JPH05292514A - Image pickup device - Google Patents

Abstract

PURPOSE:To obtain image pickup device capable of picking up a high resolution color image, which omits the registration adjustment and is not influenced by positioning accuracu of each image pickup element at the time of manufacture. CONSTITUTION:As for the image pickup device, a ray of light of an object to be photographed is divided into R, G and B images by a color separation prism 3, and a position relation of each image pickup element is detected by a memory coefficient calculating part 13, with a correlation calculation, and based on the position relation, an image signal of each image pickup element is detected by undergoing interpolating calculation by an interpolating circuit 8. For instance, with a position of the G image as a refenrece, the misalignment amount of the R image and the B image is detected, and the misalignment of the image is corrected by the interpolating calculation. Then, each image is synthesized, an image of the object is superposed and the registration is executed, and the superposing area (image pickup area) of the RGB images is formed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image pickup apparatus using a plurality of image pickup elements.

[0002]

2. Description of the Related Art Generally, in a color image pickup device, a subject light is separated into a plurality of color components by a color separation system, and the subject light of each color component is picked up by each of different image pickup devices.
There is a so-called three-plate type color camera, which is a device for forming a television signal such as a C system. In this device, the subject images of the respective color components picked up by the image pickup elements are superposed,
Creating a color image. In such a color image pickup device, it is necessary to perform positioning of each image pickup element, that is, registration adjustment, and accurate positioning accuracy.

That is, in the case where three image pickup devices are used to pick up images of red (R), green (G) and blue (B) respectively, FIG.
If the registration of the image sensor is not performed correctly, the combined image becomes a so-called color shift image in which the images of the respective colors are shifted as shown in FIG. Here, in order to remove the color shift, it is generally necessary to perform mechanical registration adjustment of each image pickup device. For example, as described in JP-A-61-288683, the adjustment directions in the registration adjustment are as shown in FIG. 15: (1) horizontal center adjustment (2) vertical center adjustment ( 3) Back focus adjustment (4) Horizontal tilt adjustment (5) Vertical tilt adjustment (6) Rotation adjustment 6 axis directions.

Such conventional adjustment is based on the difference between the pixel pitch and the repeating pitch by utilizing a repeating pattern having a repeating pitch having a predetermined positional relationship with the pixel pitch of the image sensor for registration adjustment. While observing the beat component with the oscilloscope, the adjustments of the above items (1) to (6) are performed so as to maximize the amplitude.

[0005]

However, in the above-mentioned registration adjustment, the operator requires a complicated operation and a great amount of adjustment time one by one while looking at the oscilloscope.
Further, since the registration adjustment is a mechanical adjustment, the time required for the same work varies depending on the degree of skill of the operator. Therefore, there is a problem that the number of steps required for the registration adjustment described above affects the manufacturing time of the color image pickup device, resulting in high cost.

Therefore, an object of the present invention is to provide an image pickup apparatus capable of picking up a high-resolution color image which is not affected by the positioning accuracy of each image pickup element at the time of manufacture by omitting registration adjustment.

[0007]

In order to achieve the above-mentioned object, the present invention provides a photographing optical means for picking up a subject image and a light beam separating means for separating the subject light captured by the photographing optical means into a plurality of light fluxes. Means, an image pickup means composed of a plurality of image pickup elements for picking up the subject light separated by the light beam separating means, and position shift information between two image pickup elements of the image pickup elements of the image pickup means, Positional deviation detection means for calculating a coefficient used in interpolation, storage means for storing the coefficient calculated by the positional deviation detection means, and the position of one of the image pickup elements according to the stored value stored in the storage means Corresponding to the interpolating means for calculating the image signal of the position of the other image pickup device by using the interpolation calculation, and the image information synthesizing means for synthesizing the plurality of image information output from the image pickup means and the interpolating means. It Providing that the imaging device.

[0008]

In the image pickup apparatus having the above-mentioned configuration, the positional relationship between the image pickup elements is detected by the correlation calculation instead of the mechanical registration adjustment, and the image signal of each image pickup element is interpolated based on the detected positional relationship. It is detected by calculation. That is, the positional shift amount of the R image and the B image is detected with the position of the G image as a reference, and the positional shift of the image is corrected by interpolation calculation. Then, the respective images are combined, the subject images are overlapped and registration is performed, and an overlapping area (imaging area) of the RGB images is formed.

[0009]

Embodiments of the present invention will now be described in detail with reference to the drawings. A configuration of an image pickup apparatus as a first embodiment according to the present invention is shown in FIG. 1 and will be described.

In this image pickup apparatus, a subject image (subject light) 1 is incident on a color separation prism 3 for separating colors via a photographing optical system 2. This color separation prism 3
For example, it is composed of a dichroic mirror and decomposes the subject image into light of three primary colors (red, green, blue).

The separated subject light of each color component is photoelectrically converted into an image signal by the image pickup device. The image pickup device is, for example, a CCD solid-state image pickup device or the like, and includes a red component image pickup device 3r, a green component image pickup device 3g, and a blue component image pickup device 3.
The image pickup device detects b, and these are driven by the CCD driver 16.

The outputs of the image pickup devices 3r, 3g, 3b are amplified by the preamplifiers 4r, 4g, 4b, and the A /
It is converted into a digital signal by the D converters 5r, 5g, 5b. Furthermore, a signal processing circuit (Signal Processor: SP) 6
Signal processing such as γ correction and edge enhancement is performed by r, 6g, and 6b, and the processed image signal is stored in the frame memories 7r, 7g, and 7b.

Further, the frame memories 7r, 7g, 7b
The image signals read from the image signals are input to the interpolation circuits 8r and 8b, and the image signals of the R and B images corresponding to the respective positions of the G images are interpolated according to the coefficients read from the coefficient memories 9r and 9b, which will be described later. To be done. The interpolation circuits 8r and 8b are shown in FIG.
, The interpolation circuit 8r will be described here for simplification of description.

The interpolation circuit 8r is a data read circuit 2
1 and a linear interpolation calculation circuit 22. The data read circuit 21 reads pixel values V _b , V _c , V _d , and V _{e at} four positions from the frame memory 7r according to the corresponding coordinates (IC _x , IC _y ) of the image read from the coefficient memory 9r.
Read. The linear interpolation calculation circuit 22 then interpolates the read pixel values V _b , V _c , V _d , and V _e from the coefficient memory with interpolation coefficients C _b , C _c , and C, respectively.
Interpolation is performed by multiplying _d and C _e by the multipliers 23, 24, 25 and 26, and adding by the adder 27 to calculate the pixel value V _a . That is, V _a = C _b V _b + C _c V _c + C _d V _d + C _e V _e (1)

The R and B image signals interpolated in this way are
It is input to a PS (Parallel Serial) conversion circuit 10 together with the G image signal, and is combined as a color image, for example, an NTSC television signal, and is monitored by a monitor 11, a printer 12,
It is output to the filing device 20. In this image pickup apparatus, the system controller 17 causes the CCD driver 16, the frame memories 7r, 7g, 7b, the memory coefficient calculators 13r, 13b, the coefficient memory 9r,
The drive of 9b and the PS circuit 10 is controlled.

Next, the memory coefficient calculator 13 shown in FIG.
r and 13b will be described. Here, the memory coefficient calculators 13r and 13b are both configured by the correlation circuits 14r and 14b and the coefficient calculation circuits 15r and 15b, and only the memory coefficient calculator 13r will be described for simplification of description.

First, for the interpolation coefficient, the correlation circuit 14r detects the parallel vector s and the rotation vector r as the positional deviation detection, and then the coefficient calculation circuit 15r.
Thus, the coefficients C _b , C _c , C _d , and C _e used for the interpolation calculation are calculated from the vectors r and s.

In the position shift detection, the position shift is detected separately for parallel movement and rotational movement at an arbitrary position in the image. As shown in FIG. 3A, reference areas a ₁ , a ₂ , a ₃ , a ₄ for detecting a positional deviation in the G image are shown.
To set. However, the center position of each area is p ₁ , p ₂ ,
Let p ₃ and p ₄ .

These areas are at target positions with the position c as the center, and are separated from the position c by a distance of k pixels. In addition, as shown in FIG. 3B, G is included in the R and B images.
The image area _{a 1, a 2, a 3} , a search area b ₁ for searching a position corresponding to _{a 4, b 2, b 3} , b 4 are set. Then, each of the areas a ₁ , a ₂ , a
Displacement vectors V ₁ , V ₂ , V ₃ , V corresponding to ₃ , a ₄
Detect ₄ This displacement vector is represented by V1 = vector s + vector r (2a) V2 = vector s + vector r-90, as shown in FIG. 4, by the rotation vector r and the parallel vector s at the position p _{1 about the} position c. (2b) V3 = vector s-vector r (2c) V4 = vector s + vector r + 90 (2d) where r-90 and r + 90 are r −90 degrees and +9, respectively.
Represents a vector rotated by 0 degrees. The rotation vector r is represented by the following equation. Vector r = k · tan (θ) (3) Here, θ represents a rotation angle. From equation (1), vector s = (V ₁ + V ₂ + V ₃ + V ₄ ) / 4 (4) Vector r = (V ₁ + V ₂ −V ₃ −V ₄ ) / 2 (5) It is possible to detect the displacement amount of the parallel movement and the rotational movement. The rotation angle θ of the rotational movement is obtained by θ = tan ⁻¹ (vector r / vector k) (6). FIG. 5 shows the vector s and the rotation angle θ.
The structure of the correlation circuit 14 for detecting

In the correlation circuit 14, the correlator 30
Thus, the correlation calculation between the reference area a ₁ and the search area b ₁ is performed. Similarly, the correlator 31 performs the correlation calculation between the reference area a ₂ and the search area b ₂ , the correlator 32 performs the correlation calculation between the reference area a ₃ and the search area b ₃ , and the correlator 33 performs the correlation calculation between the reference area a ₄ and the search area b _4. Then, displacement vectors V ₁ , V ₂ , V ₃ and V ₄ are detected respectively. Various methods have been conventionally proposed for this correlation calculation, but here, the calculation is performed by comparing the sums of the absolute values of the differences.

The displacement vector is the SR detector 3
4 and the parallel vector s and the rotation vector r are detected from the equations (4) and (5). The rotation vector r is input to the θ detection circuit 35, and the rotation angle θ is calculated by the calculation shown in the equations (3) and (6).

Next, referring to FIG. 6, from the vector r and the rotation angle θ, the coefficients C _b , C _c and
The coefficient calculation circuit 15 for calculating C _d and C _e will be described.

First, the linear interpolation calculation will be described. As shown in FIG. 6, pixels B, C, D, E to A of known values
The pixel value of is calculated. The intersections of a straight line (broken line) drawn perpendicularly from the pixel A to the line segments BC and DE and line segments BC and DE are defined as F and G, respectively, and BF: FC = DG: GE = m: n, FA: AG = If p: q, the value V _{f of the} pixel F is V _f = (nV _b + mV _c ) / (m + n) (7) The pixel value V _g of the pixel G is V _g = (nV _d, + mV _e ) / (M + n) (8) Therefore, V _a is V _a = (qV _f + pV _g ) / (m + n) (9) Here, if the inter-pixel distance is “1”, m + n = p + q =
Therefore, V _a is V _a = q (nV _b + mV _c ) + p (nV _d, + mV _e ) = (1-p) (1-m) V _b + (1-p) mV _c + p ( 1-m) V _d + pmV _e (10) Then, by comparing the equation (10) with the equation (1), the interpolation coefficients are _Cb = (1-p) (1-m), _Cc = (1-p) m, and _Cd = p. (1-m), C _e = pm (11) Further, the coordinates of the pixel A are (C _x ,
C _y ), the coordinates of the pixels B, C, D and E are expressed by the following equations. Pixel B = (IC _x , IC _y ) Pixel C = (IC _x +1, IC _y ) Pixel D = (IC _x , IC _y +1) Pixel E = (IC _x +1, IC _y +1) (12) Here, IC _x is the integer part of IC _x , and IC _y is the integer part of C _y . By the way, the positions X _r and X _b on the R and B images corresponding to the position X _{g of} the G image are: X _r = R (θ _r ) · (X _g + S _r ) ... (13) X _b = R (θ _r ) ・ (X _g + S _b ) ... (14)

Can be obtained as Where S _r ,
S _b , θ _b , and θ _b are parallel vectors and rotation angles between the R and B images and the G image, respectively. Where X _r , X _g , X _b
Is a two-dimensional vector having xy coordinates as elements. Also, R (θ) is

[0025]

[Equation 1] Is. Next, the configuration of the coefficient calculator 15 is shown in FIG. 7 and will be described.

In FIG. 7, the coordinate conversion circuit 36 performs the operations of the equations (13) and (14) and outputs the corresponding coordinates C _x and C _y (real numbers) of the R and B images. The corresponding coordinates C _x and C _y are input to the integer circuits 37 and 38, respectively, and the coordinates IC _x and IC _y shown in FIG. 6 are output. Further, the subtractor 39 and 40, respectively, m (= C _x -
IC _x ) and p (= C _y −IC _y ) are output. Next, the coefficient calculator 41 calculates the interpolation coefficient from the values of the coefficients m and p based on the equation (11), and the interpolation coefficients C _b , C _c and C are calculated.
Output _d and C _e . FIG. 8 shows the structure of the coefficient memory 9r. The coefficient memory 9b is the coefficient memory 9r.
Since it is the same as the configuration of the above, only the coefficient memory 9r will be described.

A memory 5 for storing the interpolation coefficients C _b , C _c , C _d , and C _e output from the coefficient calculation circuit 15, respectively.
1, 52, 53, 55 and memories 56, 57 for storing the coordinate values IC _x , IC _y . The operation of the first embodiment configured as described above will be described.

The operation of the present embodiment is roughly divided into two, one of which detects the coefficient by using the memory coefficient calculation unit 13 shown in FIG. 1 and stores the data in the coefficient memories 9r and 9b. And the other is the second step of imaging a general subject. At the time of the second step, in the configuration of the image pickup apparatus, the memory coefficient calculation unit 13 is useful only for the calculation of the coefficient, and therefore it is not necessary in the second step and may be removed.

First, the first step will be described with reference to FIG. For example, it is assumed that a test chart of a monochrome image is captured on the subject 1. Here, a black-and-white image has a large correlation between RGB images, and the amount of positional deviation between the R image and the G image and between the B image and the G image can be accurately obtained.
Further, it is preferable that this test chart is a subject having as many spatial frequency components as possible so that the correlation signal can be correctly obtained at each position.

When the image pickup of the test chart 1 is started, the image pickup optical system 2 is operated by a distance measuring system and a photometric system (not shown).
And the exposure time of the image pickup devices 3r, 3g, 3b are adjusted. Then, the subject light is the color separation prism 3
Are separated into RGB images by the image pickup device 3
An image signal is generated by photoelectric conversion by r, 3g, and 3b.
This image signal is amplified by the preamplifiers 4r, 4g, 4b so that white balance is maintained, and converted into a digital signal by the A / D conversion circuits 5r, 5g, 5b.

Then, the SP circuits 6r, 6g and 6b respectively perform signal processing such as γ correction and contour enhancement, and are recorded in the frame memories 7r, 7g and 7b. And the correlation circuit 1
4r, the reference area a of the G image from the frame memory 7r
₁ , a ₂ , a ₃ , a ₄ are read, the search areas b ₁ , b ₂ , b ₃ , b ₄ are read from the R image, the parallel vector S _{r of the} R image and the G image, and the rotation angle θ _r. Is detected.
Next, the vectors S _r and θ _r are calculated by the coefficient calculation circuits 15g, 1
5r, the corresponding coordinates of the R image corresponding to the coordinates Xg of the G image IC _x , IC _y and the interpolation coefficients C _b , C _c ,
C _d and C _e are calculated. These values are stored in the coefficient memory 9
It is stored at a predetermined address of r.

Then, the corresponding coordinates IC _x , IC _y and C _b , C _c , C _d , C are defined as the predetermined range of the G image, which is the range (imaging area) where the RGB images shown in FIG. 9 overlap.
_e is detected and stored in the coefficient memory 9r. The image pickup area is determined by the system controller according to the output values of the correlation circuits 14r and 14b, but may be set in advance.

Similarly, by the memory coefficient calculator 13b, G
Positional deviation amount between the image and the B image is the detection, the corresponding coordinates IC _x of the imaging range of the B image, IC _y and the interpolation coefficient _{C b, C c, C d} , C e is calculated, stored in the coefficient memory 9g To be done. As described above, the interpolation coefficient for the image pickup area where the RGB images overlap is obtained, and the first step as the registration is completed.

The first step is a step performed when the image pickup apparatus is manufactured, and is completed before the apparatus is passed to a general user (photographer). Therefore, the memory coefficient calculation unit 13
Is no longer needed and is designed to be removable. Next, referring to FIG. 1, the imaging of a general subject, which is the second step, will be described. In this step, the memory coefficient calculation unit 13 is assumed to be removed.

The subject to be photographed is captured by the photographer in the image pickup area, and image pickup is performed by the image pickup devices 3r, 3g, 3b. In the subsequent steps, signal processing is performed in the same manner as in the first step, and the image signals picked up by the respective image pickup elements are transferred to the frame memory 7r,
It is stored in 7g and 7b.

Next, the system controller 17 specifies the coordinates of the G image in the image pickup range, and the G image at that coordinate is read from the frame memory 7g. And from the coefficient memory 9r, coefficients corresponding to the designated G image position, i.e., the corresponding coordinates IC _x, IC _y and the interpolation coefficient _{C b, C c, C d} , C e is read.

With the corresponding coordinates IC _x and IC _y , the R image data used for the interpolation calculation is read from the frame memory 7r, and the interpolation calculation is performed by the interpolation circuit 8r to correspond to the designated G image position. The pixel value of the R image is calculated.

Similarly, the pixel value of the B image corresponding to the designated G image position is also calculated, and the PS circuit 1
Output to 0. In the PS circuit 10, the monitor 11, the printer 12,
A television signal that should be output to the filing device 20 or the like is generated. From the above, in the image pickup apparatus of the present invention,
It is possible to provide a high-resolution image pickup device that does not cause color shift in a plate-type color image pickup device.

Since the color misregistration is corrected by utilizing the interpolation calculation, it is not necessary to move and adjust the image sensor itself as in the conventional example, and it is not necessary to perform complicated registration adjustment. Further, even if the positioning accuracy of each image sensor is poor, the image signal is corrected, so that a high-resolution color image can be captured.

Further, the registration adjustment is
The interpolation coefficient can be calculated at high speed because it is performed by processing an electric signal instead of the conventional mechanical adjustment. Further, in this embodiment, since the correlation calculation is performed between the R image and the G image and the B image and the G image, which have a strong correlation, the positional deviation amount can be detected with high accuracy.

Although the four areas shown in FIG. 3 are selected as the reference areas in this embodiment, it is also possible to select only a ₁ and a ₃ or a ₂ and a ₄ . However, the present invention is not limited to this, and more areas may be used. Further, although the linear interpolation calculation is used in the interpolation circuits 8a and 8b, the present invention is not limited to this, and a higher-order correlation calculation such as spline or SING interpolation may be used. Further, the memory coefficient calculation unit 13 may be detachably attached to the camera body by a connector or the like.

In the embodiment described above, two memory coefficient calculators were used to calculate the coefficients for the deviation correction between the R image and the B image, respectively.
It is also possible to use only one and switch between the R image and the B image.

Next, the configuration of an image pickup apparatus as a second embodiment of the present invention is shown in FIG. 10 and described. Here, the members of the second embodiment which are equivalent to the members of FIG. 1 are designated by the same reference numerals, and the description thereof will be omitted.

In the first embodiment described above, the coefficient memory 9
Although r and 9b are used, in the second embodiment, the memories 60 and 61 that store the vector r and the rotation angle θ are adopted, and the amount of memory to be stored is greatly reduced. However,
In this case, it is necessary to calculate the interpolation coefficient and the corresponding coordinates by the coefficient calculation circuits 15r and 15b in the second step, that is, when the general subject is imaged.

Next, FIGS. 11 and 12 show the structure of an image pickup apparatus as a third embodiment according to the present invention, which will be described. Here, FIGS. 11 and 12 show only the characteristic parts of the third embodiment, and the other constituent members are the same as those of the first embodiment. The third embodiment reduces the amount of memory to be stored in the coefficient memory 9 and has a processing speed comparable to that of the first embodiment.

As described above, in the first embodiment, a high-resolution color image can be picked up from a plurality of coordinates represented by real numbers ( _Cx , C) even _if the positioning accuracy of each image pickup element is poor.
_y ) The position A was obtained by interpolation. However, since the obtained coordinates of the position A are real numbers, innumerable interpolation coefficients C _b , C _c , C
_d and C _e were present.

Therefore, as shown in FIG. 13, one image is L ×
By dividing the block into L equal parts and using the interpolation coefficient of the center coordinates as a representative of the block, the required set of interpolation coefficients is L ² Become individual. Then, information about which block to use the interpolation coefficient is stored. That is, 1 to L ^{2 for} each block The numbers up to are given and stored as block numbers. In the configuration of the third embodiment, the coefficient calculation circuit 15 and the coefficient memory 9 are different from those of the first embodiment.

In the coefficient calculating circuit 15a of the third embodiment, a block number calculator 62 is provided instead of the coefficient calculator 41. This block number calculator 62 uses m,
The block number N is calculated from the value of p based on the following equation. N = [{m / (1 / L)} + 1] + [{q / (1 / L)} + 1] × L (16) where 0 ≦ m <1, 0 ≦ p <1

Further, the coefficient memory 9a of the third embodiment has the corresponding coordinates IC _x , IC output by the coefficient calculating circuit 15a.
Memories 63, 64, 6 for storing _y and coefficient N respectively
5, interpolation coefficients C _b , C _c , C _d corresponding to each block,
Memories 66, 67, 68, 69 for respectively storing C _e
It is composed of Here, the interpolation coefficient memory 66,
67, 68, and 69 are L ² respectively It is composed of a single storage memory, which is significantly reduced as compared with the first embodiment, which required several pixels in the imaging area.

From the above, the image pickup apparatus of the present invention adopts the interpolation circuit for the processing of the image signal to generate the image in which the positional deviation of the images picked up by the plurality of image pickup elements is corrected by the interpolation, and the image pickup element is picked up. It is possible to realize a high-precision color image pickup device which does not require mechanical registration adjustment and is low in cost. Further, the present invention is not limited to the above-described embodiments, and it goes without saying that various modifications and applications can be made without departing from the scope of the invention.

[0051]

As described in detail above, according to the present invention, registration adjustment for mechanical adjustment is omitted, and a high-resolution color image that is not affected by the positioning accuracy of each image sensor during manufacturing can be captured. An imaging device can be provided.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of an image pickup apparatus as a first embodiment according to the present invention.

FIG. 2 is a diagram showing a specific configuration of the interpolation circuit shown in FIG.

FIG. 3 (a) is a diagram showing a reference area for detecting a position shift in a G image, and FIG. 3 (b) is a view showing a position corresponding to a predetermined area in the G image. It is a figure which shows the search area of R and B image.

FIG. 4 is a diagram showing a displacement vector of the first embodiment.

FIG. 5 is a diagram showing a specific configuration of the correlation circuit shown in FIG.

FIG. 6 is a diagram showing predetermined pixels for explaining an interpolation calculation.

FIG. 7 is a diagram showing a specific configuration of the coefficient calculator shown in FIG.

8 is a diagram showing a specific configuration of the coefficient memory shown in FIG.

FIG. 9 is a diagram showing an imaging area in which RGB images of the present invention overlap each other.

FIG. 10 is a diagram showing a configuration of an image pickup apparatus as a second embodiment of the present invention.

FIG. 11 is a diagram showing a configuration of a coefficient calculation circuit of an image pickup apparatus as a third embodiment according to the present invention.

FIG. 12 is a diagram showing a configuration of a coefficient memory of an image pickup apparatus as a third embodiment according to the present invention.

FIG. 13 is a diagram showing divided blocks obtained by equally dividing one image into L × L.

FIG. 14 is a diagram showing a configuration of registration in which three image pickup devices are superimposed and displayed.

FIG. 15 is a schematic view showing an adjustment direction of conventional registration adjustment.

[Explanation of symbols]

1 ... Subject image (subject light), 2 ... Photographing optical system, 3 ... Color separation prism, 3r ... Red component image sensor, 3g ... Green component image sensor, 3b ... Blue component image sensor, 4r, 4g, 4b
... Preamplifier, 5r, 5g, 5b ... A / D converter, 6
r, 6g, 6b ... Signal processing circuit (Signal Processor: S
P), 7r, 7g, 7b ... Frame memory, 8r, 8b
... Interpolator, 9r, 9b ... Coefficient memory, 10 ... PS (Pa
rallel serial) conversion circuit, 11 ... Monitor, 12 ... Printer, 13 ... Memory coefficient calculator, 13r, 13b ... Memory coefficient calculator, 14r, 14b ... Correlation circuit, 15r, 1
5b ... Coefficient calculation circuit, 16 ... CCD driver, 17 ... System controller, 20 ... Filing device, 21 ...
Data read circuit, 22 ... Linear interpolation calculation circuit, 23, 2
4, 25, 26 ... Multiplier, 27 ... Adder, 30, 31,
32, 33 ... Correlator, 34 ... SR detector, 35 ... θ detection circuit, 51, 52, 53, 54, 55, 56, 57, 6
0, 61, 63, 64, 65, 66, 67, 68, 69
…memory.

Claims (1)

[Claims] 1. An image pickup optical means for picking up a subject image, a light beam splitting means for splitting the subject light captured by the image pickup optical means into a plurality of light fluxes, and an image of the subject light split by the light flux splitting means. An image pickup means including a plurality of image pickup elements for detecting the positional shift information between two image pickup elements of the image pickup elements of the image pickup means, and calculating a coefficient used in interpolation; A storage unit that stores the coefficient calculated by the shift detection unit, and an image signal at the position of the other image sensor corresponding to the position of the one image sensor is interpolated according to the storage value stored in the storage unit. An image pickup apparatus comprising: an interpolating unit that calculates by using an arithmetic operation; and an image information synthesizing unit that synthesizes a plurality of image information output from the image pickup unit and the interpolation unit.