JPH03173287A - Color image pickup device - Google Patents

Abstract

PURPOSE:To improve an S/N and brightness reproducibility by placing priority on a 2nd luminance component with excellent S/N when noise is eminent in the state of an object and placing priority on a 1st luminance component with excellent brightness reproducibility when noise is not eminent. CONSTITUTION:A low pass filter 121 is provided, which extracts a low frequency component of a signal subject to A/D conversion (103) from a signal of a sensor 101 to form a 2nd luminance signal. A signal from the 121 is gamma-converted by a gamma converter 119 to be a 2nd luminance signal YL2 and it is inputted to a mixer circuit 123 together with a conventional luminance signal YL1. The mixture ratio depends on color difference signals R-Y, B-Y by an evaluation circuit 122, a sensor output of a white balance sensor 120 giving color temperature information and a signal from an output level signal generating circuit 124. Thus, the 1st luminance component with bad S/N and excellent brightness reproducibility and the 2nd luminance component with excellent S/N and bad brightness reproducibility are mixed to place priority on the brightness reproducibility or the S/N as required to improve the quality of the picture.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an imaging device such as a color video camera or a color still video camera, and particularly relates to the generation of a luminance signal.

(Prior Art) Conventionally, in this divine device, a solid-state image sensor r is equipped with three or more color filters as shown in FIG.
By performing signal processing as shown in FIG. 8, the luminance signal and the two color difference signals R-Y and B-Y are finally converted into IB? This is Nishi-dori.

In such conventional color signal processing, first,
In this process, arithmetic processing is performed from a color difference signal obtained by subtracting the IX power from pixels that are adjacent in one direction and are equipped with color filters of different types. For example, if the color filter array shown in FIG.
4, a subtraction result of C1=(Mg-Gr) is obtained, and the even number column [1 obtains a subtraction result of C2=(Ye-Cy). In contrast, the color signal processing unit 805 performs color processing calculations such as white balance and γ conversion using an appropriate method.

Next, these line-sequential color difference signals CI/C2
The synchronization circuit 806 synchronizes them using an IH (water scanning time) delay line, etc., and passes them through the color difference matrix circuit 807 to appropriately rotate the color difference axes, and finally Two color differences No. 111 RY and B-Y are obtained.

However, this type of color processing method has the following two fundamental problems.

(A) White balance is difficult to achieve.

For tube type cameras and RGB primary color (pure color) type cameras,
White balance is achieved by changing the ratio of R (red) and B (1't) to G (edge) according to the color temperature, but in this type of device, color information is based on the shape of color difference. For example, depending on the color temperature, some percentage of the luminance signal is added to or subtracted from the color difference signal to force the color difference signal for white to zero, thereby achieving white balance. This method is incorrect in principle, and it is extremely difficult to achieve accurate white balance over a wide color temperature range.

(B) Color reproducibility is poor because γ conversion is performed without changing the color difference.

For RGB primary color type cameras, NTSC
system (national television system)
The color-separated outputs R, G, and B are multiplied by γ to obtain Rr, a r, and Br, and then two color differences R'-Y and B'-Y are obtained. However, Y (luminance (u'j) haY =0.30Rr+0.5
9G r+0. IIB r”'C.

However, in a complementary color type camera, the color signal is first subtracted and then multiplied by γ (Mg-Gr).
Like r, the difference is multiplied by γ. Therefore, no matter how much correction is made later, a color signal corresponding to the regular NTSC cannot be obtained, and the color quality is not good.

In order to solve the problem mentioned above, for example, as shown in FIG.
One conceivable method would be to use the luminance signal YL' that has passed through the .

According to this, white balance in R, G, and H states,
Since the γ conversion can be performed, the problems such as the example shown in FIG. 8 described above can be solved to some extent.

However, in the -L method, which creates a color difference based on the -degree horizontal output difference and then performs color processing based on this,
There was a problem in that color processing that matched the spectral sensitivity of the filter could not be performed, resulting in poor color reproduction.

Therefore, in order to improve color reproducibility, the following digital processing methods can be considered.

FIG. 10 shows its block diagram.

Four types of color filters as shown in FIG. 7 are arranged in the CCD sensor (1! image element) 101.

The image signal read out pixel by pixel from the sensor 101 using interlaced scanning is! 02 AGC (auLcooa
After the amplitude is adjusted by the tie gain control), the A/D converter]03 sets the readout clock to [-
Analog-to-digital conversion is performed at the timing of U1 of the month. For color processing to be performed later, this A/D converter 103 uses a linear characteristic, quantity? - (From the point of view of E error, 8bi
It is desirable to carry out the test at t or more.

A high frequency component of the luminance signal is detected by a bypass filter (luminance signal high frequency component generating means) 116.

Low-frequency component Y of luminance obtained by the method described later
, are added by an adder 117, digital-to-analog converted by a D/A converter 118, and output.

On the other hand, the output of the A/D converter 102 is manually input to four interpolation filters 106, 107, 108, and 109, respectively, and is synchronized to produce color signals Mg and cy.

Ye, Gr (magenta, cyan, yellow, green).

These colors (RGB for +3) are input manually to the conversion section 110 and converted into R, G, B, 3f, and 4. This is done by the following matrix calculation.

Here, matrix A is sensor 10! Mg, Gr, C
y, spectral characteristics Mg(λ) of Ye.

Gr (λ), Cy (λ), Ye (λ) are RGB ideal spectral characteristics R (λ), G (λ) defined by the NTSC standard.
, B(λ).

Next, the white balance unit 111 determines the white balance by converting the RGB signal from R, G, and B to αR, G, and βB based on the color temperature information obtained by the white balance sensor 102. .

Next, the γ conversion unit 112 performs table conversion to convert RGB
The signal is gamma transformed.

The color difference matrix unit 113 performs conversion in accordance with the NTSC standard, and converts the multiplexed luminance low-frequency components Y, 4 and the color difference signal R-Y.

Something like B-Y is generated. The color difference signals R-Y and B-Y are subsequently digital-to-analog converted by D/A converters 114 and 115 and output. Low frequency component Y of luminance signal
, is added to the high-frequency component Y of the luminance signal detected by the bypass filter 116 in an adder 117,
A D/A converter 118 performs digital-to-analog conversion and outputs the signal.

(The invention seeks to solve this problem: The low-frequency component YL of the luminance signal obtained from equation (1) is N
RG B (;N
White balance section 111, which was created from the No.
Since the spectral characteristics of brightness match the visibility of the eye, the brightness is adjusted to No. 13 with good brightness reproducibility.

However, this luminance signal Yl, is
Noise is mixed in by processing such as RGB conversion at the subsequent stage of 06 to 109.

That is, this signal YL has a normal luminance i1T characteristic, but is characterized by a poor S/N ratio.

The present invention was made under these circumstances.
The object of the present invention is to provide a color imaging device with good brightness reproducibility and inconspicuous noise.

[Means to solve the problem]

In order to achieve the above objects, the present invention configures a color imaging device as shown in 1) to (4) below.

(1) The first luminance component obtained from the image sensor f output and the second luminance component obtained from the image sensor f output.
the first luminance component has a lower SN and better luminance reproducibility than the second luminance component; is a color imaging device that has good SN and poor brightness reproducibility compared to the first brightness component.

(2) In the color imaging device according to (1) above, the mixing ratio of the mixing means is variably set according to the color M information of the subject.

(3) +? In the color imaging device according to (1) of i+-, the mixing ratio of the γU combining stage is variably set according to brightness information of the subject.

(4) In +ii (1), 7u of the correction mixed means
A color imaging device in which the combination ratio is variably set according to the color information and brightness information of the subject.

[Effect]

In the present invention as described above, the first luminance component has poor SN or good luminance reproducibility, and the first luminance component has good SN or good luminance reproducibility.
By mixing the second luminance component with poor reproducibility, the luminance reproducibility can be improved to SNf! : It is possible to improve the image quality by %.

In addition, by setting the mixing ratio by 1 degree depending on the color information and/or brightness information of the subject, the brightness can be adjusted to 1 degree depending on the subject.
By giving priority to either the brightness or the SN, it was possible to give priority to the luminance reproducibility when there was no noise or noise, and to give priority to the SN when the noise was noticeable.

〔Example〕

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

FIG. 1 is a block diagram of a "color imaging device" which is a first embodiment of the present invention. Portions corresponding to those in FIG. 1O are designated by the same reference numerals, and description thereof will be omitted here.

In this embodiment, in order to generate the second luminance signal, a low-pass filter 121 is provided for deriving the low frequency component of the analog-to-digital converted signal from the sensor 101.

The signal obtained by the low-pass filter 121 is transmitted to the γ converter 11
9 and becomes the second luminance signal YL2, which enters the mixing circuit 123 together with the luminance signal YL of 7JIJl, which is equivalent to the luminance A3 YL in FIG. The mixing ratio here is determined by the evaluation circuit 122, which receives the color difference signals R-Y, B-Y and the color temperature information, and calculates the white and balance sensor 1.
20 and a signal from the output level signal generation circuit 124.

The M5J generation circuit 124 is configured as shown in FIG.

Same 1. ! At tJ, the low-frequency component YL' of the luminance signal output from the low-pass filter 121 is clamped at a certain threshold value T1. This threshold value T is the threshold value r of the comparator 204 (see FIG. 2) in the evaluation circuit described below.
Same as ef. The clamped signal is sent to AGC circuit 1
The average output level signal YD of the sensor 101 obtained from 02 is added to the adder 803, and outputted as an output level signal YD.

By the way, the second luminance signal is an output obtained by switching the output of each pixel and reading out the signal as it is from the sensor 101 equipped with the color filter shown in FIG. .

Therefore, the spectral characteristics are Mg+Or for each line.

It is equal to the average of the transmittance of the filter, which is Ye+Cy. Therefore, the second luminance signal has a spectral characteristic (unlike the visibility of the eye, the luminance reproducibility is 6 to λ), and on the other hand, the second luminance signal undergoes calculations such as RGB conversion. Since this process is not carried out, the S/N ratio is good.

The present invention takes advantage of both the first luminance signal with good luminance +Ii ISi +i but poor S/N and the second luminance signal No. 43 with poor luminance + IG performance but good S/N. It is a combination. That is, by using the object color +lIi information waiting in the processing system No. 5 and the information on the incident light ID obtained from the output level No. 5 generation circuit 124, When the saturation is high, priority is given to the first luminance I word with good visibility (
In addition, if noise is likely to occur (in other words, if the color saturation is low), the second luminance signal with good SN is given priority, and apart from the color saturation, the amount of incident light is If the amount of incident light is too much (bright) and the noise is not noticeable, the first brightness signal is given priority, and if the amount of incident light is small and the amount of noise is high, the second brightness signal is given priority.

Furthermore, in this embodiment, when the color temperature becomes high, that is, when the noise in the subject of the front thread is noticeable,
When there are many i\\-based colors, the second luminance signal with good SN is given priority, and when there are many red-based colors, the first luminance signal with good luminance reproducibility is given priority. On the other hand, when the color temperature is low, that is, when noise in a red subject is noticeable, the first luminance signal is prioritized when there are many colors in the front thread,
When there are many red colors, priority is given to the second luminance signal.

In this embodiment, each element (color saturation level 1
The evaluation circuit 122 and the iu combining circuit 123 are provided to change the mixing ratio of the first and second luminances (Z) according to the color temperature and light amount.

The evaluation circuit 122 receives color saturation, color temperature T, ie color information, and output level signal Y. is evaluated and the mixing ratio M is determined. That is, M=F, (RY, BY, T, Yr,)...
(2) In Figure 2, the mixing ratio M is M + , M21
An evaluation circuit that takes four values M3M4 is shown.

In the same I'XI, the output AW from the white balance sensor
B or comparator 201 manually, a certain threshold ref
If the color temperature is higher than the color temperature given by , a "1" signal is output, and if it is lower than the color temperature, a "0" signal is output. Similarly, the color difference A and ; of B-Y and R-Y and the output level signal YD are also output to the comparator 2.
02.203°204 manually, both at a certain threshold ref! If the signal is higher than the level of the signal A or --, a signal of "1" is output, and if it is smaller, a signal of "0" is output.

The output results are as shown in the table below.

The mixing ratio is set such that the ratio of the low frequency component of the first luminance signal increases in the order of Mn<M3<M2<Ml.

FIG. 4 shows a block diagram of a "degree color imaging device" which is a second embodiment of the present invention.

In this embodiment, as shown in the figure, the mixing ratio M is determined by the evaluation circuit 12.
2, color signals R, G, B and output level 1, No. 1 Y.

is determined using That is, M=F, + (R,
G, B, Yo) """ (3) In Figure 5, the mixing ratios are M, , M2, M3.M,.

M6. Evaluation circuit 1 that takes eight values of M, , M7, and Mn
22 is shown.

In the figure, white balance-adjusted color signals B, G, and R are shown.
input to the calculation circuits 601 and 602, and the difference number B-G
, R-G are created. This difference No. 13 is comparator 60
3,604, and the threshold r with their absolute value
When it is compared with ref and is smaller than the threshold ref in both cases, the mixture ratio M is output as Ml. The color signal at this time is close to an achromatic color. Let W be a set of signals where M=M. Signals that do not belong to W are classified into the following F regions by the evaluation circuit 122 in FIG.

(R>Gf”IR≧B)nWn (Yo>Tr)”
' (4) (G ≦Rn G>R) nWn (
Yo>Tr) −(5)(B>R>G)nW
n (Yo > Tr) − (6) (B≧G≧R
)nWn (Yo > Tr) ”” (7)
(R>GnR≧B)nWn(Y,≦TI”)−(8)
(G≧RnG>B)nWn (Y,≦T I )−(9
) (B>R>GUB≧G≧R) nWn(Y,≦TI)
・-(+0) (4), (5), (6), (7),
(8) In the regions shown by (9) and (10), the mixing ratio M is M2, M3. Mn, M5. Mt...
It becomes M7M8. When the region where the mixing ratio M takes a value of M, to M8 is shown in the L"u"■0 color space, it becomes as shown in FIG.

In addition, in the following examples, the mixing ratio M was set as a discrete value of 1j, but the present invention is not limited to this.
For example, in the configuration shown in FIG. 1, CI and C2 may be given as continuous values such as proportionality constants.

In addition to the complementary color mosaic type shown in FIG. 7, the sensor of the present invention may also have complementary color stripes. Even if the color filter is of the RGB type, the present invention can be applied as long as the RGB signals are regenerated by matrix conversion or the like.

Furthermore, the RGB conversion shown in FIG. 1 etc. is not limited to the linear attack conversion such as the so-called affine transformation using a matrix, but also the R=FR(c+, C2, -=・Co)−・−・−(9
)G=FC(CI,C2,・=−Co)・・”−(
10) B-FR (C1, C2,...Cn)...
... (11) However, Lc i (i = 1, -,
The present invention is similarly applicable even if n) is a non-M''J transformation of the color form of the color filter of the sensor.

Note that the first aspect of the present invention. As the second luminance component,
1) The information is not limited to those described in 4, and may be extracted by other methods.

(Effects of the Invention) As is clear from the above description, according to the present invention, when the subject state has noticeable noise, priority is given to the second luminance component with good SN, and when the noise is not noticeable, the luminance is reproduced. By prioritizing the first luminance component with good brightness, it is possible to improve both the SN and luminance reproducibility for human visual perception.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a first embodiment of the present invention, FIG. 2 is a circuit diagram of an evaluation circuit in the same embodiment, FIG. 3 is a block diagram of an output level signal generation circuit in the same embodiment, and FIG. A block diagram of a second embodiment of the present invention, FIG. 5 is a circuit diagram of an evaluation circuit in the same embodiment, and FIG. 6 shows an area in 1llu*■@color space divided by the evaluation circuit in the same embodiment. 7 are diagrams showing the arrangement of color filters, and FIGS. 8, 9, and 10 are block diagrams of conventional examples. 101... CCD sensor (imaging device) 106~
109...Interpolation filter 110...-RG
B conversion unit 113... Color difference matrix unit 17... Adder 20... White balance sensor 2...
...Low pass filter 22...Evaluation circuit 23...Mixing circuit

Claims (4)

[Claims] (1) The first luminance component and the second luminance component obtained from the image sensor output
the first luminance component has a lower SN and better luminance reproducibility than the second luminance component, and the second luminance component has a better luminance reproducibility than the second luminance component; 1. A color imaging device characterized by having better SN and poorer luminance reproducibility than the luminance component of No. 1. (2) The color imaging device according to claim 1, wherein the mixing ratio of the mixing means is variably set according to color information of the subject. (3) The color imaging device according to claim 1, wherein the mixing ratio of the mixing means is variably set according to luminance information of the subject. (4) The mixing ratio of the mixing means is variably set according to color information and brightness information of the subject.
The color imaging device described.