JPH0628456B2 - Position measurement method for solid-state image sensor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Hereinafter, the present invention will be described in the following order.

A. Industrial application fields B. SUMMARY OF THE INVENTION C. Conventional Technology D. Problems to be Solved by the Invention E. Means for Solving Problems F. Action G. Example G-1 Configuration of Registration Position Adjusting Device (See FIG. 1) G-2 Configuration of Registration Measurement Test Chart (See FIGS. 2, 3, and 4) G-3 Procedure of Adjustment Operation (See FIGS. 5 to 12) Effect of Invention A. TECHNICAL FIELD The present invention relates to a charge coupled device (CCD) used as an image sensor in a solid-state imaging device.
The present invention relates to a position measuring method for a solid-state image sensor, which is applied when performing so-called registration adjustment or the like in which the solid-state image sensor is arranged at a predetermined position of the image pickup optical system such as e).

B. SUMMARY OF THE INVENTION The present invention, when performing so-called registration adjustment or the like for arranging a solid-state image sensor used in a solid-state image sensor at a predetermined position of an image pickup optical system, repeats in a predetermined relationship with a pixel pitch of the solid-state image sensor. Based on the difference between the picture element pitch and the repetition pitch, by imaging the stripe-shaped repetitive pattern having an inclination with respect to the direction of the picture element array of the solid-state imaging element The position shift of the solid-state image pickup device can be easily measured from the beat component.

C. 2. Description of the Related Art Generally, in a color image pickup device, image pickup light is separated into a plurality of color components by a color separation system, and a subject image of each color component is obtained by a solid-state image sensor such as CCD or an image pickup tube. A standard color television signal of the NTSC system or the like is formed and output from each image output. In a color image pickup apparatus using a plurality of image sensors, it is necessary to surely maintain a state in which subject images of respective color components picked up by the image sensors are superposed, that is, a so-called registration adjustment is accurately performed.

In the conventional so-called three-tube color video camra, a color separation three-color separation prism and three image pickup tubes are mechanically attached to a housing formed by die-casting, etc. The registration of each image pickup tube is adjusted by mechanically adjusting the position and further finely adjusting the image pickup screen position using the deflection system of the image pickup tube.

For such a three-tube color video camera, CC
In a solid-state color image pickup device that uses a solid-state image pickup device such as D as an image sensor, it is not possible to electrically adjust the image pickup screen position like an image pickup tube. Therefore, in each solid-state image sensor that picks up a subject image of each color component, A holder made of iron or the like is joined, and a holder is also joined to each light emitting part of the color separation prism of the color separation system.
For each image sensor, the adjustment direction is indicated by the arrow in FIG. 13, (1) Horizontal center (arrow ± X direction) adjustment (2) Vertical center (arrow ± Y direction) adjustment ( 3) Back focus (arrow ± Z direction) adjustment (4) Horizontal tilt (arrow ± RX direction) adjustment (5) Vertical tilt (arrow ± RY direction) adjustment (6) Rotation (arrow ± RZ direction) adjustment After adjusting the registration in the 6-axis directions, the holders are heat-welded by soldering or the like to attach and fix the solid-state image sensors to the emission surfaces of the color separation prism.

D. Problems to be Solved by the Invention In general, a solid-state image sensor such as a CCD cannot adjust the position of an electric image-pickup screen like an image pickup tube. Therefore, in a color image pickup apparatus using a plurality of solid-state image sensors, Therefore, it is necessary to maintain the registration of each image sensor with extremely high accuracy. Particularly, in the solid-state color image pickup device which adopts the so-called spatial pixel shift method to achieve high resolution, the above-mentioned ± X direction, arrow ± Y direction, ± Z direction, ± RX, ± RY direction and ± RZ It is necessary to secure a registration on the order of 1 μm in the six axial directions, and conventionally, it took a lot of time and effort to adjust the registration.

Therefore, an object of the method of the present invention is to enable the above-mentioned registration adjustment in the 6-axis directions to be performed with high accuracy by a simple operation.

E. Means for Solving the Problems In order to solve the above-mentioned problems, a solid-state image sensor position measuring method according to the present invention measures a relative positional deviation of a plurality of solid-state image sensors that capture the same image. In doing so, the light and shade are repeated at a repeating pitch that has a predetermined relationship with the pixel pitch of the solid-state image sensor, and an image of a striped repetitive pattern with an inclination with respect to the pixel array direction of the solid-state image sensor is displayed on each solid-state image sensor. An image is picked up, and a beat component for at least one cycle based on the difference between the picture element pitch and the repetition pitch included in the image pickup output signal obtained from each solid-state image pickup element is detected,
A beat component detected from the reference solid-state image sensor,
By detecting the phase difference with the beat component detected from the solid-state image sensor to be measured for measuring the position shift, the positional shift of the solid-state image sensor to be measured with respect to the reference solid-state image sensor in the horizontal direction and the vertical direction can be determined. It is characterized in that the phase difference is measured.

F. In the position measuring method of the solid-state image sensor according to the present invention, the light and shade are repeated at a repeating pitch that has a predetermined relationship with the pixel pitch of the solid-state image sensor, and the stripes are inclined with respect to the direction of the pixel array of the solid-state image sensor. An image pickup output obtained by picking up an image of a repeating pattern in the shape of a solid by the solid-state image sensor includes a beat component based on the difference between the pixel pitch and the repeating pitch. The amplitude of the beat component changes depending on the back focus of the image pickup optical system and the amount of tilt of the solid-state image pickup element, that is, the registration error in the ± Z direction and the ± RX, ± RY directions described above, and the phase of the beat component changes. It changes depending on the above-mentioned registration error in the ± X direction, the arrow ± Y direction, and the ± RZ direction. As a phase difference between the beat component included in the image pickup output signal of the reference solid-state image pickup device and the beat component included in the image pickup output signal of the measured solid-state image pickup device that measures the positional deviation, The displacement of the measured solid-state image sensor in the horizontal and vertical directions is measured.

G. Embodiment Hereinafter, one embodiment of a position measuring method for a solid-state image sensor according to the present invention will be described in detail with reference to the drawings.

The embodiment shown in FIGS. 1 to 12 is a three-plate CCD.
The present invention is applied to a registration adjusting device for performing registration adjustment of a color video camera.

G-1. Configuration of Registration Adjusting Device In the registration adjusting device in this embodiment, as shown schematically in FIG. 1, the master lens 20 has a registration measuring test chart 10 on which a predetermined test pattern is formed. It is arranged in front of. The registration measurement test chart 10
The image pickup light of the test pattern is guided to the color separation prism 30 of the three-plate CCD color video camera through the master lens 20, and the color separation prism 30 converts the image pickup light into three primary color lights R, G, B, that is, red. Component light R, green component light G, and blue component light B are separated into three CCs.
The D image sensors 31r, 31g, 31b are irradiated. Each CCD image sensor 31
The r, 31g, and 31b are attached to position adjusting jigs 40r, 40g, and 40b for performing six-dimensional alignment, respectively. Then, each CCD image sensor 31
r, 31g, and 31b are driven by a CCD drive circuit 51 that operates based on a reference signal provided by a reference signal generator 50, and an image, that is, a test, by the imaging light of each color component separated by the color separation prism 30. Image the pattern. Each CCD image sensor 31r, 31
The image pickup outputs obtained by g and 31b are output from the preamplifiers 41r, 41g, and 41b to the gate circuits 42r and 41r, respectively.
It is supplied to the matrix circuit 43 via 42g and 42b. Here, each of the gate circuits 42r, 42g, 42b
Is supplied with a gate pulse generated by the gate pulse generating circuit 52 that operates based on the reference signal.

The image output signal obtained by the matrix circuit 43 is
The high-resolution monitor 46 is supplied to the oscilloscope 44 for observing waves and also through the video signal processing circuit 45.
It is supplied to the microcomputer 48 via the digitizer 47.

G-2. Configuration of Registration Measurement Test Chart The above registration measurement test chart 10 comprises:
As schematically shown in FIG. 2, the CCD image sensor 3
Double the pitch 2τc for one picture element pitch τc
.Tau.p = 2.tau.c. +-.. DELTA., Which is shifted by .DELTA.p1 ...

Here, Δ is the value of the CCD image sensor 31 so that the image pickup output by the CCD image sensor 31 includes n beat components in one horizontal scanning period (1H) as shown in FIG. 4, for example. It is set to Δ = nτcτp / τo (n is an arbitrary number of 1 or more), where τo is the effective image size in the pixel array direction.

The registration measuring test chart 10 in this embodiment has a vertical striped repeating pattern PTsh in which light and shade are repeated along the horizontal center line Lsh, as the test pattern is schematically shown in FIG.
And two vertical striped repeating patterns PTah, PT which are arranged symmetrically with respect to the horizontal center line Lsh and which are parallel to each other.
bh and two horizontal striped repeating patterns PTav and PTbv which are symmetrically arranged with the center line Lsv in the vertical direction sandwiched therebetween.
What is formed is used.

Here, each of the vertical striped repeating patterns PTsh, PT
The ah and PTbh are formed so as to repeat the light and shade at the repeating pitch τph which has the relationship of the above-mentioned first equation with respect to the horizontal picture element pitch τch of the CCD image sensor 31.
In addition, the horizontal striped repeating patterns PTav, PTbv
Is the vertical pixel pitch τ of the CCD image sensor 31.
Repetition pitch τ that has the relationship of the above-mentioned first equation with respect to cv
It is formed to repeat the light and shade in pv.

In this embodiment, for the CCD image sensor of two types of picture element arrays corresponding to the CCIR and EIA standards, one test chart is shared so that registration measurement can be performed. , EIA standards are arranged in parallel.

Registration measurement test chart 1 as described above
The image pickup output obtained by the CCD image sensor 31 irradiated with the image pickup light of the test pattern by 0 is the repeating pitch τp of the repeating pattern PT and each pixel pitch τ.
A moire pattern is generated based on the difference from c, and the signal level changes as shown in FIG. 2 according to the relative position between the repeating pattern PT and each picture element S. That is, the imaging output for the horizontal repeating pattern PTh includes the fourth
As shown in the figure, one horizontal scanning period (1H) includes n beat components.

The amplitude of the beat component included in the imaging output signal obtained by imaging the repetitive pattern of the registration measuring test chart 10 by the CCD image sensor 31 has the amplitude of the back focus of the imaging optical system and the amount of tilt of the solid-state imaging device. That is, it changes depending on the registration error in the ± Z direction and the ± RX and ± RY directions described above, and the phase thereof changes in the center position and rotation of the solid-state image sensor, that is, in the ± X direction and the arrow ± Y.
Direction and ± RZ direction depending on the registration error.

G-3. Therefore, in this embodiment, in this embodiment, the CCD image sensors 31r, 31g, 31b mounted on the position adjusting jigs 40r, 40g, 40b of the registration adjusting device are mounted.
The driving state, the repeating pattern of the test chart 10 for installation measurement is imaged, and each CCD
With respect to the imaging output obtained by the image sensors 31r, 31g, 31b, each beat component is determined by the oscilloscope 4 described above.
Waveform observation by 4 or the above microcomputer 4
While performing waveform analysis with 8, position adjustment jigs 4
By operating 0r, 40g, and 40b, as shown in the flowchart of FIG. 5, a first adjustment operation for back focus adjustment, a second adjustment operation for tilt adjustment, and a coarse adjustment for center adjustment and rotation adjustment are performed. Each of the CCD image sensors described above in the order of a third adjustment operation to be performed, a fourth adjustment operation to perform rotation adjustment, a fifth adjustment operation to perform vertical center adjustment, and a sixth adjustment operation to perform horizontal center adjustment. Six-dimensional alignment is performed with respect to 31r, 31g, and 31b, and 1/2 offset adjustment of the CCD image sensor 31g for green image pickup is performed by the seventh adjustment operation.

In the first adjustment operation, the registration measurement test chart 10 is performed by the CCD image sensors 31r, 31g, 31b through the master lens 20.
The image output of the central portion AS of the vertical striped repeating pattern PTsh which repeats shading along the horizontal center line Lsh schematically shown in FIG. Each of the gate circuits 42r, 42g, 42b is selectively taken out.
That is, the image output of the central portion of the image pickup screen which is less affected by the axial aberration and the chromatic magnification aberration of the image pickup optical system is taken out. The amplitude of the beat component included in the image output of the central portion AS changes depending on the above-mentioned ± Z direction registration error, that is, the back focus of the image pickup optical system.
It becomes maximum when the CD image sensors 31r, 31g, 31b are in the just focus position.

Therefore, while observing the waveform of the beat component included in the image output of the central portion AS by the oscilloscope 44, the position adjusting jigs 40r, 40g, and 40b are operated, and the belly partial amplitude of the beat component is operated. By moving each CCD image sensor 31r, 31g, 31b to the position where
Position adjustment, that is, back focus adjustment, can be performed by setting r, 31g, and 31b to the just focus position.

The back focus adjustment is performed by setting the diaphragm of the image pickup optical system to an open state to reduce the depth of focus so that the CCD image sensors 31r and 31 are passed through the master lens 20.
The registration measurement test chart 10 is imaged by g and 31b.

In the next second adjustment operation, each CCD image sensor 31
With respect to r, 31g, and 31b, both ends Ba and Bb of the vertical striped repeating pattern PTsh that repeats shading along the horizontal center line Lsh of the registration measurement test chart 10 are symmetrical with the center line Lsh. Repeating pattern PT with two vertical stripes arranged in parallel with each other
Using central portions Cu and Cd of ah and PTbh, horizontal and vertical tilt adjustments are performed.

In the horizontal tilt adjustment operation in the second adjustment operation, an image obtained by imaging the repetitive pattern of the registration measurement test chart 10 by the CCD image sensors 31r, 31g, 31b via the master lens 20. Regarding the output signal, both sides B of the vertical striped repeating pattern PTsh in which light and shade are repeated along the horizontal center line Lsh schematically shown in FIG.
Image outputs of a and Bb are output to the respective gate circuits 42r, 42g, 4
2b selectively takes out.

The amplitudes of the beat components included in the image output of the both side portions Ba and Bb are the same as those of the CCD image sensors 31r and 31.
Horizontal amount of g and 31b, that is, ± RX described above
There is a level difference depending on the registration error.

Therefore, while observing the waveform of the beat component included in the image output of the both side portions Ba, Bb by the oscilloscope 44, the position adjusting jigs 40r, 40g, 40b are obtained.
By operating the CCD image sensors 31r, 31g, 3 until the position where the partial amplitude of the antinode of the beat component is maximized.
1b is moved in parallel so that each CCD image sensor 31 for both side portions Ba and Bb of the repeating pattern PTsh.
Just focus positions Pa, P of r, 31g, 31b
b is detected.

The position of the back focus adjusted CCD image sensor 31 obtained by performing the first adjustment operation is set as the initial position Po, and the initial position P is set as shown in FIG.
The amount of movement from 0 to each just focus position Pa, Pb is measured to obtain each measurement output Δa, Δb, and the tilt amount of the CCD image sensor 31 is calculated from each measurement output Δa, Δb, and Δa = Δb = 2 adjustments to make it 0
・ Repeat three times to adjust the tilt.

In the above-mentioned tilt adjustment, | Δa | = | Δb | does not always hold due to field curvature, but it is made to converge by repeating the adjustment.

Here, the center line L of the registration measuring test chart 10 used for the horizontal tilt adjustment.
Vertical striped repeating pattern P that repeats shading along sh
As shown in an enlarged view in FIG. 7, both side portions Ba and Bb of Tsh are repeatedly shaded at a repeating pitch having the relation of the above-mentioned formula 1 with respect to the horizontal pixel pitch of the CCD image sensor 31, and the CCD Repetitive patterns PTsh 'are formed in vertical stripes having an inclination with respect to the direction of the picture element array of the image sensor 31, that is, the horizontal direction.

Thus, the striped repeating pattern P having an inclination with respect to the direction of the picture element array of the CCD image sensor 31.
When Tsh 'is imaged by the CCD image sensor 31, the phase of the beat component changes between the horizontal lines Ln and Ln ± 1 adjacent in the vertical direction according to the inclination of the repetitive pattern PTsh'. Therefore, each of the gate circuits 42
By selecting a horizontal line with r, 24g, and 42b, it is possible to adjust so that the abdomen of the beat component can be reliably gated.

Further, the vertical tilt adjustment operation in this second adjustment operation is performed by each CCD image sensor 31r, 31g, 3
Regarding the imaging output signal obtained by imaging the repetitive pattern of the registration measurement test chart 10 by 1b, two vertical striped repetitive patterns P which are symmetrically arranged and parallel to each other with the horizontal center line Lsh therebetween.
While observing the waveforms of the beat components included in the image outputs of the central portions Cu and Cd of Tah and PTbh with the oscilloscope 44, the position adjusting jigs 40r, 40g, 4
By operating 0b, the same procedure as the horizontal tilt adjustment operation described above is performed.

In this embodiment, CC is also provided outside the central portions Cu and Cd of the two vertical striped repeating patterns PTah and PTbh of the registration measuring test chart 10.
Striped repetitive patterns PTah 'and PTbh' having an inclination with respect to the direction of the pixel array of the D image sensor 31 are provided.

The back plane adjustment and the tilt adjustment described above determine the imaging planes of the CCD image sensors 31r, 31g, 31b.

In the next third adjustment operation, each CCD whose image pickup plane is determined by the back focus adjustment and the tilt adjustment described above.
For the image sensors 31r, 31g, and 31b, the horizontal and vertical center lines Lsh and Lsv of the registration measurement test chart 10 are used to perform coarse adjustments of horizontal and vertical center adjustments and rotation adjustments. Like this.

That is, since the picture element located at the center of the image pickup surface of the CCD image sensor 31 is determined, a cross-shaped reference image centered on this picture element is displayed at the high resolution as shown by the solid line in FIG. The test chart 1 for registration measurement is displayed on the screen of the monitor 46.
0 is imaged by each CCD image sensor 31r, 31g, 31b, and is monitored in the horizontal direction of the registration measuring test chart 10 which is monitored on the screen of the high resolution monitor 46, for example, as shown by the broken line in FIG. And the position adjusting jigs 40r, 40 so that the images of the center lines Lsh, Lsv in the vertical direction match the cross-shaped reference image.
g, 40b.

By this third adjustment operation, each CCD image sensor 31
The horizontal and vertical centers and rotations of the r, 31g, and 31b are roughly adjusted to about ± 3 μm, and the CCD image sensors 31r, 31g, and 31b are halved to the registration measurement test chart 10. Registration adjustment is done within the bare pitch.

In the next fourth adjustment operation, regarding the imaging output signal obtained by imaging the repetitive pattern of the registration measuring test chart 10 by the CCD image sensors 31r, 31g, 31b through the master lens 20, the third Horizontal center line L schematically shown in the figure
The image outputs of the central portions Cu and Cd of the two vertically striped repeating patterns PTah and PTbh which are symmetrically arranged with the sh interposed therebetween are selectively taken out by the gate circuits 42r, 42g and 42b. Each beat component included in the image output of the central portions Cu and Cd on both sides has a phase difference corresponding to the registration error in the ± RZ directions, that is, the rotation error of the CCD image sensor 31.

Therefore, while observing the waveforms of the beat components included in the image outputs of the central portions Cu and Cd on both sides with the oscilloscope 44 as shown in FIG. Operate each CD image sensor 3 to the position where the phase of each beat component matches.
Rotation adjustment can be performed by rotating 1r, 31g, and 31b.

Here, the phase difference of each beat component included in the image output of the central portions Cu and Cd on both sides, that is, the rotation error, is roughly adjusted within 1/2 pixel pitch by the above-described third adjustment operation. Therefore, the rotation adjustment can be performed reliably and with extremely high accuracy by the fourth adjustment operation.

In the next fifth adjustment operation, the registration measurement test chart 10 is performed by the CCD image sensors 31r, 31g, 31b through the master lens 20.
Regarding the image pickup output signal obtained by picking up the image of the repetitive pattern, the two horizontal striped repetitive patterns PTav, PTbv which are symmetrically arranged and parallel to each other with the vertical centerline Lsv schematically shown in FIG. The image outputs of the central portions Da and Db of the above are selectively taken out by the respective gate circuits 42r, 42g and 42b. Each beat component included in the image output of the central portions Da and Db has a phase difference according to a vertical registration error of the CCD image sensor 31.

Therefore, while observing the waveforms of the beat components included in the image output of the central portions Da and Db on both sides with the oscilloscope 44, the position adjusting jigs 40r and 40 are adjusted.
g, 40b to operate the CCD image sensors 31r, 31g, until the positions where the phases of the beat components match each other.
By moving 31b, center adjustment in the vertical direction can be performed.

In the next sixth adjusting operation, the central portions Cu and C of the two vertical striped repeating patterns PTah and PTbh which are parallel to each other are provided.
Using d, similarly to the above-described vertical center adjustment, each position adjustment is performed while observing the waveform of each beat component included in the image output of the central portions Cu and Cd on both sides with the oscilloscope 44. The jigs 40r, 40g, and 40b are operated to move the CCD image sensors 31r, 31g, and 31b in the horizontal direction to the positions where the phases of the beat components coincide with each other, to adjust the center in the horizontal direction.

Each CCD image sensor 31r, 31g, 31b
Is (1) Horizontal center (arrow ± X direction) adjustment (2) Vertical center (arrow ± Y direction) adjustment (3) Back focus (arrow ± Z direction) adjustment (4) Horizontal tilt ( (+/- arrow direction) adjustment (5) Vertical tilt (arrow +/- RY direction) adjustment (6) Rotation (arrow +/- RZ direction) adjustment. The adjustment operation completes with extremely high accuracy.

Here, in this embodiment, the CCD image sensor 31 having a pixel pitch of τc forms a repetitive pattern PT which repeats light and shade at the repetitive pitch τp shown in the first formula provided in the registration measurement test chart 10. The beat component included in the obtained image pickup output is, as shown in an enlarged view of the waveform near the zero-cross position in FIG. 10, for each picture element of the image pickup output corresponding to each picture element S of the CCD image sensor 31. The signal level of changes continuously in the same direction twice at the zero-cross portion. Therefore, it is possible to detect the zero-cross position of the beat component, that is, the phase of the beat component by detecting the change state of the signal level of each pixel of the image output.

Therefore, in this embodiment, the signal of the beat component is supplied to the oscilloscope 44 at the time of performing the center adjustment and the rotation adjustment in which the adjustment work is performed by using the phase of the beat component included in the imaging output as described above. The signal level of the beat component is digitized by the digitizer 47 and supplied to the microcomputer 48 so that the signal level changes twice in the same direction in succession. The zero-cross position is calculated with high accuracy by interpolation processing from a plurality of signal level data corresponding to the picture element at the zero-cross position and picture elements near the picture element, and the calculation result is output to the printer 49. is there.

Further, in this embodiment, when the registration adjustment of the color video camera adopting the spatial picture element shifting method is performed, it is possible to perform 6 by performing the above-mentioned first to sixth adjustment operations.
For each CCD image sensor 31r, 31g, 31b whose axial registration has been adjusted, the seventh image of the procedure shown in the flowchart of FIG. 11 is utilized by utilizing the image pickup output of the central portion AS of the vertical stripe repeating pattern PTsh.
And the beat component included in the imaging output captured by the CCD image sensor 31g for green imaging,
The CCD image sensor 31g for green image pickup is set so that the beat components included in the image pickup output picked up by the other CCD image sensors 31r and 31b have a phase difference of 90 ° as shown in FIG. Adjust 1/2 offset by moving horizontally.

In this seventh adjustment operation, first, for example, CC for red imaging
The digitizer 47 digitizes the signal level of the beat component of the imaging output that can obtain the registration measurement test chart 10 by the D image sensor 31r, and the central portion AS of the vertical stripe repeating pattern PTsh.
The zero-cross position of the beat component included in the image pickup output is calculated by the microcomputer 48 from the signal level data.

Next, the CCD image sensor 31g for green image pickup is roughly adjusted by 1/2 offset in a predetermined direction, and the registration measuring test chart 10 is imaged by the CCD image sensor 31g for green image pickup.

Further, the CCD image sensor 31g for green image pickup
The digitized signal level of the beat component of the imaging output of the central portion AS of the vertical striped repetitive pattern PTsh obtained by the above is digitized by the digitizer 47 and supplied to the microcomputer 48. For the beat component included in the image pickup output of the CCD image sensor 31g for use in the image pickup, the first zero-cross position is started from the previously calculated zero-cross position of the beat component of the CCD image sensor 31r for image pickup, Calculated from

The phase of the beat component of the CCD image sensor 31g for green image pickup, that is, the zero-cross position calculated in this way is 90 with respect to the phase of the beat component of the CCD image sensor 31r for red image pickup, that is, the zero-cross position.
It is determined whether or not the phase difference of °, that is, the offset is 1/2. When the zero-cross position of the beat component of the green image pickup CCD image sensor 31g is not within the desired adjustment accuracy range, the green image pickup CCD image sensor 31g is finely adjusted, and again By calculating the zero-cross position of the beat component with respect to the imaging output obtained by the CCD image sensor 31g for green imaging, determining the zero-cross position, and performing the fine adjustment operation until a desired adjustment accuracy range is obtained, CCD image sensor 31g for green image pickup
1/2 offset adjustment is completed.

That is, in this embodiment, in the first adjustment operation, the center portion AS of the vertical striped repeating pattern PTsh that repeats light and shade along the horizontal center line Lsh of the registration measuring test chart 10 is used for backing up. Adjust the focus. In the second adjustment operation, horizontal tilt adjustment is performed using both side portions Ba and Bb of the vertical striped repeating pattern PTsh, and symmetrically arranged parallel to each other with the horizontal center line Lsh interposed therebetween. Cu and C at the center of each of the two vertical striped repeating patterns PTah and PTbh
Vertical tilt adjustment is performed using d. In the third adjusting operation, the horizontal and vertical center lines Lsh and Lsv of the registration measuring test chart 10 are utilized to perform the horizontal and vertical center adjustments and the coarse rotation adjustments. In the fourth adjustment operation, rotation adjustment is performed by using the central portions Cu and Cd of the two vertically striped repeating patterns PTah and PTbh which are symmetrically arranged with the horizontal center line Lsh interposed therebetween. In the fifth adjusting operation, the center portions Da and Db of two horizontal striped repeating patterns PTav and PTbv which are symmetrically arranged with the center line Lsv in the vertical direction of the registration measuring test chart 10 interposed therebetween are used. And adjust the center in the vertical direction. In the sixth adjustment operation, center adjustment in the horizontal direction is performed using the central portions Cu and Cd of the two vertical striped repeating patterns PTah and PTbh that are parallel to each other. In the seventh adjustment operation, 1/2 offset adjustment of the CCD image sensor 31g for green image pickup is performed using the central portion AS of the vertical striped repeating pattern PTsh.

H. EFFECTS OF THE INVENTION As is apparent from the above description of the embodiments, in measuring the relative positional deviation of a plurality of solid-state image pickup devices that capture the same image of the solid-state image pickup device according to the present invention, the pixels of the solid-state image pickup device are measured. The light and shade are repeated at a repetitive pitch that has a predetermined relationship with the pitch.An image of a striped repetitive pattern having an inclination with respect to the pixel array direction of the solid-state image sensor is captured by each solid-state image sensor and obtained from each solid-state image sensor. A beat component detected from a reference solid-state image sensor, and a beat component detected from a reference solid-state image sensor. By detecting the phase difference with the beat component detected from the measurement solid-state image sensor, the horizontal direction and the vertical direction of the solid-state image sensor to be measured with respect to the reference solid-state image sensor. The positional deviation in the direct direction can be used as the above-mentioned phase difference and can be measured with high accuracy by a simple operation, and the intended purpose can be sufficiently achieved.

[Brief description of drawings]

1 to 12 are drawings showing an embodiment of a registration adjusting device to which the present invention is applied, and FIG. 1 is a block diagram schematically showing the configuration of the registration adjusting device of this embodiment. FIG. 2 is a schematic diagram for explaining the relationship between the test pattern of the registration measurement test chart and the pixel array of the CCD image sensor in the above embodiment, and FIG. 3 is the resist used in the above embodiment. FIG. 4 is a schematic diagram showing the configuration of a registration measurement test chart, and FIG. 4 is a waveform diagram showing a beat component included in an imaging output obtained by imaging a test pattern of the registration measurement test chart with a CCD image sensor. FIG. 5 is a flow chart showing the operation procedure of registration adjustment in the above embodiment, and FIG. FIG. 7 is a schematic diagram for explaining the tilt adjustment operation of the CCD image sensor in the above embodiment, and FIG. 7 is an enlarged CCD image of a part of the repetitive pattern provided in the registration measurement test chart in the above embodiment. FIG. 8 is a schematic diagram showing the relationship with the pixel array of the sensor, FIG. 8 is a schematic diagram for explaining a rough adjustment operation of the registration of the CCD image sensor in the above embodiment, and FIG. FIG. 11 is a waveform diagram showing a beat component used in a rotation adjustment operation of the CCD image sensor in the example, FIG. 10 is an enlarged waveform diagram near the zero cross position of the beat component, and FIG. 6 is a flowchart showing a procedure of 1/2 offset adjustment of the CCD image sensor in the embodiment. Figure above 1/2
It is a wave form diagram which shows the beat component used for offset adjustment. FIG. 13 is an external perspective view of a CCD image sensor for explaining an adjusting operation direction in general registration adjustment. 10 ... Test chart 20 ... Master lens 30 ... Color separation prism 31r, 31g, 31b ... CCD image sensor

Claims (1)

[Claims] 1. When measuring the relative positional deviation of a plurality of solid-state image pickup devices for picking up the same image, the solid-state image pickup device is repeatedly shaded at a repeating pitch having a predetermined relationship with a pixel pitch of the solid-state image pickup device. An image of a striped repetitive pattern having an inclination with respect to the pixel array direction is captured by each solid-state image sensor, and the difference between the pixel pitch and the repetitive pitch included in the imaging output signal obtained from each solid-state image sensor. A beat component for at least one cycle based on the following, and a beat component detected from the reference solid-state image sensor,
By measuring the phase difference with the beat component detected from the solid-state image sensor to be measured for measuring the position shift, the horizontal and vertical position shift of the solid-state image sensor to be measured with respect to the reference solid-state image sensor can be determined. A method for measuring the position of a solid-state imaging device, characterized in that the phase difference is measured.