference optical axis K are obtained. Even if there is movement in the plane, it is extremely small and within a negligible range.

  In contrast, when the camera unit 30 is moved to the off-axis target position in order to perform off-axis measurement, for example, after the camera unit 30 is moved to the front end of the defocus range on the reference optical axis K, The camera unit 30 moves greatly from the reference optical axis K in the X-axis direction corresponding to the off-axis target position. The movement process in the X-axis direction is performed by inputting an X coordinate value from the system controller 40 to the actuator drive control IC 48 and driving the three-axis stage 28 with the number of drive pulses corresponding to the X coordinate value. However, the amount of movement of the camera unit 30 at this time is significantly larger than the amount of movement during the centering process described above.

  However, in reality, it is extremely difficult to position all three axes X, Y, and Z of the three-axis stage 28 with respect to the upper surface of the surface plate 2 and the reference optical axis K with high accuracy. When the bottom surface of the triaxial stage 28 is used as a reference surface for positioning, the X axis and the Z axis are set within the accuracy range with no practical problem by matching the bottom surface with the upper surface of the surface plate 2. Although it is possible to make the Y-axis parallel to the upper surface of the plate and the Y-axis perpendicular to the upper surface of the surface plate 2, it is often required for a lens inspection machine only by measuring and adjusting the mechanical position. Therefore, it is very difficult to align the Z axis parallel to the reference optical axis K to such an extent that the accuracy to be satisfied is satisfied.

  For these reasons, an inclination error of several seconds to several tens of seconds or more remains between the reference optical axis K and the Z axis of the triaxial stage 28. Correspondingly, an XY coordinate plane of the triaxial stage 28 corresponds to this. Is inclined with respect to the reference optical axis K, and even when the camera unit 30 is moved only in the X-axis direction of the triaxial stage 28, the camera unit 30 is moved in the reference optical axis K direction. Since the amount of movement in the direction of the reference optical axis K increases in proportion to the amount of movement in the direction of the X-axis, if the light is moved about several millimeters from the on-axis measurement position to the off-axis measurement position as described above, the reference light It may move 10 μm or more in the axis K direction. As a result, when the camera unit 30 is stopped at the front end of the defocus range of the off-axis target position, the degree of defocus increases and the camera becomes out of focus, so that proper centering cannot be performed in normal sequence processing. It may also decrease.

  In view of this, in this lens inspection machine, the deviation from the target position, which is likely to occur when the camera unit 30 is moved to the target position by the three-axis stage 28, can be easily performed before measuring the lens 7 to be calibrated. The correction coefficient can be obtained by simple operation. In order to obtain the correction coefficient, as shown in FIG. 8, a check plate 65 serving as a focus check jig is first attached to the lens mount 6 to which the lens holder 8 is attached.

  As with the lens holder 8, the check plate 65 has a ring 66 that functions as a mounting member for the lens mount 6, and a display fixed to the inside thereof. Plate 67. When the ring 66 is tightly screwed to the lens mount 6, the rear end surface of the ring 66 comes into contact with the mount reference surface of the lens mount 6 in the same manner as when the lens holder 8 is fixed, and the display plate 67 is perpendicular to the reference optical axis K. And is positioned at a fixed position on the reference optical axis K.

The back surface of the display plate 67 is a pattern display surface on which a design pattern for focus check is printed. As an example of the design pattern, a pattern indicated by a solid line in FIG. 9 is used. The design pattern has a plurality of cross-hair regions where vertical lines and horizontal lines intersect, a central cross-line pattern of the center portion 67a, a peripheral cross-line pattern of the peripheral portion 67b horizontally away from the central portion 67a, and a center And a peripheral cross line pattern of the peripheral part 67c that is separated from the part 67a in the vertical direction. The central cross line pattern of the central portion 67a is covered by the imaging screen 60 of the image sensor 68 of the camera unit 30 on the reference optical axis K, and the peripheral cross line patterns of the peripheral portions 67b and 67c are based on the camera unit 30. When moved in the horizontal and vertical directions from the optical axis K, the image sensor 68 covers the image and captures the images individually. It is possible to make the symbol pattern of the central portion 67a and the symbol patterns of the peripheral portions 67b and 67c different from each other, but it is desirable that they are the same as described above.

  A correction coefficient for correcting the movement of the triaxial stage 28 is obtained by the processing procedure shown in FIG. When the correction coefficient calculation processing sequence prepared in the sequence memory 41 is started after mounting the check plate 65 as described above, the camera unit 30 operates the Z axis passing through the origin of the XY coordinate plane by the operation of the triaxial stage 28. The Z-axis coordinate moves to the target position set with reference to the back surface (pattern display surface) of the display plate 67. The thicknesses of the ring 66 and the display plate 67 of the check plate 65 are known, their mounting positions are uniquely determined with respect to the reference optical axis K, and the back surface of the display plate 67 is regarded as the imaging surface of the lens 7 to be tested. Thus, the position in the Z-axis direction at the target position can be set in advance as the front end of the defocus range.

  After the camera unit 30 is moved to the target position in this way, the centering of the camera unit 30 described above is performed so that the central cross line of the symbol pattern on the back surface of the display plate 67 coincides with the center of the imaging screen 60 as shown in FIG. Is performed, and imaging is performed. Thereafter, in exactly the same way as in the measurement process described above, five types of measurement image data can be obtained by sequentially repeating the centering and imaging while the camera unit 30 is sent at a constant pitch P within the defocus range. . Furthermore, based on the measurement image data obtained in this way, the best focus position ZP0 can be obtained for the central cross line of the display plate 67 in accordance with the analysis procedure described in FIGS.

  If the Z-axis coordinate value of the best focus position ZP0 obtained in this way is “Z3 + (U0 / T) P”, this value is the X and Y coordinate values (X0, Y0) is stored in the measurement data memory 55 as the first position data. Subsequently, the best focus position ZP1 is obtained in the same procedure by performing centering and imaging for the peripheral cross line deviated in the horizontal direction from the central cross line in the pattern pattern on the back surface of the display plate 67. If the coordinate value of the best focus position ZP1 obtained for the peripheral cross line is “Z3 + (U1 / T) P”, this is similarly measured along with the X and Y coordinate values (X1, Y1) at that time. Is stored as second position data. When the check plate 65 is attached to the lens mount 6, it is desirable for the sequence processing that the center cross line and the peripheral cross line are arranged horizontally as shown in FIG. 9, but each cross line is slightly inclined. However, as long as the centering process is possible, the correction coefficient calculation process sequence can be used as it is.

  These first and second position data (X0, Y0, ZP0), (X1, Y1, ZP1) are read by the correction coefficient calculation unit 56 to calculate a correction coefficient α used for movement correction of the three-axis stage 28. Is done. As described above, if the Z-axis of the three-axis stage 28 is parallel to the upper surface of the surface plate 2 and only tilted with respect to the reference optical axis K, when the camera unit 30 is moved in the X-axis direction, Since the camera unit 30 also moves in the Z-axis direction along with the movement, it is necessary to prevent this movement in the Z-axis direction.

Since the movement amount of the camera unit 30 in the X-axis direction can be obtained as “X1-X0” and the movement amount in the Z-axis direction can be obtained as “ZP1-ZP0”, the unit movement amount in the X-axis direction Can be calculated as “(ZP1−ZP0) / (X1−X0)”. For example, in the above example, the amount of movement in the X-axis direction from the first position to the second position is 5 mm, the number of drive pulses T for moving by 50 μm, which is a constant pitch P, is 10, and U0 = 2, U1 When = 9, the correction coefficient α in the Z-axis direction with respect to the unit movement amount of the camera unit 30 in the positive X-axis direction is 7 × 10 ⁻³ .

After the correction coefficient α is obtained and written in the correction coefficient memory 57, the check plate 65 is removed from the lens mount 6, and the lens holder 8 holding the test lens 7 is attached to the lens mount 6 instead. Thereafter, the measurement of the test lens 7 is started according to the procedure already described, and first, the MTF measurement is performed while stepping the camera unit 30 in parallel in the defocus range on the reference optical axis K using the on-axis collimator 14. Done.

  Even during this MTF measurement, the camera unit 30 moves in the Z-axis direction of the three-axis stage 28. However, the centering process is performed so that the camera unit 30 is centered on the chart image 62 incident from the on-axis collimator 14. Feedback control is performed so as to capture at the center of the imaging screen 60. Therefore, when the check plate 65 is used first to obtain the correction coefficient α, the display plate 67 is correctly mounted so that the back surface of the display plate 67 is orthogonal to the reference optical axis K, and the optical axis of the test lens 7 is the reference light. When the lens holder 8 is mounted on the mount 6 so as to coincide with the axis K, there is no problem, and measurement can be performed while moving the camera unit 30 in parallel with the reference optical axis K. When the measurement in the defocus range on the reference optical axis K is finished, the camera unit 30 is moved to the front end of the defocus range on the reference optical axis K by the operation of the triaxial stage 28, and further determined by the first centering process. Set to the specified position.

  When the off-axis MTF measurement is subsequently performed using the off-axis collimator 15 or 16, the camera unit 30 is moved toward the off-axis target position. If the Z-axis of the triaxial stage 28 is parallel to the reference optical axis K, the front end of the defocus range at the off-axis target position coincides with the front end of the defocus range used during measurement on the axis. Therefore, thereafter, the measurement data can be sequentially obtained by performing imaging and step feed while centering the camera unit 30 at the off-axis position in exactly the same manner. However, as described above, when the camera unit 30 is largely moved in the X-axis direction by the three-axis stage 28, the distance from the imaging surface of the lens 7 to be tested perpendicular to the reference optical axis K according to the amount of movement. Is different from the front end position of the intended defocus range at the off-axis target position.

  Therefore, when the correction coefficient α is read from the correction data memory 57 by the system controller 40 and the camera unit 30 is moved in the X-axis direction by the three-axis stage 28, the camera unit 30 may also move in the Z-axis direction. The movement correction process is performed so as not to occur. When the camera unit 30 is moved, for example, 10 mm from the front end of the defocus position on the axis to the front end of the defocus range of the off-axis target position, the system controller 30 causes the actuator drive control IC 48 to “− (10 mm × An adjustment signal “α) = − 70 μm” is sent.

Accordingly, the actuator drive control IC 48 supplies 1 × 10 ³ drive pulses to the X-axis direction stepping motor of the drive unit 32 in order to move the 3-axis stage 28 by 10 mm in the X-axis positive direction. Seven drive pulses for moving the axis stage 28 in the negative direction of the Z-axis are supplied to the stepping motor for driving in the Z-axis direction of the drive unit 32. As a result, the camera unit 30 is moved to the front end of the off-axis defocus range without changing the distance to the imaging plane of the lens 7 to be examined perpendicular to the reference optical axis K. Accordingly, the chart image is hardly greatly out of focus even outside the axis, and the centering process can be started immediately.

  The above description is based on the premise that the Z-axis of the three-axis stage 28 is inclined in the horizontal plane with respect to the reference optical axis K, the X-axis is parallel to the upper surface of the surface plate 2, and the Y-axis is vertical. However, when these are not satisfied, the movement correction data reading process and the correction coefficient calculation unit are performed in the same procedure as shown in FIG. 56, it is necessary to obtain the correction coefficient in the Y-axis direction. In this case, of course, the correction coefficient calculator 56 also reads the number of drive pulses in the Y-axis direction from the movement data memory 49. The imaging optical axis of the camera unit 30 may not be strictly parallel to the Z-axis of the triaxial stage 28, but at the time of imaging related to the calculation of the correction coefficient and at the time of imaging when measuring the lens to be measured. When the measurement image data is obtained, the imaging optical system of the camera unit 30 is quite common, so that there is almost no problem practically.

  As described above, in the present invention, when capturing the image data necessary for calculating the correction coefficient, the camera unit 30 used for measuring the test lens is used, and the procedure is also performed using a common sequence. Therefore, the cost burden in hardware and software is hardly increased. Further, by adopting such a method, accumulation of errors can be avoided, and the reliability at the time of moving the camera unit is not impaired. In carrying out the present invention, for example, the setting of the defocus range, the number of step movements within the defocus range, the number of drive pulses supplied to the triaxial stage when sending a constant pitch P, etc. It is possible to set appropriately according to the optical specifications of the analyzing lens.

  Furthermore, the present invention can be applied to any lens inspection machine that calculates the measurement data that can be imaged based on the obtained measurement image data by imaging the imaging state of the test lens using the camera unit, The present invention can be used not only for handling the MTF values described in the above embodiment, but also for an apparatus that performs performance evaluation based on, for example, CTF (Contrast Modulation Transfer Function). Further, the process of calculating the correction coefficient α using the check plate 65 is executed every time the sample of the lens 7 to be tested is taken into consideration in consideration of the measurement accuracy and the like, and the data in the correction coefficient memory 57 is rewritten each time. However, it may be performed for each lot of measurement or periodically.

  Further, when the check plate 65 is attached to the lens mount 6, it is convenient that the screw cut on the outer periphery of the ring 66 as described above is shared with the screw of the lens holder 8. If it can be mounted in a posture perpendicular to the reference optical axis K, the display plate 67 is held by another mounting member instead of the ring 66, and the mounting member is used by using a mounting portion provided separately on the lens mount 6. You may wear it. Even if the position of the display plate 67 on the reference optical axis K is deviated from the virtual image plane of the lens 6 to be examined, if the deviation is small, it can be absorbed by the defocus processing and is large. Even if it is a case, there is no problem because it can be obtained in advance.

2 Surface plate 6 Lens mount 7 Lens to be tested 8 Lens holder 14 On-axis collimator 15 and 16 Off-axis collimator 28 Three-axis stage 30 Camera unit 32 Drive unit 35 Imaging unit 40 System controller 54 MTF calculation unit 56 Correction coefficient calculation unit 57 Correction data Memory 60 Imaging screen 65 Check plate 67 Display plate 68 Image sensor

Claims (6)

It has a lens mount that aligns the optical axis of the mounted test lens with the reference optical axis, and positions and holds the test lens in the direction of the reference optical axis, and passes the measurement light from the test chart through the collimating lens with the lens mount. There is an inclination error with respect to the reference optical axis by the camera unit that is incident on the held test lens and the image of the test chart formed on the focal plane of the test lens is supported by a triaxial stage. In the camera unit movement adjustment method of a lens inspection machine that performs imaging while moving in a plane perpendicular to the imaging optical axis and parallel to the imaging optical axis, and inspecting the lens to be inspected based on the obtained imaging signal,
Prior to mounting the test lens, a first step of positioning a display plate on which a pattern pattern for focus check is displayed on the lens mount so as to be perpendicular to the reference optical axis;
The position of the camera unit where the central portion of the symbol pattern is sequentially imaged while moving the camera unit on the imaging optical axis by the three-axis stage, and the central portion of the symbol pattern is imaged in an optimally focused state. A second step of taking in the first position data from the three-axis stage;
After the camera unit is moved in the in-plane direction perpendicular to the imaging optical axis by the triaxial stage, the periphery of the symbol pattern is sequentially imaged while the camera unit is moved in the imaging optical axis direction. A third step of capturing the position of the camera unit that is imaged in the optimum focus state as the second position data from the three-axis stage;
Based on the first position data and the second position data, a ratio between the movement amount of the camera unit in the in-plane direction perpendicular to the imaging optical axis and the movement amount of the camera unit in the reference optical axis direction is calculated. A fourth step of obtaining a correction coefficient to be expressed and storing it in a memory,
When the camera unit is moved after the lens to be tested is mounted on the lens mount instead of the focus check jig, the operation of the triaxial stage is controlled based on the correction coefficient read from the memory. A method for adjusting the movement of a camera unit of a lens inspection machine.   The central portion of the symbol pattern includes a central cross line pattern, and in the second step, the camera unit is orthogonal to the imaging optical axis by operating the triaxial stage while imaging the central cross line pattern with the camera unit. The reference position data is captured after the camera unit is centered by controlling the operation of the three-axis stage so that the center of the center cross pattern coincides with the center of the screen of the camera unit. 2. The camera unit movement adjustment method for a lens inspection machine according to claim 1, wherein imaging is performed while moving the camera unit on the imaging optical axis.   The peripheral part of the symbol pattern includes a peripheral cross line pattern. When the camera unit is moved in the in-plane direction perpendicular to the imaging optical axis in the third step, the camera unit captures the peripheral cross line pattern. While operating the three-axis stage, the camera unit is moved in a plane orthogonal to the imaging optical axis, and the three-axis stage is operated so that the center of the peripheral cross line pattern coincides with the center of the screen of the camera unit. 3. The camera unit movement adjustment method for a lens inspection machine according to claim 2, wherein the centering of the camera unit is performed by controlling the camera unit.   In the second step and the third step, when performing imaging while moving the camera unit on the imaging optical axis by the triaxial stage, the triaxial stage moves the camera unit at a constant pitch each time imaging is performed. The camera unit movement adjustment method for a lens inspection machine according to claim 1, wherein the movement is performed. It has a lens mount that aligns the optical axis of the mounted test lens with the reference optical axis, and positions and holds the test lens in the direction of the reference optical axis, and passes the measurement light from the test chart through the collimating lens with the lens mount. There is an inclination error with respect to the reference optical axis by the camera unit that is incident on the held test lens and the image of the test chart formed on the focal plane of the test lens is supported by a triaxial stage. It is used in a lens inspection machine that performs imaging while moving in a plane perpendicular to the imaging optical axis and parallel to the imaging optical axis, and inspects the test lens based on the obtained imaging signal. A focus check jig attached to the lens mount for measuring an inclination error of the imaging optical axis with respect to the reference optical axis prior to inspection,
A display plate with a plurality of design patterns on one side;
A mounting member that holds the display plate and is attached to the lens mount such that the one surface is perpendicular to the reference optical axis;
The plurality of symbol patterns include a central symbol pattern written at a position whose center coincides with the reference optical axis, and a peripheral symbol pattern written at a certain distance from the central symbol pattern to the peripheral side. The focus check jig is characterized in that the plurality of design patterns are individually imaged when the camera unit is moved in a plane perpendicular to the imaging optical axis on the three-axis stage. The focus check jig according to claim 5, wherein the plurality of design patterns are the same pattern.

Images