JPS61206256A - Color solid-state image pickup device - Google Patents

Abstract

PURPOSE:To optimize the spectral-response characteristics of a luminance signal by forming a color filter for a single plate color solid-state image pickup element by a dyed layer by two complementary colors such as Ye and Cy. CONSTITUTION:Patterns for a color filter are repeated while using eight picture elements in total of patterns Cy1, Cy2, Ye1, Ye2, G1-G4 for the color filter as a minimum unit. The title device is divided into green filter sections 41 coating the whole surface, photosensitive sections 42 for Cy picture elements, openings 43 for Ye dyed layers, openings 44 for Cy dyed layers, openings 45 for Ye dyed layers, etc. Odd lines in Odd fields form the repeated signal rows of (Cy1+Ye1), (G1+G2), (Cy1+Ye1), (G1+G2)..., and even lines shape the repeated signal rows of (Cy2+G4), (G3+Ye2), (Cy2+G4), (G3+Ye2).... These signal rows are employed as luminance signals by band-limiting signal rows made to be contained in one lines to reading frequency or less. The same applies to both lines in even numbered fields.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a color solid-state imaging device, and particularly to improvements in the color separation filter portion of a device used in a single-chip color camera.

[Technical background of the invention and its problems]

In general, solid-state image sensors: - In order to equalize the sensitivity of each pixel, the amount of light incident on the photosensitive portion of the semiconductor substrate is limited by a fixed aperture of an optical seal r made of aluminum or the like that covers the top surface of the substrate. Do a.

A sectional view thereof is shown in FIG. 11 in the figure is a semiconductor substrate;
12 is an At layer, and 13 is a 9th layer made of red.

A single-plate color solid-state image sensor usually has two
Color filters with spectral transmittance characteristics that exceed color are repeatedly formed. Here, in a conventional single-plate solid-state image sensor, a single color filter is formed for one photosensitive pixel in a semiconductor substrate. As a specific example, Y・(yellow),'
Two complementary color filters of Cy (cyan) are m- and Ye
* Cross-sectional structure kl/E of a color device using a single-plate color multiplexing method in which three color filter layers of Cy #G (green) are formed
Figure 14 KyF. In the figure, 21 is the C7 staining layer, 22 is the Y/
Staining layer, 23 E photosensitive pixel, 24 G photosensitive pixel, 25
is a C7 photosensitive pixel.

As another conventional example, in the color filter array of the color difference sequential method, there are actually many parts in which two color filters are not provided within one pixel. The purpose of this is that when extracting a color difference signal as a modulation component in a horizontal signal line, if there is a color component, the modulation component is generated as a color difference signal every horizontal period, but Regardless of the color, the luminance component is made equal for each horizontal period.

Here, the above-mentioned conventional technology has the following problems (-11 to 2). It is necessary to form a color filter with unique spectral transmittance characteristics for the unique spectral characteristics of the photosensitive pixels of the element, and the total spectral sensitivity characteristics of each color pixel are determined arbitrarily by the characteristics of each dyeing layer. The method for controlling the total spectral sensitivity characteristics of these color pixels requires controlling the spectral transmittance characteristics of the dyed layer or the spectral sensitivity characteristics of the light-sensitive pixels of the solid-state image sensor.

For the purpose of suppressing the occurrence of false signals, the brightness signal of a single-chip color camera is ideally taken from all the light-sensitive pixels of the solid-state image sensor. The surface of the photosensitive pixel of the solid-state image sensor is covered with a multiplexing color filter. In order to use all pixel signals as a luminance signal, including the spectral characteristics of the color filter, it is necessary to limit the band of the horizontal readout signal to a band below the color multiplex frequency, but the spectral sensitivity characteristics appropriate for a luminance signal In order to obtain this, a visibility correction filter is used to correct the distribution of incident light. If that processing is insufficient, it is common to electrically calculate the red and blue signal components of the luminance signal, and the S/N of the signal is
This results in a decrease in the ratio.

(3) In the color filter of the color difference sequential method, Ye +Vh
, Cy, G (D 4 color combination and Re
The advantage of forming a color filter by combining the four colors of G + Cy * G is that it is not possible to form a color filter with only two specific colors, that is, two dyed layers, so it is possible to form a color filter by overlapping two layers, etc. However, it is difficult to develop such a method, and in the end, filters of three or more colors are required, which is accompanied by technical difficulties in manufacturing.

[Purpose of the invention]

In most solid-state image sensing devices that are color-modulated (by carrier frequency), the main portion of the luminance signal is obtained by band-limiting the signal for one horizontal period. The spectral sensitivity characteristic of the luminance signal is the product of the spectral sensitivity characteristic of the solid-state image sensor before forming the color filter and the average transmittance characteristic of the gold filter present in one horizontal line. To control the spectral sensitivity characteristics of the luminance signal, you can either install a correction filter in the optical path as described above, add or subtract the red and blue signals after color separation to the luminance signal obtained by the above method, or filter the luminance signal using a color filter. This allows direct control of the spectral transmittance characteristics.

Therefore, the present invention aims to optimize the spectral sensitivity characteristics of luminance signals by minimizing the addition and subtraction of the signals, controlling the spectral transmission characteristics of color filters, and minimizing the use of visibility correction filters in the optical path. .

[Summary of the invention]

The color filter of the single-plate color solid-state imaging device realized by the present invention is formed from dyed layers of two complementary colors, for example, Ye and Cy. When a single Y and ay dye layer covers a photosensitive pixel, assuming that the spectral characteristics of the solid-state image sensor are flat, the spectral sensitivity of the pixel reflects the spectral transmittance characteristics of the dye layer. When both Ye and cy layers overlap on the same photosensitive pixel, that pixel will have more of the spectral sensitivity characteristic of G. In fact, photosensitive pixels are separated on the surface of the semiconductor substrate, and pixels with spectral sensitivities of Y and ay are CyeY
*Made with openings in two dyed layers. That is, Y
They are formed so that there is no Ay layer above the pixel having the E sensitivity characteristic, and there is no Ye layer above the pixel having the Cy sensitivity characteristic. Here, the average spectral sensitivity of one horizontal line becomes the spectral characteristic of the luminance signal. In fact, the number ratio of color filter pixels provided in this invention is Y: C,: G〒1:
The ratio is 1:2.

In the present invention, for example, when forming Y pixels, K C
The aperture provided in the seventh dyeing layer was made smaller than the actual photosensitive pixel size, so that the Y pixel had a shape as if it were surrounded by G pixels. The spectral sensitivity of K and higher luminance decreases in the red band. In other words, in consumer video cameras, the light source used for standard setting is 3100 to 5000, which has a large red component. This means that it is sufficient to use a thinner one, that is, one that has a weaker effect.

[Embodiments of the invention]

In realizing the present invention, an interline transfer type COD with an image size of A-former or T was used as a solid-state image sensor. FIG. 1 shows a schematic diagram of the imaging plane. 31 in the diagram
3 is a photosensitive pixel, 32 is a charge transfer vertical CCD, 33 is a charge transfer water "FCCD", and 34 is a semiconductor substrate.

The CCD image sensor was driven in a field P accumulation mode in which the signals of the upper and lower two pixels were added and read out.

Figure 2 shows the driving method. Figure 2 (al is a conceptual diagram of charge movement during odd field readout, Figure 2 (b)
) is a conceptual diagram of charge movement during even field readout. That is, on the imaging surface, interlaced readout is performed for each field by shifting the combination of pixels to be added up and down by 1 bi between odd and even fields.The color filter considered in the present invention FIG. 3 shows the shape of the imaging surface of the array before implementing the present invention. As shown in the figure, the color filter /translation Cyl −Cy2 *Yel, Y
・2. In the figure, 41 is the green filter part covering the entire surface, 42 is the photosensitive part of the cy pixel, 43 is the opening of the Ye dyeing layer, and 44 is the opening of the cy dyeing layer. The opening 45 is an opening in the Ye dyeing layer. When read out as shown in FIG. 3, the odd lines of the odd field are (Cyl+Y・1), (Gl+
G2), (Cyl+Y・1).

(Gl +G2) ,...) return signal string, the even line of the odd field is (Cy2
+ G4) e(G3+Ye2), (Cy2+
G4), (G3+Y*2) = ''. By band-limiting the signal strings included in the l line to below the readout frequency, these signal strings can be used as luminance signals. The same process can be performed on both sides of the busy-even field P, Y, pixel,
Spectral sensitivity characteristics of Cy pixel and G pixel are respectively Y・(λ)
ecy (λ) and G (λ), the spectral sensitivity characteristic Y (λ) of luminance is Y (λ) = (Ys (λ) + Cy (
J)+2G(λ))/2.

Using the same color filter array as shown in FIG. 3, FIG. 4(a) shows the planar structure of the imaging surface of the color filter realized by the present invention, and FIG. 4(bJK) shows the cross-sectional structure.

This is an example of a case corresponding to the conventional FIG. 14. In FIG. 4, 51 is an opening in the Y dyeing layer, 52 is an opening in the C7 dyeing layer, 53 is an opening in the Y dyeing layer, 54 is the photosensitive area of the Ye pixel, 55 is the photosensitive area of the G pixel, and 56 is the cy pixel. This is the photosensitive part of the The features of this configuration are as shown in the figure.
It is made by providing a Cy-dyed layer with an aperture smaller than the effective photosensitive area.

That is, the Ya pixel is divided into a portion a where a filter with a Y@ spectral transmittance characteristic is formed and a portion a where a filter with a G spectral transmittance characteristic is formed.

Assuming that the area ratio of the steep part and the part b is A:B, the spectral sensitivity characteristic Y1λ) of the luminance signal in the color image sensor considered according to the present invention is rltλ)=["C7(λ)+
TFM” 'λ)+G(λ)+2G(λ]]/4...
...(11) This second term or shim containing the red component of equation (11)
In the term Y・(λ)/(A+8), the spectral sensitivity characteristic Y'(J
), the sensitivity to the red component gradually decreases.

When photosensitive pixels of each color are formed using a conventional filter structure, if a luminance signal is obtained by band-limiting the signal read out in one horizontal period, the spectral characteristic of the luminance signal Y (
λ) is the average of the pixel outputs of each color, and is expressed by the following formula.

Y(λ)=CYs(λ)+CF(λ)+2G(λ)
]/4 Each color image If the general equal-energy spectral sensitivity characteristics of each color are shown in FIGS. 5 to 7, the luminance spectral characteristics Y(λ) of the device with this color filter structure are shown in FIG. 8.

When actually used in a color camera, the color temperature is 300.
In many cases, a white light source of about 0.0 is used, and in reality the amount of light in the red band is very large.In order to eliminate the imbalance between the signal energy of the red band and the front band, 01 color visibility correction is performed. A filter is installed in the optical path to reduce the actual amount of light in the red band. On the other hand, the Ye0 pixel spectral sensitivity characteristics of a device in which the aperture of the cy dye layer realized by the present invention is made smaller than the size of the photosensitive pixel to form a Ye color pixel are as follows:
The whistle can be controlled as shown in Figure 9.

If the aperture of the ay dye layer is larger than the size of the photosensitive pixel, the situation is as shown in FIG. 6, but as the size of the photosensitive pixel becomes smaller, the sensitivity of the Ye0 pixel red decreases. If the size of the cy dye layer is used as the size of the photosensitive pixel, the Ye0 pixel spectral sensitivity will be as shown in FIG.

Furthermore, when the structure of the color filter according to the present invention is adopted, false color signals between pixels of different colors, where the light-sensitive pixel is exposed to light multiplied by the filter of the other color pixel, are generated. It can be reduced. In Figure 11,
This figure shows how Ye light that has passed through the Y pixel color filter of a conventional device leaks into the G pixel on the right side. That is, according to the filter structure of the present invention, the aperture of the cy dye layer on the Ye pixel is 74% smaller, and less Y light leaks into the G pixel on the right. As much as the amount of light leaking is reduced,
It is possible to widen the aperture, such as kl, which restricts the aperture of a photosensitive pixel.

[Effects of the Invention] As explained above, according to the present invention, the spectral sensitivity characteristics of the luminance signal can be optimized, so that conventional signal addition/subtraction, control of the spectral transmittance characteristics of color filters, and control of the visibility correction filter in the optical path can be performed. A color solid-state photographic device can be provided which has advantages such as minimizing usage and reducing light leakage between adjacent pixels.

[Brief explanation of drawings]

The figures are for the purpose of explaining one embodiment of the present invention.
s) uses this CCD solid-state imaging device in a field storage mode.
Fig. 2 is a conceptual diagram showing the charge movement state in an odd field when driven with (BL is a conceptual diagram showing the charge movement state in an even field), and Fig. 3 is an arrangement of color filters considered in the present invention. A plan view of the imaging surface before implementing the present invention in
FIG. 4(a) is a plan view of the imaging surface of the embodiment of the present invention;
Figure (b) is the same cross-sectional view, Figure 5 is the spectral sensitivity characteristic diagram of the cy pixel, Figure 6 is the spectral sensitivity characteristic diagram of the Y pixel, Figure 7 is the spectral sensitivity characteristic diagram of the G pixel, and Figure 8 is the spectral sensitivity characteristic diagram of the G pixel. A spectral characteristic diagram when a luminance signal is created from the output signal of each pixel with the spectral sensitivity characteristics shown in Figures 5 to 6, and Figure 9 is a spectral characteristic diagram of a Y pixel with a small aperture of the Cy dye layer. Figure 10 is a luminance spectral sensitivity characteristic diagram of a color element with a small aperture in the cy dye layer;
The figure is a conceptual diagram of light leakage when a color filter is formed using the conventional method, Figure 12 is a conceptual diagram of light leakage when a color filter is formed according to the present invention, and Figure 13 is a conceptual diagram of light leakage when a color filter is formed using the present invention. FIG. 14 is a sectional view of a black-and-white solid-state device before forming a filter, and FIG. 14 is a sectional view of a conventional color solid-state device. 11... Semiconductor substrate, 21... cy dyed layer, 22...
...Y6 dyeing layer, 54...ye pixel photosensitive area, 55.
...Photosensitive part of G pixel, 56...Photosensitive part of C7 pixel. Applicant's agent Patent attorney Takehiko Suzue Figure 1 Figure 4 (b) 1+ Munekikaku〈zuφ枳4〈 Figure 11 Figure 13 Figure 14

Claims (1)

[Claims] a first dyed layer having a first spectral transmittance characteristic;
A second dyed layer having a spectral transmittance characteristic of A color solid-state imaging device characterized by being formed by overlapping.