JP3018101B2 - Color imaging device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color imaging device, and more particularly to an imaging device in which a light receiving element has an offset sampling structure.

[Conventional technology]

FIG. 16, FIG. 18, and FIG. 20 are diagrams showing an example of an arrangement configuration of color filters of a conventionally known color solid-state imaging device. In FIG. 16, the green light transmitting filter (hereinafter referred to as “Gr filter”) is vertically striped,
A red light transmitting filter (hereinafter referred to as “Rd filter”) and a blue light transmitting filter (hereinafter referred to as “Bl filter”) are arranged between the Gr filters in every two rows and two columns.

In FIG. 18, the magenta light transmitting filter (hereinafter referred to as “Mg filter”), the Gr filter, the cyan light transmitting filter (hereinafter referred to as “Cy filter”), and the yellow light transmitting filter (hereinafter referred to as “Ye filter”) are arranged in the horizontal direction. Two pixels,
Eight color filters of four pixels in the vertical direction are set as one unit and arranged in the order shown in the figure.

Further, FIG. 20 shows a solid-state image sensor having an offset sampling structure as described in Japanese Patent Application No. 1-24433, in which Rd, Gr, and Bl filters each have a horizontal pitch of 3 pixels and a vertical pitch of 3 pixels. It is arranged in an offset sampling structure with one pixel and an offset amount of 1.5 pixels in the horizontal direction.

FIGS. 17, 19, and 21 correspond to FIGS. 16, 18, respectively.
FIG. 21 is a characteristic diagram of the first quadrant when the color light carrier in the color filter array configuration of FIG. 20 is represented by a two-dimensional frequency plane (f _H , f _V ). In each case, the pixel pitch in the horizontal direction is
p _H , the pixel pitch in the vertical direction is p _V, and 0 ≦ f _H ≦ 1 / p _H , 0 ≦
It represents the range of f _V ≦ 1 / 2p _V. In each of the figures, the arrows indicate carriers of each color, the length of the arrow indicates the size of the carrier, and the direction indicates the phase relationship.

In this specification, “direction” indicates one direction,
"Direction" is two directions, one direction and 180 degrees different from this.
One direction.

In FIG. 17, the colored light carriers are, in addition to (0,0),
_{(1 / 2p H, 0)} , (1 / p H, 0), (0,1 / 4p V), (1 / 2p H, 1 / 4p
_V), it is generated in the _{(1 / p H, 1 /} 4p V). Of these, (0,
(0) and (1 / p _H , 0) are carriers generated for achromatic light and cause aliasing distortion, and the other carriers completely cancel each other for achromatic light. Although it disappears, it does not disappear for chromatic light and causes aliasing distortion.

Similarly, in FIG. 19, the color light carrier is (0,0)
Other than (1 / 2p _H , 0), (1 / p _H , 0), (1 / 2p _H , 1 / 4p _V ),
_{(0,1 / 2p V), (} 1 / 2p H, 1 / 2p V), (1 / p H, 1 / 2p V) to occur, these and _{(0,0) (1 / p H} , 0) is a carrier generated for achromatic light, and the other carriers are extinguished for achromatic light.

Furthermore, in the FIG. 21, when the offset amount in the horizontal direction of the solid-state imaging device and p _H / 2, the color light carrier (0,
0), (2 / 3p _H , 0), (1 / 3p _H , 1 / 2p _V ), (1 / p _H , 1 /
2p _V ), of which (0,0) and (1 / p _h , 1 / 2p _V ) are carriers generated for achromatic light, and the others disappear for achromatic light. Career. The advantage of such an offset sampling structure is that the rectangular sampling structure shown in FIGS. 16 and 18 has a horizontal frequency of (1 / p _H , 0) at a frequency of achromatic light. Since there is a carrier generated, f _H = 1 / 2p _H becomes the Nyquist frequency, and no higher frequency component can be obtained, and f _H = 1 / f
But the horizontal resolution obtained only up 2p _H, the offset sampling structure shown in FIG. 20, the horizontal direction (1 / p _H,
The position of 0) no carriers generated against achromatic light, that can be a f _H = 1 / p _H and the Nyquist frequency is generally known. Thus, despite the same sampling pitch as the rectangular sampling structure, the horizontal resolution can be obtained up to multiples of f _H = 1 / p _H.

[Problems to be solved by the invention]

However, a general subject is not always achromatic and generally has a color.
The aliasing distortion occurs from the color light carriers shown at all positions in the figure, and it is very unsightly in some scenes. For this reason, harmful colored light carriers must be cut using an optical low-pass filter or the like in the imaging optical system, resulting in a reduction in resolution.

In addition, in the color solid-state imaging device having the offset sampling structure shown in FIG. 20, since a color light carrier is generated at the position of (2 / 3p _H , 0), a frequency component of f _H = 2 / 3p _H or more is horizontally generated. since the required optical low pass filter for cutting, horizontal resolution only a place where the horizontal resolution to the original free if color is f _H = 1 / p _H obtained to its two-thirds of the f _H = 2 / 3p _H Can not get.

The present invention has been made in order to solve such a problem, and obtains a horizontal resolution up to f _H = 1 / p _H even with a horizontal pixel pitch p _H and a vertical pixel pitch p _V. It is an object of the present invention to provide a color image pickup apparatus which has good resolution and less occurrence of moiré.

[Means for solving the problem]

In order to achieve the above object, in the present invention, a color imaging device is configured as in the following (1) and (2).

(1) An image sensor having a pixel having an offset sampling structure in which a horizontal pitch is P _H , a vertical pitch is P _V , and a horizontal offset is P _H / 2, and each pixel of the image sensor is A color filter array composed of four types of color filters provided in the same manner, wherein each type of color filter has a horizontal pitch of 2P _H , a vertical pitch of 2P _V and a horizontal offset amount of P _H In a color image pickup device comprising a color filter array having a certain offset sampling structure, the output of each pixel provided with the four types of color filters is 3
A conversion means for converting into an RGB signal by performing an operation using a matrix of four rows and A = (a _ij ) (i = 1,2,3; J = 1,2,3,4), The matrix A has the following formulas (i) to
(Iii) A color imaging apparatus comprising a conversion unit that satisfies any one of the above.

(I) a _i1 + a _i3 = a _i2 + a _i4 (i = 1,2,3) (ii) a _i1 + a _i4 = a _i2 + a _i3 (i = 1,2,3) (iii) a _i1 + a _i2 = a _i3 + a _i4 (i = 1,2,3) (2) In the above (1), provided in the imaging optical system, the incident light beam is converted into an angle of θ counterclockwise with respect to the scanning direction of the imaging device. An optical low-pass filter that divides light into two light beams separated by a distance D in a direction that forms an angle of θ in a clockwise direction with respect to a direction formed or a direction opposite to the direction of scanning of the imaging element, and A color imaging device comprising an optical low-pass filter satisfying the following.

However, [Operation] With the configurations of (1) and (2) above, a resolution up to f _H = 1 / p _H is obtained, and the occurrence of moire is reduced.

〔Example〕

 Hereinafter, the present invention will be described in detail with reference to Examples.

FIG. 1 is a diagram showing an arrangement of color filters in a “color imaging device” according to a first embodiment of the present invention. As shown in the figure, the solid-state imaging device has a horizontal pixel pitch.
It has an offset sampling structure of p _H , a pixel pitch p _{V in} the vertical direction, and a pixel offset amount p _H / 2 in the horizontal direction. Each of the magenta, green, cyan, and yellow light transmission filters Mg, Gr, C
A color filter array in which y and Ye are arranged at positions corresponding to each pixel is provided.

Each color filter, horizontal 2p _H, vertical 2p _V, the offset amount in the horizontal direction is offset sampling structure of p _H.

The color light carrier in such a color filter arrangement is 2
Dimension frequency plane (f _H, f _V) by a characteristic diagram Figure 2 expressed in, shows a range of the first quadrant of _{0 ≦ f H ≦ 1 / p} H, 0 ≦ f V ≦ 1 / 2p V . In the figure, the color optical carrier other than (0,0), (1 / p H, 0), (1 / 2p H, 1 / 4p V), (0,1 / 2p V), (1 /
p _H , 1 / 2p _V ), of which (0,0) and (1 / p _H ,
1 / 2p _V ) is a carrier generated for achromatic light,
Other carriers are carriers that disappear with respect to achromatic light.

As is apparent from the figure, no color light carrier is generated up to the position of (1 / p _H , 0) in the horizontal direction, so that it is possible to obtain a frequency component up to f _H = 1 / p _H. it can. That is, the color image pickup device in the present embodiment has a horizontal resolution of the color image pickup device shown by the sampling structure in FIG.
A horizontal resolution up to 1.5 times f _H = 1 / p _H can be obtained.

Furthermore, the color light carrier closest to the origin is (1 / 2p _H , 1 / 4p _V ), which has a sufficient distance to the origin, and for achromatic light, Since the color light carrier disappears, no significant aliasing distortion occurs for the horizontal and vertical frequency components.

As described above, the color image pickup apparatus according to the present embodiment has a high resolution and less occurrence of moire.

Next, the signal processing of the color imaging device included in the color filter shown in FIG. 1 will be described. FIG. 3 shows a block diagram thereof.

In the CCD sensor 1, which is a solid-state image sensor, a color filter array including four types of color filters as shown in FIG. 1 is arranged. The pixel signal read out for each pixel from the sensor 1 is gain-adjusted by the AGC 2 and then A / D-converted by the A / D (analog-digital) converter 3 at a timing synchronized with the read-out clock. For color processing to be performed later, the A / D converter 3 has a good linear characteristic, and is preferably performed with 8 bits or more from the viewpoint of quantization error.

Luminance signal, after interpolating the offset sampling structure two-dimensionally by the interpolation filter 4 is detected high frequency components in the high-pass filter 5, a low-frequency component Y _L of the luminance obtained in the manner described below adder 6 , D / A converted by a D / A (digital-analog) converter 7, and output.

On the other hand, the output of the A / D converter 3 has four interpolation filters 8,
The color signals Mg, Cy, Ye,
Gr.

By controlling the order of the horizontal lines input to the interpolation filters 8, 9, 10, 11 and 4 by the clock, it is possible to obtain an output signal by interlace scanning and an output signal by non-interlace scanning.

The color signals synchronized by the interpolation filters 8 to 11 are RG
The signal is input to the B conversion unit 12 and converted into R, G, B3 signals. This is based on the following matrix operation.

... (1) where the matrix Is the spectral characteristic Mg (λ), Gr of Mg, Gr, Cy, Ye of the sensor 1.
(Λ), Cy (λ), Ye (λ) are defined by the NTSC RGB
Is a matrix of 3 rows and 4 columns optimized to approximate the ideal spectral characteristics R (λ), G (λ), and B (λ). here, And Now, if the gain in black and white of an object is adjusted to a constant, Mg, Gr, Cy, baseband component of Ye is on the frequency space function α (f _H, f _V) with, Mg (f _{H, f V) = Gr (} f H, f V) = Ye (f H, f V) = Cy (f H, f V) = α (f H, f V) ...... (3) and reveal it Can be. Here, it is assumed that the base band components of each color are sufficiently band-limited by two-dimensional filters such as interpolation filters 8 to 11. In this case, the carrier component at the position of FIG. 2 _{(1 / 2p H, 1 /} 4p V) _{are, Mg (f H, f V} ) = Cy (f H, f V) = α (f H, f _{V), Gr (f H,} f V) = Ye (f H, f V) = α (f H, f V) from the ... (4), the carrier component at this point in the RGB signals,
From equations (1), (2), (3), and (4), they are expressed as follows.

_{R (f H, f V)} = (a 11 -a 12 + a 13 -a 14) α (f H, f V), G (f H, f V) = (a 21 -a 22 + a 23 -a 24 _{) α (f H, f V} ), B (f H, f V) = (a 31 -a 32 + a 33 -a 34) α (f H, f V) ...... (5) In this case, the matrix In each row, if the sum of the coefficients in the first and third columns is equal to the sum of the coefficients in the second and fourth columns, ie, a _i1 + a _i3 = a _i2 + a _i4 (i = 1,2,3 ) if Naritate relationship that ... (6), the carrier component of the RGB signal at this point _{(1 / 2p H, 1 /} 4p V) is reduced, thus, the carrier component of the color signal at this point does not occur. Similarly, a point (−) in the third quadrant symmetrical to the origin with respect to FIG.
1 / 2p _H, can be eliminated by satisfying the carrier component also (6) of the relationship of the color signal at -1 / 4p _V).

FIG. 4 shows positions where carriers of the luminance signal and the chrominance signal occur near the origin at this time. (±
Since the carriers at 1 / 2p _H , ± 1 / 4p _V ) (in the same order as the double sign) have disappeared, the aliasing distortion to the baseband is reduced as compared with the case where the relationship of the expression (6) is not satisfied.

In this way, the RGB conversion unit 12 shown in FIG. 3 converts the respective signals of Mg, Gr, Cy, and Ye into RGB signals with reduced aliasing distortion.

Next, the white balance unit 13 converts the RGB signals into R and R based on the color temperature information obtained from the white balance sensor 18.
By converting G, B into αR, G, βB, white balance is obtained.

Next, the γ conversion unit 14 performs γ conversion on the RGB signal by table conversion.

In the color difference matrix section 113, Performs conversion was in the standard of the NTSC system of the low-frequency component Y _L and the color difference signals R-Y of the luminance described above, and the B-Y are generated. The color difference signals R-Y and B-Y are converted by the D / A converters 16 and 17
/ A Converted and output. As described above, the low-frequency component Y _L of the luminance is added to the high-frequency component of the luminance detected by the high-pass filter 5 by the adder 6, D / A converted by the D / A converter 7, and output.

Although the present embodiment may be configured in a hard-wired manner according to the block diagram, a DSP (digital signal processor) may be used.
And the like, and may be configured by software.

As described above, in the color imaging apparatus according to the first embodiment, a carrier component causing aliasing occurs at the position shown in FIG. In order to minimize aliasing distortion in such a color image pickup apparatus, as shown in FIG. 5, the carrier (1 / 2p _H , ± 1 / 4p) closest to the origin is located in front of the sensor 1.
_V ) An optical low-pass filter 50 for trapping the vicinity of this point is required to prevent aliasing distortion from (in the same order as decoding).

As described above, the image pickup optical system according to the first embodiment further provided with an optical low-pass filter will be described as a second embodiment of the present invention. As shown in FIG. 6, the optical low-pass filter 50 converts the incident light beam into two light beams separated by a distance D in a direction clockwise forming an angle θ with respect to the direction opposite to the scanning direction. And satisfies the following conditions.

However, It is. When the value of D exceeds the lower limit of the inequality (7), the aliasing distortion increases, and when the value exceeds the upper limit, the resolution decreases.

In this embodiment, And In this way, the optical low-pass filter 50 traps the position shown by the dotted line in FIG. Therefore, not only (1 / 2p _H , ± 1 / 4p _V )
Since the carrier of the color signal at the positions of (± 1 / p _H , 0) and (0, ± 1 / 2p _V ) is also trapped, it is possible to minimize the aliasing distortion of the color signal and obtain a favorable image quality. .

It should be noted that the same characteristics can be obtained even when the light beam splitting direction of the optical low-pass filter 50 is a point symmetrical direction as shown in FIG. 6, that is, a direction that forms an angle of θ clockwise with respect to the scanning direction. The optical low-pass filter 50 may be constituted by a birefringent plate using a uniaxial crystal such as quartz, or any other material such as a prism having a property of dividing an incident light beam into two light beams. Is also good.

Further, as shown in FIG. 8, the incident light beam is separated by a distance of 45 ° clockwise with respect to the direction opposite to the scanning direction. Other optical low-pass filter 81 consisting of birefringent plate is divided into two light beams separated by the birefringent plate for dividing the light beam incident on the two light beams parallel apart p _H / 2 and the scanning direction An optical low-pass filter 80 configured by combining an optical low-pass filter 82 composed of At this time, the spatial frequency characteristic of the optical low-pass filter 80 traps the position indicated by the dotted line in FIG. Thus the color signal carrier _{(1 / 2p H, ± 1} / 4p V) not only trap (double signs in same _{order), (± 1 / p H} , 0), (0, ± 1 / 2p V) The carrier of the color signal having the position is also trapped in the vicinity thereof, so that sufficient suppression is performed. Furthermore, (± 1 / p _H , ± 1/2
Since the carrier of the luminance signal in (p _V ) (arbitrary decoding) is also trapped, a good image in which aliasing distortion is well suppressed can be obtained.

Next, a third embodiment of the present invention will be described. This embodiment differs from the first embodiment only in the matrix of the RGB conversion unit.

2-dimensional frequency plane (f _H, f _V) of the color light carrier of the color solid-state imaging device shown in FIG. 1 that expresses the second quadrant on a Figure 10. In the figure, the color photocarriers are (0,
0), (−1 / p _H , 0), (−1 / 2p _H , 1 / 4p _V ), (0,1 / 2p
_V ) and (−1 / p _H , 1 / 2p _V ), of which (0,
(0) and (−1 / p _H , 1 / 2p _V ) are carriers generated for achromatic light, and the others are carriers generated for chromatic light. These are has assumed symmetrical with respect to FIG. 2 and f _V axis, - they have different code only for each color of the carrier as the _{(1 / 2p H, 1 /} 4p V).

These color signals Mg, Gr, Cy, Ye are converted to an RGB converter shown in FIG.
When converting to an RGB signal at 110, (-1 / 2p _H , 1/4
Using the equation (3), the carrier component at the position of (p _V ) is given by: Mg (f _H , f _V ) = Ye (f _H , f _V ) α (f _H , f _V ), Gr (f _H , f _V ) _{V) = Cy (f H,} f V) α (f H, f V) from the ... (10), the carrier component at this point in the RGB signals (1), (2), (3), From equation (10), it is expressed as follows.

_{R (f H, f V)} = (a 11 -a 12 -a 13 + a 14) α (f H, f V), G (f H, f V) = (a 21 -a 22 -a 23 + a 24 ) Α (f _H , f _V ), B (f _H , f _V ) = (a ₃₁ −a ₃₂ −a ₃₃ + a ₃₄ ) α (f _H , f _V ) (11) , If the sum of the coefficients in the first and fourth columns is equal to the sum of the coefficients in the second and third columns, ie, a _i1 + a _i4 = a _i2 + a _i3 (i = 1,2,3 If the relationship (12) is established, the carrier component of the RGB signal at this point (−1/2 p _H , 1/4 p _V ) disappears, and therefore the carrier component of the color signal at this point does not occur. Similarly, the point in the fourth quadrant (1/1 /
2p _H , −1 / 4p _V )
2) It can be eliminated by satisfying the relationship of the expression.

FIG. 11 shows positions where carriers of the luminance signal and the chrominance signal are generated near the origin at this time.

As in the first embodiment, in order to minimize aliasing distortion in such a color imaging apparatus, as shown in FIG.
The carrier closest to the origin before sensor 1 (± 1 / 2p _H , ±
1 / 4p _V) (to prevent aliasing from decoding the same order) is required optical low-pass filter 50 to trap near this point.

The fourth embodiment of the present invention will be described in which the image pickup optical system according to the third embodiment is further provided with an optical low-pass filter. As shown in FIG. 12, the optical low-pass filter 50 divides an incident light beam into two light beams separated by a distance D in a direction at an angle of θ counterclockwise with respect to the scanning direction. Which satisfies the following conditions.

However, It is. When the value of D exceeds the lower limit of the inequality (7), the aliasing distortion increases, and when the value exceeds the upper limit, the resolution decreases.

In this embodiment, And In this way, the optical low-pass filter 50 traps the position shown by the dotted line in FIG. Therefore, not only (± 1 / 2p _H , ± 1 / 4p _V )
Since the carrier of the color signal at the positions of (± 1 / p _H , 0) and (0, ± 1 / 2p _V ) is also trapped, it is possible to minimize the aliasing distortion of the color signal and obtain a favorable image quality. .

Also, the light separation direction of the optical low-pass filter 50 is
The same characteristics can be obtained even when the direction is point-symmetrical to that shown in FIG. 12, that is, the direction forms an angle θ counterclockwise with respect to the direction opposite to the scanning direction. The optical low-pass filter 50 may be constituted by a birefringent plate using a uniaxial crystal such as quartz, or any other material such as a prism having a property of dividing an incident light beam into two light beams. Is also good.
Further, in order to obtain the same effect as the optical low-pass filter of the first embodiment shown in FIG. 8, the light incident on the optical low-pass filter 81 is rotated counterclockwise by 45 ° with respect to the scanning direction. Distance in angle direction Optics consisting of a birefringent plate that splits the incident light into two light beams parallel to the scanning direction and separated by _pH / 2. The optical low-pass filter 80 combined with the dynamic low-pass filter 82 may be used. At this time, the spatial frequency characteristic of the optical low-pass filter 80 traps the position indicated by the dotted line in FIG. Therefore, the color signal carrier (± 1 / 2p
_H , ± 1 / 4p _V ) (double sign same order)
Carriers of the color signals at the positions (± 1 / p _H , 0) and (0, ± 1/2 p _V ) are also trapped in the vicinity thereof, so that sufficient suppression is performed. Furthermore, since the carrier of the luminance signal at (± 1 / p _H , ± 1/2 p _V ) (arbitrary decoding) is also trapped, a good image in which aliasing distortion is well suppressed can be obtained.

 Further, a fifth embodiment of the present invention will be described.

The carrier component of the color signal at the positions of (1 / p _H , 0) and (0,1 / 2p _V ) in FIG. 2 can be expressed by the following equation (3): Mg (f _H , f _V ) = Gr a _{(f H, f V) =} α (f H, f V), Cy (f H, f V) = Ye (f H, f V) = α (f H, f V) ...... (13) Is converted into an RGB signal by the RGB conversion unit 12 shown in FIG. 3, the carrier components at this point are (1), (2),
From equations (3) and (13), they are expressed as follows.

_{R (f H, f V)} = (a 11 + a 12 -a 13 -a 14) α (f H, f V), G (f H, f V) = (a 21 + a 22 -a 23 -a 24 ) Α (f _H , f _V ), B (f _H , f _V ) = (a ₃₁ + a ₃₂ −a ₃₃ −a ₃₄ ) α (f _H , f _V ) (14) , If the sum of the coefficients in the first and second columns is equal to the sum of the coefficients in the third and fourth columns, ie, a _i1 + a _i2 = a _i3 + a _i4 (i = 1,2,3 If the relationship (15) holds, the points (1 / p _H , 0) and (0,1 / 2p
The carrier components of the RGB signal in _V ) disappear, so that no carrier components of the chrominance signal occur at these points. Similarly, a point (−1 / p _H ,
0) and (0, can be eliminated by satisfying the carrier component even (15) in relation of the color signal at -1 / 2p _V).

FIG. 15 shows positions where the carriers of the luminance signal and the chrominance signal are generated near the origin at this time.

In order to minimize aliasing distortion in such a color image pickup apparatus, the optical low-pass filter having the light beam splitting characteristic shown in FIG. 6 used in the first embodiment and the optical low-pass filter shown in FIG. It is sufficient to use a combination of optical low-pass filters having the light beam splitting characteristics shown in (1), and the division distance of any of the optical low-pass filters only needs to satisfy the condition shown in Expression (7).

〔The invention's effect〕

As described above, according to the present invention, it is possible to provide a color imaging device that can obtain a horizontal resolution up to a frequency of f _H = 1 / p _H , has good resolution, and has little moiré. it can.

[Brief description of the drawings]

FIG. 1 is a diagram showing a color filter array used in the first embodiment of the present invention, FIG. 2 is a two-dimensional frequency characteristic diagram (first quadrant) of a color light carrier by the color filter array of FIG. 1, and FIG. FIG. 4 is a block diagram of the first embodiment, FIG. 4 is an explanatory diagram of characteristics of the first embodiment, FIG. 5 is a diagram showing an arrangement of an optical low-pass filter of a second embodiment of the present invention, and FIG. , FIG. 7 is a characteristic diagram of the optical low-pass filter of the embodiment, FIG. 8 is a diagram showing a modification of the optical low-pass filter of the embodiment, and FIG. FIG. 8 is a characteristic diagram of the optical low-pass filter shown in FIG. 8, FIG. 10 is a two-dimensional frequency characteristic diagram (second quadrant) of the color light carrier by the color filter arrangement of FIG. 1, and FIG. 11 is a third embodiment of the present invention. FIG. 12 is an explanatory diagram of characteristics, and FIG. 12 is a view for explaining the configuration of an optical low-pass filter according to a fourth embodiment of the present invention. Figure, FIG. 13 characteristic diagram of the optical low-pass filter of the embodiment, Fig. 14 characteristic diagram of a modification of the optical low-pass filter of the embodiment, FIG. 15 of the present invention 5
FIG. 16 is a diagram showing a conventional color filter array example 1, FIG. 17 is a two-dimensional frequency characteristic diagram of a color light carrier of the same color filter array example 1, and FIG. 18 is a conventional color filter. 19 shows a two-dimensional frequency characteristic diagram of the same color filter array example 2, FIG. 20 shows a conventional color filter array example 3, and FIG. 21 shows a color light carrier of the same color filter array example 3. 3 is a two-dimensional frequency characteristic diagram of FIG.

Claims (2)

(57) [Claims] 1. An image sensor having a pixel having an offset sampling structure in which a horizontal pitch is P _H , a vertical pitch is P _V , and a horizontal offset is P _H / 2, and each pixel of the image sensor is provided. And a color filter array comprising four types of color filters provided for each of the color filters. Each type of color filter has a horizontal pitch of 2P _H , a vertical pitch of 2P _V , and a horizontal offset amount of P _{In a} color imaging apparatus including a color filter array having an offset sampling structure of _H , a matrix of three rows and four columns is obtained from the output of each pixel provided with the four types of color filters. A = (a _ij ) (i = 1,2 , 3; J = 1,2,3,4), which is a conversion means for converting into an RGB signal by using the following formula (i) to
(Iii) A color imaging apparatus comprising a conversion unit that satisfies any one of (1) and (2). (I) a _i1 + a _i3 = a _i2 + a _i4 (i = 1,2,3) (ii) a _i1 + a _i4 = a _i2 + a _i3 (i = 1,2,3) (iii) a _i1 + a _i2 = a _i3 + a _i4 (i = 1,2,3) 2. An imaging optical system, comprising:
A distance D in a direction that forms an angle of θ counterclockwise with respect to the scanning direction of the image sensor or in a direction that forms an angle of θ clockwise with respect to the direction opposite to the direction of scanning of the image sensor 2
An optical low-pass filter for splitting into light rays of a book,
2. The color imaging apparatus according to claim 1, further comprising an optical low-pass filter satisfying the following condition. However,