JPH1098735A - Single ccd color camera and method for separating color signal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a single CCD color camera with a color separating circuit which has high horizontal and vertical resolutions and suppresses the generation of a false color signal. SOLUTION: A CCD 10 is driven by a driving circuit 102 to perform all-pixel independent read driving. A CL CR CB generating circuit 104 receives and holds a signal outputted from the CCD 10 in a RAM 106, and operates a luminance signal and a color difference signal for a pixel (x, y) in a (y)th row and an (x)th column receiving the CCD 10 on the basis of output signals of selected two rows and two columns according to the position of the pixel (x, y) and the kind of a corresponding color filter. A matrix circuit 108 receives the output of the CL CR CB generating circuit 104 and performs a specific operation to separately generate RGB signals for all the pixels of the CCD 10.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color camera, and more particularly to a solid-state imaging device including photoelectric conversion elements arranged in an array corresponding to color filters arranged in an array according to a color difference sequential system. The present invention relates to a single-chip color camera having a color separation circuit for processing signals and a color signal separation method.

[0002]

2. Description of the Related Art In a color camera, a CCD (Charge Coupled D) widely used as an imaging device at present is used.
evice) merely changes the amplitude of the output signal according to the brightness of the received light, and the output signal does not include color information. Therefore, in order to obtain color information, it is necessary to use optical means to apply a filter to light incident on the CCD.

[0003] In a home color camera, a so-called single-panel system for extracting three primary color signals from a single CCD is adopted, and a so-called simultaneous color imaging system using a color filter array on the light receiving surface side of the CCD. Has been adopted.

[Configuration of Interline Transfer CCD] FIG.
Reference numeral 1 denotes an interline transfer CCD 1 that is generally used as a configuration of a CCD in a home color camera.
FIG. 3 is a schematic block diagram illustrating a configuration of a zero.

[0005] The interline transfer CCD 10 includes a photosensitive section 12 composed of pn junction type photodiodes arranged in an array, a transfer section 14 including an analog shift register composed of CCDs, and a charge transferred by the transfer section 14. And a horizontal transfer register 16 which receives the signal charges and converts the sequentially transferred signal charges into a voltage, and transfers and outputs the signal in the horizontal direction.

In FIG. 11, for simplicity, the pn junction type photodiode has a configuration in which three pixels are arranged both vertically and horizontally. In fact, C used for color cameras
In a CD, for example, photodiodes corresponding to 500 pixels in the vertical direction and 500 to 800 pixels in the horizontal direction are arranged in an array.

Next, the operation will be briefly described.
When light enters the photodiode, charges are generated and accumulated in the diode. Next, by applying a predetermined voltage to the shift gate, the accumulated charges are simultaneously transferred to the analog shift register. The CCD analog shift register has clock pulse voltages φ _V1 , φ _V2 , φ
By applying _V3 , the charges are sequentially transferred to the horizontal transfer register 16. Horizontal transfer register 16
Then, after the transmitted signal charges are converted into voltages, they are driven by externally applied horizontal drive signals φ _H1 , φ _H2 , φ _H3 , and sequentially output to the outside as image pickup signal outputs.

[Drive system of interline transfer CCD]
In general, there are two modes of driving the interline transfer CCD: a frame accumulation method and a field accumulation method. Hereinafter, a field accumulation method used in correspondence with the color filter array of the color difference sequential method as described later will be described.

FIGS. 12A and 12B are schematic diagrams showing a method of reading signal charges from the interline transfer CCD 10 in the field accumulation method. FIG. 12A shows a method of reading signal charges in an odd field, and FIG. 12B shows an even field. Are described below.

As shown in FIG. 2A, in an odd field, signals of odd-numbered pixels and even-numbered pixels in the vertical direction are simultaneously transferred from a photosensitive section to a transfer section, and both signals are added in the transfer section. To be done.

In the even-numbered field, the combination is changed, and the signals of the even-numbered pixel and the odd-numbered pixel are simultaneously transferred to the transfer unit 14 so that the transfer unit 14 adds them.

When the color filter array is arranged in the color difference sequential system, such an interline transfer CC is used.
By using the transfer method of D10, the color difference signal is separated.

[Separation Method of Color Difference Signal] FIG. 13 is a schematic diagram showing a flow of signal processing of a signal output from the interline transfer CCD 10 when the color filter array is arranged in a color difference sequential method.

As shown in FIG. 13, in a color filter of a color difference sequential system, magenta (hereinafter, referred to as a color filter) is used.
Mg), green (hereinafter, represented by G), cyan (hereinafter, represented by Cy), yellow (hereinafter, represented by Ye)
Are arranged in a mosaic pattern.

Here, in the mixing of light colors, a so-called additive method is established, so that the three primary colors red (R), green,
The following relationship is established between blue (B) and complementary colors Mg, Ye, and Cy.

Mg = R + B (1) Ye = R + G (2) Cy = B + G (3) Accordingly, the above-mentioned Mg, G,
By using Ye and Cy, the intensity of a G signal having a large specific gravity with respect to a luminance signal among R, G, and B of the three primary colors can be changed to R
It can be larger than the signal and the B signal.

In the example shown in FIG. 13, an array portion of 4 rows and 4 columns is extracted from the color filter array of the color difference sequential system. In the odd-numbered rows (y direction), the Mg color filter and the G color filter
Direction).

On the other hand, in the even-numbered rows, Ye color filters and Cy color filters are arranged alternately in the horizontal direction.

The reading method from the interline transfer CCD 10 having the color filter array having such an arrangement is based on the vertical (y-direction), two-pixel addition reading.

First, in the odd field, the odd-numbered pixel and the even-numbered pixel in the vertical direction are added. In the next even field, the combination is changed, and the even-numbered pixel and the odd-numbered pixel are added. By doing so, for example, in an even field, G + Cy, Mg + Y in the nth scanning line
In the order of e,..., in the (n + 1) th scanning line, Mg + Cy,
The signals are output in the order of G + Ye,.

These signals are pulse amplitude modulated signals as shown in FIG. In FIG. 13, the complementary color signals are replaced with the three primary color signals based on the above equations (1) to (3), and are shown as amplitude modulated waveforms of the three primary color signals.

The harmonic component is omitted, and the DC component and the fundamental component are as follows.

[0023] First, the signal of the n-th scan line in the even field So is So = (Mg + Ye) + (G + Cy) +1/2 · {(Mg + Ye) - (G + Cy)} sin2πf n t = 2R + 3G + 2B + 1/2 · (2R- _{G) sin2πf n t ... (4} ) . On the other hand, the signal Se of n + 1 th scan line Se = (Mg + Cy) + (G + Ye) +1/2 · {(Mg + Cy) - (G + Ye)} sin2πf n t = 2R + 3G + 2B + 1 / a 2 · (2B-G) sin2πf n t ... (5). Here, f _n is 1 of the sampling frequency f _s .
/ 2 represents the Nyquist frequency.

The luminance signal is obtained by converting only the DC component in the above equations (4) and (5) into a low-pass filter (hereinafter, LPF).
).

On the other hand, two color difference signals 2R-G and 2B-G
Is a band-pass filter centered on the frequency f _n (hereinafter, B
PF) and are easily separated by detection.

That is, the luminance signal and the color difference signal are obtained line-sequentially. However, as it is,
When focusing on one scanning line, the color difference signal is 2
Only R-G or 2B-G can be obtained, and color signals cannot be reproduced for the scanning line.

Therefore, the actual reproduction of the luminance signal and the chrominance signal is performed by the method described below.

First, the luminance signal C _L is created from the sum of adjacent pixels in each row of the CCD output signal, that is, C _L = Ye + Cy + Mg + G = 2R + 3G + 2B. An LPF for realizing this is realized, for example, by the circuit shown in FIG.

Here, the LPF 20 includes a one-pixel delay circuit 22 for receiving a CCD output signal,
An addition circuit 24 that receives the signal from the pixel delay circuit 22 and adds and outputs the two; and an attenuator 2 that receives the output of the addition circuit 24 and halves the signal strength and outputs the signal.
6 is included.

The transfer function LPF (z) of the LPF 20 and its frequency characteristic are as follows: LPF (Z) = (1 + Z ⁻¹ ) / 2 (6) | LPF (Z) | = | cos (πf / 2f _n ) | ... and since (7), the spectrum of the Y signal, gradually decreases from the low frequency region as shown in FIG. 14 (b), a becomes zero characteristic at the Nyquist frequency f _n.

Here, assuming that the number of effective pixels of the CCD 10 is H pixels in the horizontal direction and V pixels in the vertical direction, the horizontal resolution is H and the vertical resolution is V / 2 for a luminance signal in one field. That is, the resolution in the vertical direction is halved with respect to the number of pixels arranged in the CCD.

[0032]

Next, the color signal separation processing will be described.

FIG. 15 is a schematic block diagram showing a configuration of a conventional color separation circuit 200 for separating three primary color signals R, G, and B from a CCD output signal.

The color separation circuit 200 receives the CCD output signal, performs sample hold (S & H) according to the sampling pulse SP1, and outputs the S1 signal (Ye for n lines).
+ Mg,...)
Receiving the CCD output signal, the sampling pulse SP
Sampling pulse SP2 180 degrees out of phase with 1.
, A sample-and-hold circuit 204 that outputs an S2 signal (Cy + G,... In the n-th line), an addition circuit 206 that receives signals S1 and S2 and outputs a signal S1 + S2 obtained by adding both, and a signal S1 And a subtraction circuit 208 which receives S and S2 and outputs a difference signal S1-S2 between them.

The color separation circuit 200 further includes a signal S1-
1H delay line 210 which receives S2 and outputs one horizontal scanning time delay, and receives signals S1-S2 and the output from 1H delay line 210, and switches and outputs any one according to a line selection pulse. , The color difference signal C _R (= 2R−G) and the color difference signal C _B (= 2B−G) for each line
Receiving the output from the selection circuit 212 and the signal S1 + S2, that is, the luminance signal C _L, and performing a predetermined linear operation to obtain the three primary color signals R,
And a matrix circuit 214 for separating and outputting G and B.

FIG. 16 is a schematic diagram showing a relationship between signals output from the CCD 10 and sample and hold pulses SP1 and SP2 for sampling and holding the signals.

The operation of the conventional color separation circuit 200 shown in FIG. 15 will be described below with reference to FIGS.

The following description focuses on the signal processing in the even field, but the basic operation is exactly the same as that of the odd field, except that the row in which signal addition is performed is shifted.

First, consider the signals S1 and S2 output from the sample-hold circuits 202 and 204 after the CCD outputs from the n-th line in the even field are sampled and held by the sampling pulses SP1 and SP2.

Referring to FIG. 16, sampling pulse S
The signal S1 sampled and held by P1 is a sum signal of the signal Ye and the signal Mg for the nth line.
On the other hand, the signal S2 sampled and held according to the sampling pulse SP2 is a sum signal of the signal Cy and the signal G.

Therefore, the sum and difference between the S1 signal and the S2 signal are as follows. S1 + S2 = Ye + Mg + Cy + G = 2R + 3G + 2B (8) S1-S2 = (Ye + Mg)-(Cy + G) = 2R-G (9) On the other hand, a signal obtained by sampling and holding the output signal from the n + 1 line of the CCD in the even field by the sampling pulse SP1. S1 is a sum signal of the signal Ye and the signal G, and the signal S2 sampled and held by the sampling pulse SP2 is a sum signal of the signal Cy and the signal Mg.

Therefore, the difference signal between the signal S1 and the signal S2 is as follows. S1−S2 = (G + Ye) − (Mg + Cy) = − (2B−G) (10) Hereinafter, 2R−G = C _R , − (2B−G) = C _B , 2R +
When 3G + 2B = C _L, the generating of the difference signal S1-S2, signal C _R, the signal C _B will be obtained alternately for each line.

[0043] with respect to one scanning line, the luminance signal C _L, the color difference signals C _R, is C _B are required for each one scan line in order to generate a color signal.

In the color separation circuit 200, a specific scanning
Line, for example n + 1 lines, n + 1 lines
Color difference signal C obtained from_BAnd color difference signals obtained from n lines
Issue C _R, The color signal in the (n + 1) th line
Is generated and separated.

That is, the signals S1-S2 (C _R , C _B ,...) Output from the subtraction circuit 208 and this signal are
The signals (C _B , C _R ,...) Obtained through the delay line 210
Circuit 21 which can be switched by a line selection pulse
2, the color difference signals S1-S one scanning line earlier.
2 can be used as a color difference signal for the scan line currently being read. That is, the selection circuit 212 outputs the signal S1-S2 corresponding to the scanning line currently being read, for example, the signal C1.
When _R is the matrix circuit 214 and outputted to the matrix circuit 214 the signal as a color difference signal C _R as it is, the color difference signals of one scanning line before being output from the 1H delay line 210 at the same time as a signal C _B Output. on the other hand,
If the color difference signals S1-S2 which are read from the current scan line corresponding to the signal C _B is the signal from the 1H delay line 210 to the contrary to the case of the as color difference signals C _R, subtraction circuit 20
The signal S1-S2 from 8 to output to a matrix circuit 214 as the color difference signal C _B.

The matrix circuit 214 outputs RGB signals, which are three primary color signals, from the received three signals of the luminance signal C _L and the color difference signals C _R and C _B according to the following conversion formula.

[0047] _{G = (2C L -2C R -2C} B) / 10 ... (11) R = (C L + 4C R -C B) / 10 ... (12) B = (C L -C R + 4C B) / 10 (13) Here, the signals S1 and S2 are the result of sampling and holding the output signal of the CCD at the sampling frequency fn. FIG. 17 is a graph showing frequency characteristics of the signals S1 and S2.

As a result of the sampling and holding of the signals S1 and S2, as shown in FIG. 17, the response gradually decreases from the low frequency region and has a characteristic that it becomes 0 at the frequency fn / 2.

Here, the luminance signal C _L , the color difference signals C _R and C _B, and the RGB signals generated from them have the same frequency characteristics as in FIG.

Therefore, the horizontal resolution of the RGB signal is H
It becomes -1. The vertical resolution within one field is
Issue C_RAnd signal C_BIs V / 4, and the luminance signal
Issue C _LIs V / 2. Therefore, from these
For the generated RGB signals, the vertical resolution is about
V / 2.

As described above, the resolution of RGB can be obtained only from 1/2 to 1/4 of the number of effective pixels.

The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a luminance signal and a chrominance signal with a high resolution without deteriorating the resolution with respect to the number of effective pixels. The object of the present invention is to provide a single-panel color camera capable of reproducing an image.

Another object of the present invention is to provide a single-chip color camera capable of suppressing generation of a false color signal at an edge in an image and improving image quality.

Still another object of the present invention is to provide a color separation method capable of generating a luminance signal and a color difference signal corresponding to each pixel based on output signals independently read from all pixels of a CCD. It is.

[0055]

According to a first aspect of the present invention, there is provided a single-panel color camera including a solid-state imaging device in which photoelectric conversion elements respectively corresponding to pixels are arranged in an array, and the solid-state imaging device is provided on a light receiving surface side. The color filters are arranged in a color difference sequential manner corresponding to the photoelectric conversion elements, and include a color filter array having green and three kinds of complementary filters corresponding to four pixels in an arbitrary two rows and two columns. Color separation means for receiving an output signal read out pixel-by-pixel and generating a color signal corresponding to every four pixels, further comprising a storage means for holding output signals sequentially output from the solid-state imaging means And, for each pixel group of four pixels, based on the output signal held in the storage means, a luminance signal and an intensity difference between a green signal and a predetermined one of the other primary color signals excluding the green signal among the three primary colors. Corresponding color signal Generated, and a calculating means for generating a color signal from the luminance signal and color difference signals, calculating means,
i) If the photoelectric conversion elements corresponding to the complementary color filters in the pixels of 2 rows and 2 columns that output signals including predetermined primary color signal components are arranged diagonally, the pixels of 2 rows and 2 columns Generating a color difference signal from the output signal from the group; ii) 2
If none of the photoelectric conversion elements corresponding to the complementary color filters in the pixels in row 2 that output a signal including a predetermined primary color signal component are arranged diagonally, two out of the pixels in row 2 and column 2 A color difference signal is generated from output signals from adjacent two-row, two-column pixel groups so as to share pixels.

According to a second aspect of the present invention, in the configuration of the single-panel color camera according to the first aspect, the arithmetic means comprises a photoelectric conversion element corresponding to the pixel (x, y) in the y-th row and the x-th column. Are output signals D (x, y) and D (x + y) by a linear operation according to the arrangement of the color filter array corresponding to the pixel group.
1, y), D (x, y + 1) and D (x + 1, y +
1), output signals D (x, y), D (x + 1, y),
A set of D (x, y-1) and D (x + 1, y-1) and output signals D (x, y + 1), D (x + 1, y +
1) a color difference that generates a luminance signal corresponding to a pixel (x, y) and a first color difference signal and a second color difference signal from one of a set of D (x, y + 2) and D (x + 1, y + 2) A signal generation unit; and a separation operation unit that separates three primary color signals by a predetermined linear operation on the luminance signal, the first color difference signal, and the second color difference signal.

According to a third aspect of the present invention, there is provided a single-panel color camera.
The configuration of the single-panel color camera according to claim 2, wherein
The filter array consists of alternating magenta color filters
And a plurality of first rows, including a green filter, and alternately
Multiple filters, including yellow and cyan filters
A second row, where n is a natural number.
Belongs to the rows with y = 4n-3 and y = 4n-1, and
x includes a yellow filter in even columns, and the first row has y =
4n-4 and y = 4n-2, and y = 4
If the pixel belongs to row n-4, a green filter is applied to the column where x is an odd number.
X is green in an even column when y = 4n-2
A color filter, and the color signal generating means includes a pixel (x, y)
Output signals D (x, y) from a pixel group of 2 rows and 2 columns including
D (x + 1, y), D (x, y + 1), D (x + 1, y
+1) as a first pixel group signal and an output signal D (x,
y-1), D (x + 1, y-1), D (x, y), D
A set of (x + 1, y) is defined as a second pixel group signal, and an output signal
D (x, y + 1), D (x + 1, y + 1), D (x, y
+2) and D (x + 1, y + 2) as a third pixel group signal
, The luminance signal C corresponding to the pixel (x, y)
_L(X, y), the first color difference signal C_R(X, y) and the
2 color difference signal C_BFor (x, y), i) x = 2n−
2, y = 4n-4 and x = 2n-1, y = 4n-4
Case C_L(X, y) and C_B(X, y) is the first pixel
From the group signal, C_R(X, y) is generated from the third pixel group signal.
Ii) x = 2n-2, y = 4n-3 and x = 2
C when n-1 and y = 4n-3_L(X, y) and C_R
(X, y) is calculated from the first pixel group signal as C _B(X, y)
Generated from the second pixel group signal, iii) x = 2n-2,
When y = 4n-2, x = 2n-1, and y = 4n-2
C_L(X, y) and C_R(X, y) is the first pixel group signal
From the number, C_BGenerate (x, y) from the third pixel group signal
Iv) x = 2n-2, y = 4n-1 and x = 2n
-1, y = 4n-1 C _L(X, y) and C
_B(X, y) is calculated from the first pixel group signal as C_R(X, y)
Is generated from the second pixel group signal.

According to a fourth aspect of the present invention, there is provided a color filter array and a color filter array having green and three types of complementary filters corresponding to four pixels in an arbitrary two rows and two columns. A method of separating a luminance signal in a single-panel color camera including a solid-state imaging device in which each corresponding photoelectric conversion element is arranged in an array, and for each of four pixels based on an output signal from the photoelectric conversion element, Obtaining a luminance signal from a sum of output signals from the pixel group;
For each pixel group of four pixels, i) a color difference signal corresponding to a difference in intensity between a green signal and a predetermined one of the other primary color signals excluding the green signal among the three primary colors is i) a complementary color filter in a pixel of 2 rows and 2 columns If the photoelectric conversion elements corresponding to the above are arranged in a diagonal direction to output a signal including a predetermined primary color signal component, a color difference signal is generated by an output signal from a pixel group of 2 rows and 2 columns, ii) When the photoelectric conversion elements corresponding to the complementary color filters in the pixels of 2 rows and 2 columns that output signals including predetermined primary color signal components are not arranged diagonally,
Generating a color difference signal from output signals from adjacent two-row, two-column pixels so as to share two pixels of the two-row, two-column pixel; and generating a color signal from the luminance signal and the color-difference signal. Is provided.

[0059]

BEST MODE FOR CARRYING OUT THE INVENTION

[First Embodiment] FIG. 1 is a schematic block diagram showing a configuration of a main part of a single-panel color camera 100 according to a first embodiment of the present invention.

The single-panel color camera 100 is roughly
An optical system 2 for receiving light from a subject, a CCD 10 for converting an optical image determined by the optical system 2 into an electric signal,
It includes a drive circuit 102 that performs all-pixel independent read driving on the CD 10 and a color separation circuit 101 that receives output signals from the CCD 10 and outputs three primary color signals R, G, and B corresponding to each pixel.

The configuration of the color separation circuit 101 is to be compared with the configuration of the conventional color separation circuit 200 shown in FIG.

The color separation circuit 101 outputs the output signal of the CCD 10.
The luminance signal C_LAnd color difference signal C_Rand
C_BGenerate C_LC_RC_BGenerating circuit 104 and C_LC
_RC _BRead from CCD 10 with generation circuit 104
Signal D (x, y), luminance signal C_L, Color difference signal C_Rand
C_BOf at least one scanning line
Random access memo that can hold these signals
C (hereinafter referred to as RAM) 106 and C_LC_RC_BGenerate
Receiving the signal from the circuit 104 and performing a predetermined operation
Matrix circuit 108 for separating and outputting color signals RGB
And

FIG. 2 is a schematic diagram for explaining the arrangement of the color filter array and the signal readout from the CCD 10 including the color filter array having this arrangement in the present embodiment.

Unlike the conventional field addition readout system shown in FIG. 12, in the present embodiment, an all-pixel independent readout operation is performed. That is, the driving circuit 102 generates a driving pulse which does not mix two pixels in the vertical direction in the CCD 10 unlike the conventional example. That is, the output signal of the CCD 10 is not a field readout shown in FIG. 12, but a so-called frame readout.

That is, as shown in FIG. 2, for the 0th line in the vertical direction (y direction), the color filters Mg and G are alternately arranged in the horizontal direction (x direction). The read signal from the CCD 10 is a signal in which Mg and G are alternately output for the scanning line corresponding to this line. For the next first line (corresponding to y = 1), the signal Ye and the signal Cy are output alternately.

As described above, the relationship between each pixel (x, y) in the CCD output signal and the filter color is as follows when n is a natural number.

I) When y = 4n-3 or 4n-1 (that is, y = 1, 3, 5...) If x is an even number, the color of the color filter is Ye, and if x is an odd number, , The color of the color filter is Cy.

Ii) When y = 4n-4 (that is, y = 4n-4)
y = 0, 4, 8,...) If x is an odd number, the color of the color filter is Mg, and if x is an even number, the color of the color filter is G.

Iii) When y = 4n-2 (that is, y = 2, 6, 10,...) If x is an odd number, the color of the color filter is G. If x is an even number, the color filter is a color filter. Is Mg.

In the above description, for convenience of explanation, M
The rows to which the g and G color filters belong are even rows, and Y
Although the rows to which the e and Cy color filters belong are odd rows, as will be apparent from the following description, the present invention is not limited to such a case. The same can be applied to an array of color filters that are staggered.

[Generation of Color Difference Signal from Four Pixels Fixed for Pixel (x, y)] As described above, in the present embodiment, output signals are read out from CCD 10 independently for all pixels, and these signals are read out. The signal is held in the RAM 106.

Therefore, as described below, the luminance signal C _L (x, y), the first color difference signal C _R (x, y), and the second color difference signal C _B are provided for each pixel (x, y).
(X, y) can be generated by the following operations.

Hereinafter, the pixel (x, y) in the CCD 10 will be described.
The output signal from the photoelectric conversion element corresponding to D (x, y)
Will be represented by

At this time, pixels (x, y), (x + 1, y), (x, y + 1), (x + 1, y) in two rows and two columns including the pixel (x, y)
It is possible to extract a luminance signal and two color difference signals by a linear operation on the output signal from (x + 1, y + 1).

That is, in general, the expression is as follows. Cc (x, y) = Kc (x, y) .D (x, y) + Kc (x + 1, y) .D (x + 1, y) + Kc (x, y + 1) .D (x, y + 1) + Kc (x + 1, y + 1) .D (x + 1, y + 1) (14) (where c is any of _L , _R , and _B ) At this time, the color of the filter of the pixel (X, Y) and C _L C _R C
Coefficients K _L (X, Y), K for creating _B signal
_{R (X, Y), K} B (X, Y) and shall have the following relationship. Here, X = x, x + 1 and Y = y, y + 1.

[0076] i) a pixel (X, when the color filters Y) is _{Cy K L (X, Y)} = 1, K R (X, Y) = - 1, K B (X, Y) = 1 If the color filters ii) a pixel (X, Y) is _{Ye K L (X, Y)} = 1, K R (X, Y) = 1, K B (X, Y) = - 1 iii) pixel (X, Y) when the color filters are _{Mg K L (X, Y)} = 1, K R (X, Y) = 1, K B (X, Y) = 1 iv) pixel (X, Y If the color filters are G K _L of) (X, Y) = 1 , K R (X, Y) = - 1, K B (X, Y) = - 1 or more color filters and coefficient K _L , K _R, based on the relationship between the K _B, actually explaining a process of the luminance signal and color difference signals are generated more specifically.

In the following, as an example, in FIG. 2, a luminance signal and a color-difference signal corresponding to the pixel (0, 0) are obtained.

Note that the equations (1) to (3) hold, the luminance signal C _L (0,0) is expressed by the following equation (14).

C _L (0,0) = D (0,0) + D (1,0) + D (0,1) + D (1,1) = Mg + G + Ye + Cy = (R + B) + G + (R + G) + (B + G) = 2R + 3G + 2B (14) Further, the first color difference signal C _R (0,0) is calculated from the following equation (15).

C _R (0,0) = D (0,0) + D (0,1) -D (1,0) -D (1,1) = Mg + Ye-G-Cy = (R + B) + (R + G ) −G− (B + G) = 2R−G (15) On the other hand, the second color difference signal C _B (0,0) is expressed by the following equation (1).
6).

C _B (0,0) = D (0,0) + D (1,1) -D (0,1) -D (1,0) = Mg + Cy-Ye-G = (R + B) + (B + G ) − (R + G) −G = 2B−G (16) That is, the luminance signal and the two color difference signals for the pixel (0, 0) are represented by the pixels (0, 0), (1, 0), (0,
It can be calculated using output signals from 1) and (1, 1).

In the above example, the case where the pixel of interest (0, 0) corresponds to the magenta color filter is shown. However, even when the pixel of interest is another color filter, the pixel (x , Y) containing two rows and two columns of pixels (x, y),
(X + 1, y), (x, y + 1), (x + 1, y + 1)
A luminance signal and a color difference signal can be obtained by a linear operation of the output signal from.

Therefore, since a luminance signal and two color difference signals can be obtained for each pixel, it is possible to realize a color separation circuit having a higher resolution than a conventional color separation circuit.

[Problem of Generation of Color Difference Signal from Four Fixed Pixels] However, the above-described color separation method has the following problems.

I) When the luminance changes in the horizontal direction. That is, as shown in FIG. 2, a change in the luminance level H1 in the horizontal direction is caused by the light intensity input to the CCD pixel array.
Consider a case where there is a (edge where the luminance level changes from left to right on the screen) portion.

In the following, for the sake of simplicity, an achromatic optical signal having no color component is input to this edge portion.
It is assumed that only the brightness level has changed.

At this edge portion, it is assumed that the luminance level changes by b / a (b> a) in the horizontal direction (x-axis).

Hereinafter, as an example, the luminance signal and the chrominance signal corresponding to the pixel (0, 0) are calculated as follows in the above case.

First, the luminance signal C _L (0,0) is as follows. C _L (0,0) = D (0,0) + D (1,0) + D (0,1) + D (1,1) = a × Mg + b × G + a × Ye + b × Cy = a × (R + B + R + G) + b × (2G + B) (17) Next, the first color difference signal C _R (0,0) is as follows.

C _R (0,0) = D (0,0) + D (0,1) −D (1,0) −D (1,1) = a × Mg + a × Ye−b × G−b × Cy = a × (R + B + R + G) −b × (2G + B) (18) Next, the second color difference signal C _B (0,0) is as follows.

C _B (0,0) = D (0,0) + D (1,1) −D (0,1) −D (1,0) = a × Mg + b × Cy−a × Ye−b × G = a × (R + B−R−G) + b × B (19) Here, since it is assumed that there is no color component, the following two equations hold.

2R−G = 0 2B−G = 0 Therefore, assuming that 2R = 2B = G = 2S, R = S,
When B = S and G = 2S are substituted into the above equations (17) to (19), the following results are obtained for the first and second color difference signals.

C _R (0,0) = − 5S (b−a) (20) C _B (0,0) = S (b−a) (21) Therefore, the pixel (0) originally having no color component , 0). As a result, a false color appears on the screen, which causes deterioration of image quality.

Similarly, when there is a change in the horizontal luminance level H2 shown in FIG. 2, the color difference signal for the pixel (1, 0) is calculated as follows.

C _R (1,0) = 5S (b−a) (22) C _B (1,0) = − S (b−a) (23) Therefore, there is a change in luminance level H2. Also,
A false color occurs at the pixel (1, 0).

The same can be said for all pixels having an edge of luminance change in the horizontal direction.

Ii) When the luminance changes in the vertical direction Next, as shown in FIG. 2, when there is a luminance change V1 in the vertical direction, similarly for the pixel (0, 2),
The calculation of the color difference signal is as follows.

The color difference signal C _R (0, 2) is as follows. C _R (0,2) = D (1,2) + D (0,3) −D (0,2) −D (1,3) = b × Mg + a × Ye−b × G−a × Cy = a × (R + G−GB) + b × (R + BG) (24) The color difference signal C _B (0, 2) is as follows.

C _B (0,2) = D (1,2) + D (1,3) −D (0,2) −D (0,3) = b × Mg + a × Cy−b × G−a × Ye = a × (G + B−R−G) + b × (R + B−G) (25) Similarly, by setting 2R = 2B = G = 2S, C _R (0, 2) = 0 and C _B (0,2) = 0 is obtained.

Accordingly, it can be seen that no false color occurs in the vertical direction. Similarly, when there is a change V2 in the luminance signal in the vertical direction shown in FIG. 2, the values of the color difference signals are both 0 for the pixels at the edge portion, and the false color is applied to the edge in the vertical direction. It can be seen that no problem occurs.

Accordingly, as described above, the pixel (x, y) is replaced by the pixel (x, y,
y), (x + 1, y), (x, y + 1), (x + 1, y
In the method of calculating the color difference signal based on the output signal from (+1), a false color is generated when there is a significant luminance change (edge) in the horizontal direction. For this reason, when an image is reproduced on a screen, a false color appears in a portion that originally has no color, thereby deteriorating the image quality. The signal CL, C _R, R of the three primary colors from the C _B, G, in the process of generating the B signal, a false luminance signal leads to deterioration of the luminance signal will be generated by the false color signal component.

[Relationship Between Generation of False Color Signal and Color Filter Arrangement] As described above, in the arrangement of the color difference sequential type color filters as shown in FIG. Occurs.

Therefore, the relationship between the pixel where such a false color occurs and the arrangement of the color filters will be considered below.

FIG. 3 is a diagram showing a filter arrangement for outputting a color signal containing an R signal and a B signal component in the color filter array shown in FIG. FIG. 3A is an oblique line showing an array of pixels outputting a signal having an R signal component, and FIG. 3B is an oblique line showing an array of pixels outputting a signal including a B signal component. .

Note the equations (1) to (3), the R signal is included in the output signal from the pixel corresponding to the magenta filter and the yellow filter in the color filter array shown in FIG. On the other hand, the B signal component is included in an output signal from a pixel corresponding to the magenta color filter and the cyan color filter.

Here, the case where there is a change H1 in the luminance signal in the horizontal direction as shown in FIG. 2 will be considered below.

Focusing on the pixel (0,0), among the output signals from the above-described pixel in the above-mentioned two rows and two columns including this pixel, the output signal having the R signal component can be obtained only from the side having the lower luminance level. There is no R signal component on the side with the higher luminance level.

Correspondingly, the color difference signal C _R (0,0)
Has a large false color signal component.

On the other hand, of the pixels in the above-mentioned two rows and two columns including the pixel (0, 0), the pixel outputting the signal having the B signal component has a high luminance level and a low luminance level. Existing.

Therefore, the false color signal for the color difference signal C _B (0,0) has a smaller absolute value than the false color signal component appearing for the color difference signal C _R (0,0).

[0111] Therefore, even if an edge is present in the luminance signal changes, in order to suppress the occurrence of a false color signal, the color difference signals to be obtained is, if a C _R signal, a pixel that outputs the R signal component There may be decided to calculate the color difference signals C _R on the basis of a signal from the two rows and two columns of pixel group arranged in the diagonal direction. The same is true for the color difference signals C _B.

[0112] Operation of C _L C _R C _B generating circuit 104 will now be described color separation method capable of suppressing the occurrence of a false color signal as described above.

That is, from the CCD 10 shown in FIG.
C when pixel independent reading is performed _LC_RC_BGeneration circuit
The operation of 104 will be described.

In the description so far, the color difference signal for the pixel (x, y) is converted to the pixel (x, y) including this pixel.
y), (x + 1, y), (x, y + 1), (x + 1, y
+1), which is generated from an output signal of 2 rows and 2 columns.

In the following, for the sake of convenience in describing the method of generating a color difference signal on the drawing, the luminance of the pixel at the center position of these four pixels is determined based on the output signals from the pixels in these two rows and two columns. The expression of generating a signal and a color difference signal will be used.

Also in this case, if the luminance signal and the color difference signal at the center position of the four pixels are considered as the luminance signal and the color difference signal for the pixel (x, y), the description so far and the substantive changes are as follows. Absent.

When the output signal D (x, y) is, for example, a magenta color filter, this signal is
(X, y).

FIGS. 4 to 7 show a color separation circuit 101 according to the present invention.
A C _L C _R C diagram for explaining the operation of the _B generating circuit 104 in.

FIG. 4 shows, for example, pixels (0, 0), (1, 0), (1, 2) in the color filter array shown in FIG.
FIG. 3 is a diagram showing pixels in two rows and two columns used when calculating a color difference signal at a central position surrounded by (0) and (1, 1).

FIG. 5 shows, for example, the pixel (0,
1) used to calculate a color difference signal at a central position surrounded by (1, 1), (0, 2) and (1, 2)
FIG. 3 is a diagram illustrating pixels in a row and a second column.

FIG. 6 shows, for example, the pixel (0,
2) used to calculate a color difference signal at a central position surrounded by (1,2), (0,3) and (1,3)
FIG. 3 is a diagram illustrating pixels in a row and a second column.

FIG. 7 shows, for example, the pixel (0,
3) 2 used for calculating a color difference signal at a central position surrounded by (1, 3), (0, 4) and (1, 4)
FIG. 3 is a diagram illustrating pixels in a row and a second column.

However, as will be described below, a method for calculating the luminance signal or the color difference signal shown in FIGS.
The selection of a pixel in row 2 column applies more generally to the calculation of the center position surrounded by pixels of a similar arrangement.

Referring to FIG. 4, pixels (0, 0), (1,
0), (0,1) and (1,1) to the luminance signal C _L and the color difference signal C _B in a central position surrounded is calculated both on the basis of the output signals from the four pixels. On the other hand, the color difference signal C _R is calculated based on output signals from pixels in two rows and two columns shifted by one pixel in the y direction.

Referring to FIG. 5, pixels (0, 1), (1,
The luminance signal and the color difference signal C _R at the central position surrounded by 1), (0, 2) and (1, 2) are calculated based on the output signals from these four pixels. On the other hand, the color difference signals C _B,
It is calculated from pixels in two rows and two columns shifted by one pixel in the −y direction.

Referring to FIG. 6, pixels (0, 2), (1,
2) The luminance signal C _L and the color difference signal C _R corresponding to the central position surrounded by (0, 3) and (1, 3) are calculated from the four pixels in two rows and two columns.

[0127] On the other hand, the color difference signal C _B is calculated based on the output signals from the pixels of two rows and two columns shifted by one pixel in the y direction.

Referring to FIG. 7, pixels (0, 3), (1,
3), (0,4) and (1,4) to the luminance signal C _L and the color difference signal C _B in a central position surrounded is calculated based on the output signal from the four pixels. On the other hand, the color difference signal C _R
Is calculated based on output signals from pixels in two rows and two columns shifted by one pixel in the −y direction.

The luminance signal described with reference to FIGS.
C_LAnd color difference signal C_RAnd C _BHow to calculate
More generally, it is as follows. i) For x-2n-2, y-4n-4 (n: natural number) C_L(X, y) = D (x, y) + D (x + 1, y) + D (x, y + 1) + D (x + 1, y + 1) = Mg (x, y) + G (x + 1, y) + Ye (x, y + 1) + Cy (X + 1, y + 1) C_R(X, y) = D (x + 1, y + 2) + D (x, y + 1) -D (x, y + 2) -D (x + 1, y + 1) = Mg (x + 1, y + 2) + Ye (x, y + 1) -G (x , Y + 2) −Cy (x + 1, y + 1) C_B(X, y) = D (x, y) + D (x + 1, y + 1) -D (x, y + 1) -D (x + 1, y) = Mg (x, y) + Cy (x + 1, y + 1) -Ye (x, y + 1) -G (x + 1, y) ii) When x-2n-2, y-4n-3 (n: natural number) C_L(X, y) = D (x + 1, y + 1) + D (x, y + 1) + D (x, y) + D (x + 1, y) = Mg (x + 1, y + 1) + G (x, y + 1) + Ye (x, y) + Cy (X + 1, y) C_R(X, y) = D (x + 1, y + 1) + D (x, y) -D (x, y + 1) -D (x + 1, y) = Mg (x + 1, y + 1) + Ye (x, y) -G (x, y y + 1) -Cy (x + 1, y) C_B(X, y) = D (x, y-1) + D (x + 1, y) -D (x, y) -D (x + 1, y-1) = Mg (x, y-1) + Cy (x + 1, y) ) -Ye (x, y) -G (x + 1, y-1) iii) When x-2n-2, y-4n-2 (n: natural number) C_L(X, y) = D (x + 1, y) + D (x, y) + D (x, y + 1) + D (x + 1, y + 1) = Mg (x + 1, y) + G (x, y) + Ye (x, y + 1) + Cy (X + 1, y + 1) C_R(X, y) = D (x + 1, y) + D (x, y + 1) -D (x, y) -D (x + 1, y + 1) = Mg (x + 1, y) + Ye (x, y + 1) -G (x, y) -Cy (x + 1, y + 1) C_B(X, y) = D (x, y + 2) + D (x + 1, y + 1) -D (x, y + 1) -D (x + 1, y + 2) = Mg (x, y + 2) + Cy (x + 1, y + 1) -Ye (x , Y + 1) -G (x + 1, y + 2) iv) When x-2n-2, y-4n-1 (n: natural number), C_L(X, y) = D (x, y + 1) + D (x + 1, y + 1) + D (x, y) + D (x + 1, y) = Mg (x, y + 1) + G (x + 1, y + 1) + Ye (x, y) + Cy (X + 1, y) C_R(X, y) = D (x + 1, y−1) + D (x, y) −D (x, y−1) −D (x + 1, y) = Mg (x + 1, y−1) + Ye (x, y) ) -G (x, y-1) -Cy (x + 1, y) C_B(X, y) = D (x, y + 1) + D (x + 1, y) -D (x, y) -D (x + 1, y + 1) = Mg (x, y + 1) + Cy (x + 1, y) -Ye (x, y) -G (x + 1, y + 1) v) When x-2n-1, y-4n-4 (n: natural number), C_L(X, y) = D (x + 1, y) + D (x, y) + D (x + 1, y + 1) + D (x, y + 1) = Mg (x + 1, y) + G (x, y) + Ye (x + 1, y + 1) + Cy (x, y + 1) C_R(X, y) = D (x, y + 2) + D (x + 1, y + 1) -D (x + 1, y + 2) -D (x, y + 1) = Mg (x, y + 2) + Ye (x + 1, y + 1) -G (x + 1 , Y + 2) -Cy (x, y + 1) C_B(X, y) = D (x + 1, y) + D (x, y + 1) -D (x + 1, y + 1) -D (x, y) = Mg (x + 1, y) + Cy (x, y + 1) -Ye (x + 1) , Y + 1) -G (x, y) vi) When x-2n-1, y-4n-3 (n: natural number), C_L(X, y) = D (x, y + 1) + D (x + 1, y + 1) + D (x + 1, y) + D (x, y) = Mg (x, y + 1) + G (x + 1, y + 1) + Ye (x + 1, y) + Cy (x, y) C_R(X, y) = D (x, y + 1) + D (x + 1, y) -D (x + 1, y + 1) -D (x, y) = Mg (x, y + 1) + Ye (x + 1, y) -G (x + 1 , Y + 1) -Cy (x, y) C_B(X, y) = D (x + 1, y−1) + D (x, y) −D (x + 1, y) −D (x, y−1) = Mg (x + 1, y−1) + Cy (x, y) ) -Ye (x + 1, y) -G (x, y-1) vii) For x-2n-1, y-4n-2 (n: natural number) C_L(X, y) = D (x, y) + D (x + 1, y) + D (x + 1, y + 1) + D (x, y + 1) = Mg (x, y) + G (x + 1, y) + Ye (x + 1, y + 1) + Cy (x, y + 1) C_R(X, y) = D (x, y) + D (x + 1, y + 1) -D (x + 1, y) -D (x, y + 1) = Mg (x, y) + Ye (x + 1, y + 1) -G (x + 1, y) -Cy (x, y + 1) C_B(X, y) = D (x + 1, y + 2) + D (x, y + 1) -D (x + 1, y + 1) -D (x, y + 2) = Mg (x + 1, y + 2) + Cy (x, y + 1) -Ye (x + 1) , Y + 1) -G (x, y + 2) viii) When x-2n-1, y-4n-1 (n: natural number) C_L(X, y) = D (x + 1, y + 1) + D (x, y + 1) + D (x + 1, y) + D (x, y) = Mg (x + 1, y + 1) + G (x, y + 1) + Ye (x + 1, y) + Cy (x, y) C_R(X, y) = D (x, y-1) + D (x + 1, y) -D (x + 1, y-1) -D (x, y) = Mg (x, y-1) + Ye (x + 1, y) ) -G (x + 1, y-1) -Cy (x, y) C_B(X, y) = D (x + 1, y + 1) + D (x, y) -D (x + 1, y) -D (x, y + 1) = Mg (x + 1, y + 1) + Cy (x, y) -Ye (x + 1, y) -G (x, y + 1) [Suppressing Effect of False Color Signal] C described above_LC_RC_BGenerate
The luminance signal C generated by the operation of the circuit 104_LIf
Color difference signal C_RAnd C_BCauses false color signal
You can confirm that it is suppressed as follows:
it can.

As an example, a color difference signal for the pixel (0, 0) will be derived. Pixel (0,0) is x = 2n−
2, y = 4n-4, which corresponds to the case where n = 1.

At this time, the calculation for the case where there is a luminance change H1 in the horizontal direction shown in FIG. 2 is as follows.

C _R (0,0) = D (1,2) + D (0,1) −D (0,2) −D (1,1) = b × Mg + a × Ye−a × G−b × Cy = a × (R + G−G) + b × (R + B−B−G) = − S (ba) (26) Here, since there is no achromatic color component, 2R−G =
0, 2B-G = 0 holds, and 2R = 2B = G = 2S.

Similarly, when the color difference signal C _B (0,0) is calculated, the following result is obtained.

C _B (0,0) = S (b−a) (27) Similarly, in the case where the change in the horizontal luminance level H 2 shown in FIG. 2 exists, the color difference of the pixel (1, 0) is similarly determined. Finding the signal is as follows.

C _B (1,0) = S (b−a) (28) C _B (1,0) = − S (b−a) (29) Regarding a pixel at an edge portion of another luminance signal The same can be said.

Further, when there is a change in the luminance signal level in the vertical direction shown in FIG. 2, when the same calculation is performed, both the color difference signals C _R and C _B become 0, and it can be seen that no false color difference signal is generated. .

As for the color difference signal, a color difference signal corresponding to one pixel is calculated every two rows in the vertical direction. Therefore, when the pixel has V in the vertical direction, the vertical resolution is V / V. It becomes 2. However, the characteristics of the human eye are sensitive to changes in luminance (brightness), but relatively insensitive to changes in color. Therefore, such a decrease in resolution hardly leads to a decrease in image quality.

Therefore, by using the above-described signal processing, a high-resolution image with almost no false colors can be obtained.

In this manner, the luminance signal CL and the color difference signal C _L are applied to all the pixels constituting the CCD 10.
It is possible to generate a _R and a color difference signal C _B.

[0140] Here, as described above, since the CCD output signal is composed is simple relationship between the corresponding pixel (x, y) and the color filter color, C _L the relationship based on
The C _R C _B generation circuit 104 can generate the luminance signal C _L and the color difference signals C _R and C _B by the above-described calculation.

[Separation of RGB Signals] Next, the operation of the matrix circuit 108 shown in FIG. 1 will be briefly described.

[0142] matrix circuit 108, C _L C _R C _B output signal C _L from generating circuit 104, the signal C _R, signals C _B
Then, the following calculation is performed to separate the RGB signals corresponding to all the pixels of the CCD.

[0143] G (x, y) = { 2C L (x, y) -2C R (x, y) -2C B (x, y)} / 10 ... (30) R (x, y) = {C _{L (x, y) + 4C} R (x, y) -C B (x, y)} / 10 ... (31) B (x, y) = {C L (x, y) -C R (x, y ) + 4C _B (x, y)｝ / 10 (32) Accordingly, it is possible to generate RGB signals corresponding to each of all pixels in the CCD 10. For this reason,
Unlike the conventional color separation circuit, the horizontal resolution is H-1 and the vertical resolution is V-1.

Here, -1 is added to the horizontal resolution and the vertical resolution because one pixel (x, y) included in the pixel group of 2 rows and 2 columns in this embodiment is included in the pixel group.
This is because the number of effective pixels in both the horizontal and vertical directions decreases by one column or one row in order to generate a pixel group of two rows and two columns, since the RGB signal corresponding to is generated.

[0145] In the first embodiment shown in FIG. 1 Embodiment 2, color separation circuit 101, C _L C _R C between _B generating circuit 104, read signal D from the CCD10
(X, y), the luminance signal C _L , and the color difference signals C _R and C _B are exchanged, and the RAM 106 is capable of holding these signals for at least two scanning lines.

[0146] However, on the basis of the read signal D (x, y) of all the one frame, after storing in a so-called frame memory, the read signal in the frame memory from CCD10, C _L C _R C _B product The circuit performs processing to generate a luminance signal CL and first and second color difference signals CR and C.
A configuration for generating B is also possible.

FIG. 8 is a schematic block diagram showing a configuration of a single-panel color camera 300 according to the second embodiment of the present invention having the above-described configuration.

Here, the single-chip color camera 300 of the second embodiment is different from the single-chip color camera 100 of the first embodiment.
Of the first embodiment, except that the color separation circuit 101 in FIG. 8 is replaced with the color separation circuit 301 shown in FIG.
Has the same configuration as the single-panel color camera 100 described above.

[0149] The color separation circuit 301, a frame memory 110 for storing all the one frame of the read signal D (x, y) of the CCD10, C _L C _R C _B generating circuit 304
Is configured to receive the data from the frame memory and generate a luminance signal CL and first and second color difference signals CR and CB.
01 configuration. The other points are basically the same as the configuration of the color separation circuit 101 of the first embodiment. Therefore, the same portions are denoted by the same reference characters and description thereof will not be repeated.

That is, the color separation circuit 301 of the second embodiment
Also, the luminance signal CL and the first and second color difference signals CR and CB are generated in accordance with the method described in the first embodiment with reference to FIGS.

[0151] Figure 9, of the structure shown in FIG. 8, C _L C
Is a block diagram showing the configuration of a _R C _B generating circuit 304 in more detail. In FIG. 9, the color separation circuit 301
A signal processing circuit 320 that receives a signal from the matrix circuit 108 and generates a video signal, that is, a luminance signal Y and color difference signals Cu and Cv that have been subjected to well-known γ correction and white balance correction, and a signal processing circuit 320 And an output buffer 330 for receiving a video signal from the external device and outputting the video signal to the outside.

Here, the color separation circuit 301 includes, for example,
It can also be configured as an ASIC (application specific integrated circuit).

Although the frame memory 110 is not particularly limited, when the number of horizontal pixels of the CCD 10 is H and the number of vertical pixels is V, a dynamic memory having a storage capacity capable of holding H × V pixel data is used. It can be composed of a random access memory (hereinafter, referred to as DRAM).

[0154] C _L C _R C _B generating circuit 304 includes an input buffer 306 for receiving data of a corresponding one block into a plurality of pixels necessary for performing a predetermined image processing from the frame memory 110, read from the frame memory 110 A control circuit 308 for controlling the address of the data block;
Under the control of the control circuit 308, the luminance signal CL and the first
And a color-difference signal separation circuit 310 that performs separation processing of the second color-difference signals CR and CB.

Here, as the predetermined image processing, for example, JPEG which is one of the standards of image data compression processing is used.
(Joint Photographic codin
When performing (g Experts Group) encoding, the input buffer 306 stores data corresponding to 9 × 11 pixels as one block.

Since the JPEG encoding is performed using data of 8 × 8 pixels as a unit, the luminance signal CL corresponding to each of the 8 × 8 pixels and the first and second color difference signals CR and CB are used. In order to generate, the first embodiment
This is because, when a signal corresponding to one pixel is generated from the data of 4 × 2 pixels described in the above section, data of 9 × 11 pixels is required.

The input buffer 306 is not particularly limited, but is a static random access memory (hereinafter referred to as S) having a capacity capable of holding the data of 9 × 11 pixels.
RAM).

The color difference signal separation circuit 310
8 and stored in the input buffer 306.
Data of 4 × 2 pixels are sequentially read out of the data of × 11 pixels, and the luminance signal C is read according to the method described with reference to FIGS. 4 to 7 according to the color of the color filter corresponding to each pixel.
L and first and second color difference signals CR and CB.

The matrix circuit 108 receives the output from the color difference signal separation circuit 310 and receives the three primary color signals R and G corresponding to the center positions of each 2 × 2 pixel as described with reference to FIGS.
And B, and the signal processing circuit 320 generates, based on the three primary color signals, a video signal, that is, a luminance signal Y and color difference signals Cu and Cv that have been subjected to well-known γ correction and white balance correction, based on the three primary color signals. I do.

Here, the video signal is represented by the following equation. Y = 0.2990 × R + 0.5870 × G + 0.1140 × B Cu = −0.1684 × R−0.3316 × G + 0.5000 × B Cv = 0.5000 × R−0.4187 × G−0.0813 The × B output buffer 330 converts the video signal output from the signal processing circuit 320 into one block (8 × in JPEG encoding).
(Data for 8 pixels), and outputs it to the outside in block units.

Therefore, the output buffer 330 is not particularly limited, but can be constituted by an SRAM having a capacity capable of holding the data of the above 8 × 8 pixels.

A JPEG encoding circuit (not shown) performs a JPEG encoding process based on the output from output buffer 330, and the encoded data is stored in, for example, a flash memory.

The above processing is performed by the frame memory 11
By repeating the process for one frame of data stored in 0, the compressed data for one frame of image is finally stored in the flash memory.

FIG. 10 shows the color separation circuit 301 shown in FIG.
5 is a flowchart for explaining the operation of FIG.

Referring to FIG. 10, input buffer 206
Is controlled by the control circuit 208 to store a block of processing unit data (data of 9 × 11 pixels) in the frame memory 1
10 (step S200).

Subsequently, the chrominance signal separation circuit 310 outputs data of 4 × 2 pixels for performing chrominance signal separation processing from the processing unit data in the input buffer 306 in accordance with the address specified by the control circuit 308. Is received (step S202).

The color difference signal separation circuit 310
The processing described with reference to FIGS. 4 to 7 is performed according to the color of the color filter corresponding to each pixel, which is determined based on the addresses of the two pixels. That is, in the arrangement of the color filters shown in FIG. 8, the processing corresponding to the x and y values of the pixel (x, y) corresponding to the position at which the color difference signal is to be obtained among the 4 × 2 pixels is performed. (Steps S206 to S21
2).

The matrix circuit 108 receives the output from the color difference signal separation circuit 210, and outputs three primary color signals R, G and B corresponding to the center position of each 2 × 2 pixel (step S214).

Subsequently, the signal processing circuit 320 generates Y, Cu, and Cv signals as video signals based on the three primary color signals (step S216).

The output buffer 330 is connected to the signal processing circuit 22.
0 is accumulated (step S21).
8).

Subsequently, it is determined whether or not the color difference signal separation processing for one block of data (data of 9 × 11 pixels) stored in the input buffer 306 has been completed (step S220). .

When it is determined that the processing for one block of data has not been completed, the processing returns to S202 (step S220).

When it is determined that the processing for one block of data has been completed (step S220), the output buffer 330 outputs Y, Cu,
Cv data is output, and receiving these data, JP
The EG encoding circuit performs a JPEG encoding process (image data compression process) (Step S222).

[0174] The encoded data is stored in the flash memory (step S224).

The above processing is performed by the frame memory 11
Data compression and data storage for one frame of image data are performed by repeating the process for one frame of data stored in 0.

Therefore, it is possible to generate RGB signals corresponding to all the pixels in the CCD 10. Therefore, similarly to the first embodiment, unlike the conventional color separation circuit 200, the horizontal resolution is H-1 and the vertical resolution is V-1, so that the resolution is improved.

In addition, as in the first embodiment, there is an effect that generation of a false color can be suppressed.

[0178]

As described above, according to the present invention,
When the number of effective pixels of the CCD is H in the horizontal direction and V in the vertical direction, an RGB signal can be generated for each of (H−1) × (V−1) pixels. It is possible to realize a higher resolution as compared with. Furthermore, even when there is an edge where the brightness of the subject changes and the luminance level changes abruptly, the generation of a false color signal can be suppressed, and a high-quality image without image quality degradation can be reproduced. .

[Brief description of the drawings]

FIG. 1 shows a single-panel color camera 1 according to a first embodiment of the present invention.
It is a schematic block diagram which shows the structure of 00.

FIG. 2 is a schematic diagram for explaining an arrangement of a color filter array of a color difference sequential system and an operation of independent reading of all pixels.

3A and 3B are diagrams for explaining an arrangement of pixels that output a primary color signal component in the color filter array arrangement illustrated in FIG. 2, wherein FIG. 3A outputs an R signal and FIG. 3B outputs a B signal; 1 shows an array of pixels to be displayed.

Figure 4 is a C _L C _R C first schematic diagram for explaining the operation of the _B generating circuit 104 of the present invention.

5 is a second schematic diagram for explaining the operation of the C _L C _R C _B generating circuit 104.

6 is a third schematic diagram for explaining the operation of the C _L C _R C _B generating circuit 104.

7 is a fourth schematic diagram for explaining the operation of the C _L C _R C _B generating circuit 104.

FIG. 8 shows a single-panel color camera 3 according to the second embodiment of the present invention.
It is a schematic block diagram which shows the structure of 00.

FIG. 9 is a schematic block diagram showing a configuration of a color separation circuit 301 shown in FIG.

FIG. 10 is a flowchart for explaining the operation of the color separation circuit 301.

FIG. 11 is a schematic diagram showing a configuration of an interline transfer CCD.

12A and 12B are schematic diagrams showing field accumulation reading of an interline transfer CCD, wherein FIG. 12A shows a reading operation for an odd field, and FIG. 12B shows a reading operation for an even field.

FIG. 13 is a conceptual diagram showing a color separation process from a conventional color difference sequential color filter array.

FIGS. 14A and 14B are diagrams for explaining the configuration and operation of the low-pass filter. FIG. 14A is a schematic block diagram illustrating the configuration of the low-pass filter, and FIG. 14B is a graph illustrating frequency characteristics of the low-pass filter.

FIG. 15 is a schematic block diagram illustrating a configuration of a conventional color separation circuit.

FIG. 16 is a conceptual diagram illustrating a color filter array of a color difference sequential system and interlaced reading.

FIG. 17 is a diagram illustrating a frequency characteristic of an output signal of the conventional color separation circuit 200.

[Explanation of symbols]

Second optical system 10 CCD 12 exposed portion 14 transferring portion 16 a horizontal transfer registers 100 and 300 single-chip color camera 101, 301 color separation circuit 102 driving circuit 104 C _L C _R C _B generating circuit 106 RAM 108 matrix circuit 200 conventional color separation circuit

Claims (4)

[Claims] 1. A single-panel color camera, comprising: a solid-state imaging device in which photoelectric conversion elements respectively corresponding to pixels are arranged in an array; wherein the solid-state imaging device corresponds to the photoelectric conversion element on a light receiving surface side. A color filter array having a color filter array having green and three kinds of complementary color filters corresponding to four pixels in an arbitrary two rows and two columns, wherein all the pixels are independently read from the solid-state imaging means. Color separation means for receiving the output signal received and generating a corresponding color signal for each of the four pixels, wherein the color separation means stores the output signal sequentially output from the solid-state imaging means. And a luminance signal and a green signal for each of the four pixel groups based on the output signal held in the storage means.
Calculating means for generating a color difference signal corresponding to an intensity difference from a predetermined one of the other primary color signals excluding the green signal among the primary colors, and generating the color signal from the luminance signal and the color difference signal, The arithmetic means includes: i) when photoelectric conversion elements corresponding to complementary color filters in the pixels in the 2 rows and 2 columns that output signals including the predetermined primary color signal components are arranged in a diagonal direction, Generating the color difference signal from an output signal from the pixel group in the 2 rows and 2 columns; and ii) including the predetermined primary color signal component among photoelectric conversion elements corresponding to complementary color filters in the pixels in the 2 rows and 2 columns. In the case where the signal output is not arranged in the diagonal direction, the color difference signal is output from the adjacent two-row, two-column pixel group so as to share two pixels among the two-row, two-column pixels. Generate a single-plate color card La. 2. The image processing device according to claim 1, wherein the output signal from the photoelectric conversion element corresponding to the pixel (x, y) in the y-th row and the x-th column is D (x, y). Output signals D (x, y), D from the pixel group are obtained by a linear operation according to the arrangement of the color filter array.
(X + 1, y), D (x, y + 1) and D (x + 1,
y + 1), and output signals D ((x, y), D (x + 1,
y), a set of D (x, y-1) and D (x + 1, y-1) and output signals D (x, y + 1), D (x + 1, y)
+1), D (x, y + 2) and D (x + 1, y + 2)
And a color difference signal generating unit that generates a luminance signal corresponding to the pixel (x, y), a first color difference signal, and a second color difference signal, and the luminance signal and the first color difference signal. 2. The single-chip color camera according to claim 1, further comprising: a separation operation unit that separates three primary color signals by performing a predetermined linear operation on the second color difference signal. 3. 3. A color filter array comprising: a plurality of first rows including alternately arranged magenta and green filters; and a plurality of first rows including alternately arranged yellow and cyan filters. 2 where n is a natural number, wherein the second row belongs to a row of y = 4n−3 and y = 4n−1, and includes a yellow filter in an even column of x. The first row belongs to the rows with y = 4n-4 and y = 4n-2, and if it belongs to the row with y = 4n-4, x has a green filter in an odd column, and y = 4n- When the pixel belongs to row 2, x has a green filter in an even column, and the color signal generation unit outputs an output signal D (x, y) from a pixel group of row 2 and column 2 including the pixel (x, y). ), D (x + 1, y), D (x, y +
1) and D (x + 1, y + 1) as a first pixel group signal, and output signals D (x, y-1), D (x + 1, y-1), D
A set of (x, y) and D (x + 1, y) is defined as a second pixel group signal, and output signals D (x, y + 1), D (x + 1, y + 1), D
When a set of (x, y + 2) and D (x + 1, y + 2) is used as a third pixel group signal, a luminance signal C _L (x, y) corresponding to the pixel (x, y)
y), the first color difference signal C _R (x, y) and the second color difference signal C _B (x, y): i) x = 2n−2, y = 4n−4 or x = 2n−
From 1, y = 4n-4 when C _L (x, y) and C _B (x, y) of the first pixel group signals, from C _R (x, y) of the third pixel group signal Ii) x = 2n-2, y = 4n-3 or x = 2n-
1, y = 4n-3 where C _L (x, y) and C _R (x, y) from the first pixel group signals, C _B (x, y) from the second pixel group signal Iii) x = 2n-2, y = 4n-2 or x = 2n
-1, y = 4n-2 of the case C _L (x, y) and C _R (x, y) from the first pixel group signals, C _B (x, y) of the third pixel group signal Iv) x = 2n-2, y = 4n-1 or x = 2n-
1, when y = 4n-1 C _L (x, y) and C _B (x, y) are obtained from the first pixel group signal, and C _R (x, y) is obtained from the second pixel group signal. 3. The single-panel color camera according to claim 2, wherein the color camera is generated. 4. Arbitrary two rows and two rows arranged in a color difference sequential system
Separation of Luminance Signals in a Single-Plate Color Camera Including a Color Filter Array Having Green and Three Complementary Color Filters Corresponding to Four Pixels in a Column and a Solid-State Image Sensor in which Photoelectric Conversion Elements Corresponding to Pixels Are Arranged in Array A method for obtaining a luminance signal from a sum of the output signals from the pixel group for each of the four pixels based on an output signal from the photoelectric conversion element; and for each of the four pixel groups. A) a color difference signal corresponding to the intensity difference between the green signal and a predetermined one of the other primary color signals excluding the green signal among the three primary colors: i) photoelectric conversion corresponding to the complementary color filter in the pixels in the 2 rows and 2 columns When the elements that output signals including the predetermined primary color signal component are arranged in a diagonal direction, the color difference signal is generated by an output signal from the pixel group of 2 rows and 2 columns, i. If the photoelectric conversion elements corresponding to the complementary color filters in the pixels in the 2 rows and 2 columns that output signals including the predetermined primary color signal components are not arranged diagonally, the 2 rows and 2 columns Generating the color-difference signals from output signals from adjacent two-row, two-column pixel groups so as to share two of the pixels; and generating the color signals from the luminance signal and the color-difference signals. A method for separating color signals, comprising: