JPS6189791A - Solid color image pick-up device - Google Patents

Abstract

PURPOSE:To execute the high quality color pick-up by executing registration correcting processing to respective chrominance components with an electrical signal processing means based upon the said correcting data while the corrected data formed by the previous interpolating processing read at the time of an actual image pick-up action. CONSTITUTION:When an actual pick-up is operated, a control part 10 reads respective color registration corrected data Dcr and Dcb from corrected data memories 15 and 16, supplies through decoders 17 and 18 to a registration correcting processing part 6 and controls an action. For the image pick-up output obtained by respective CCD image sensor 3R, 3G and 3B of an image pick-up part 3 by operating the registration correcting processing part 6 in accordance with respective coefficients (a) and (b) and h-k given to the control part 10, a highly accurate registration adjustment is executed by a signal processing and a high quality color image pick-up is executed, when the image pick-up action is actually performed.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention uses a charge coupled device (CCn) as an imager for photographing a subject.
The present invention relates to a solid-state color imaging device using a solid-state image sensor such as an imaging optical system, and particularly to one in which an error in superimposition (so-called registration) of images of each color occurring in an imaging optical system is corrected by electrical signal processing.

[Prior art] In a television camera for color photography, a subject image is separated into three optical images of red, green, and blue, and the color image is created by superimposing the images of each color. It is designed to be reproduced. Therefore, the superimposition error between images of each color must be extremely small since it appears as color blurring in the reproduced color image.

Conventionally, color television cameras using image pickup tubes
In order to eliminate the above-mentioned image overlay error, so-called registration adjustment is performed by mechanically adjusting the position of each imaging surface and electrically adjusting the deflection system. In addition, in solid-state color television cameras that use solid-state image sensors such as COD, the imaging area cannot be adjusted using a deflection system like in image pickup tube cameras, so mechanical registration adjustment must be performed with high precision. Ta. Furthermore, as a method for electrically adjusting the registration of a solid-state color television camera, a technique for controlling the frequency of driving pulses of a solid-state image sensor was disclosed in Japanese Patent Laid-Open No.
This is shown in, for example, Japanese Patent No.-59690.

[Problems to be Solved by the Invention] Generally, in a color imaging device, there is an error in overlaying images of each color, that is, a registration error due to chromatic magnification aberration of the imaging optical system. Assuming that the overlay error due to the chromatic magnification aberration is zero at the center of the screen, the amount of error tends to increase as the distance from the center of the screen increases. Therefore, there is a problem in that simply by eliminating the overlay error at the center of the screen through mechanical registration adjustment, it is not possible to eliminate the overlay error caused by the -F color scheme magnification aberration.

In addition, in a solid-state color imaging device, if registration adjustment is performed by controlling the frequency of the driving pulse of the solid-state image sensor, the time axis of the imaging output read from the solid-state image sensor will fluctuate, so the signal processing The problem is that it becomes difficult. moreover,#
Since solid-state image sensors with a numerical pixel structure spatially sample the subject image to obtain imaging output, it is not possible to properly adjust the registration of a solid-state color imaging device by simply manipulating the time axis. There is a problem.

In view of the above-mentioned conventional problems, the present invention provides highly accurate registration adjustment for overlay errors caused by one-color magnification aberration by electrical signal processing of the imaging output obtained from a solid-state image sensor. The object of the present invention is to provide a solid-state color imaging device with a novel configuration that enables high-quality color imaging.

[Means for Solving the Problems] Therefore, in order to solve the above-mentioned problems, the present invention provides a method for each color signal obtained by capturing an image using color-separated imaging light of each color using a solid-state image sensor. , storage means for storing correction data formed by interpolation processing by detecting the difference between one color signal serving as a reference and other color signals; The present invention is characterized by the configuration of a solid-state color imaging device including a signal processing means that performs registration correction processing on stored color signals.

[Function] The solid-state color imaging device according to the present invention reads out correction data formed in advance by interpolation processing from the storage means during actual imaging operation, and uses the electric signal processing means based on the correction data. Perform registration correction processing on each color signal.

[Embodiment] Hereinafter, an embodiment of the solid-state color imaging device according to the present invention will be described in detail with reference to the drawings.

The embodiment shown in the block diagram of FIG.
This is applied to an OD digital television camera, in which the imaging light from the subject passes from the imaging lens 1 to the color separation prism 2.
An imaging unit 3 for photographing a subject is composed of three COD image sensors 3R, 3G, and 3B that are guided and irradiated through the COD image sensor 3R, 3G, and 3B.

In this embodiment, the two CCD imagers 3R, 3 are transmitted from the contiguous image lens 1 through the color separation prism 2.
The above imaging light guided to G and 3B is.

The color separation prism 2 separates the three primary color components, that is, the red light R, the color light G, and the blue light B.
, G, and B are formed on each imaging surface of each CCD image sensor 3R, 3G, and 3B of the imaging section 3.

The imaging unit 3 captures each optical image of the three primary color lights R, G, and B by a reporter CCD image sensor 3R.

Photographing is performed in 3G and 3B, and three primary color signals, that is, a red signal Er, a green signal Eg, and a blue signal Eb are output as the photographic output. Then, the three primary color signals Er, Eg, and Eb obtained in the ``-image distribution section 3 are sent to the preamplifiers 4R and 4, respectively.
The signal is supplied to an analog/digital (A/D) converter 5 via G and 4E. The A/D converter 5 converts the three primary color signals Er, Eg, and Eb into three A/D converters 5R, 5G,
5B to form each color data Dr, Dg, and Db. Each color data Dr, Dg obtained by this A/D converter 5.

Db is supplied to the process processing unit 7 via the registration storage correction processing unit 6, and the three process processing circuits 7R, 7G, and 7B of the process processing unit 7 perform process processing such as gamma correction and clamp processing. After the process processing is performed, the data is converted into, for example, NTSC color image data Dout by a color encoder 8, and is output from an output terminal 9.

Each color data Dr, D supplied from the A/D converter 5
A registration correction processing section 6 performs registration correction processing on g and Db.

Based on the green signal Eg obtained by the CCD image sensor 3G for photographing a green optical image of the imaging section 3, registration correction processing is performed on the other red signal Er and blue signal Eb, and the A/D Regarding each color data Dr, Dg, and Db obtained by digitizing the three primary color signals Er, Eg, and Eb in the conversion section 5, a delay circuit 6G that delays the green data Dg by a predetermined time, a red data Dr, and It is composed of two correction circuits 6R and 6B that perform correction processing on the blue data Db.

In the registration correction processing section 6 of this embodiment, the delay circuit 6G for delaying the green data Dg is configured to operate at pixel pitches Py and Py of the CCD image sensor, which will be described later.
two delay circuits 6gy having a delay amount corresponding to g;
The configuration is as shown in Fig. 2, in which 6g x 6g x 100m are connected in series. By delaying the green data Dg, phase-advancing time axis processing is made possible. Also, red data D
As shown in FIG. 3, the two correction circuits 6R and 6B that perform correction processing on r and blue data Db include a data selector unit 60 as a pre-stage to perform neighborhood processing, and a calculation circuit that executes the neighborhood processing. It consists of a processing section 61.

The data selector section 60 of the registration correction processing section 6 includes two delay circuits DLya and D having a delay amount corresponding to the pixel pitch Py of the CCD image sensor.
Lyb, and - six 8-length circuits D Lx having a delay amount corresponding to the pixel pitch Px of the COD image sensor.
a, D Lxb, D Lxc, D Led.

It is composed of D Lie, D Lxf, a selector switch SWb with two circuits and two contacts, and a selector switch SWa with two contacts in one circuit. Each selector switch S above
Wa and SWb are controlled according to coefficients a and b given by the control unit 1°, as will be described later.

Further, the arithmetic processing unit 61 of the registration correction processing unit 6 includes 4 @ multipliers MPh, MPi, which are supplied with each data selected by the data selector unit 60.
MPj, MPk are multiplied by coefficients by the control unit 10, which will be described later,
Four latch circuits LCh, LC given i, j, k
i, LCj, LCk, and each of the above multipliers M P h, M
It is composed of an adder ADD that adds the multiplication outputs of P i , M P j , and M P k. The above four multipliers M P h, M P i, M P j , M P
k is weighted by multiplying each data from the data selector section 60 by each of the coefficients, i, j, and k supplied through the latch circuits LCh, LCi, LCj, and LCk.

Further, the A/D converter 5 converts the three primary color signals E r,
Each color data D obtained by digitizing E g and E b
r, Dg, and Db are also supplied to a control section 10 for controlling the operation of the distortion correction processing section 6 described in -h.

The control unit IO includes three image memories 11, 12, and 13 for storing each color data Dr, Dg, and Db supplied from the A/D converter 5. Above image memory 1
1, 12, and 13, the three primary color signals Er, Eg, and Eb obtained by photographing a predetermined test chart TC by the imaging unit 3 during the registration adjustment operation are digitized by the A/D converter 5 for each color. Data Dr, Dg, Db
is memorized.

The control unit 10 also controls the image memories 11, 12, .
Based on the color data Dr, Dg, and Db stored in the registration correction processing section 6, registration correction processing is performed on the other red data Dr and blue data Db using the green data Dg as a reference. A central processing unit that calculates each color registration correction data Dcr and Dcb to be applied in the circuits 6R and 6B by interpolation processing! (CPU) 14, each color registration correction data D cr calculated by this CPU 14,
Each correction data memory 15.D cb.
It is equipped with 16.

Then, during the actual photographing operation, the control section 10 reads each color registration correction data D cr and D cb from each correction data memory 15.16 and performs the registration correction via the decoder 17.18. The signal is supplied to the processing unit 6 to control the operation of the registration correction processing unit 6. The operation of each of the correction data memories 15 and 16 and each of the decoders 17 and 18 is controlled by a memory controller 19.

Here, regarding the basic principle of the registration correction process in this embodiment, the concept of the registration error of two optical images by red light R and green light G and its correction will be explained.

FIGS. 4A and 4B show, in a one-dimensional model, the concept of the overlay error of images of each color, that is, the registration error, caused by the ratio aberration of the imaging optical system and its correction process.

In this -dimensional model, as shown in FIG. ) and the image point P formed by the red light R component.
It is assumed that a registration error of Δ has occurred between R (hereinafter referred to as R image point). In addition, a solid-state image sensor such as a CCU having a discrete pixel structure performs spatial sampling of a subject image at each pixel pitch P as shown in FIG. It is assumed that the G image point Pg of ``2'' coincides with the sample point Pm on the imaging surface SC of the solid-state image sensor.

Then, if the sample value of the red light R component at the sample point Pm is R(Pn+), then the estimated value R''(Pm
−△) is R″(Pm −△) = α, R(Pm−+) + (
1-α)·R(P+++) (where α=Δ/P) It can be given by linear interpolation. Therefore,
In the registration correction process, the estimated value R''(Pm-△
) is delayed and replaced as the sample value at the sample point Pm.

In other words, the signal R' after the registration correction process
(Pm) is R'(Pm)=R''(P+n-△)=cx-R(Pm-1)+(1-a:LR(
Pm).

Furthermore, the concept of registration correction using the one-dimensional model described above can be extended to a two-dimensional model, as shown in FIG. 5, for example. In this two-dimensional model, the signal R'( m, n), R'(m, n) = a-β-R(m-4n-+), +
(1-(X)-β-R(m, nJ+α-(1-β)
・R(m-+, n) + (1-a)・(1-β)・
R (m, n) (where α=ΔX/P!, β=
Δy/Py).

Here, in the registration error caused by the debt ratio aberration of the imaging optical system, if the error amount is zero at the center of the screen, the error amount △ tends to increase as you move away from the center of the screen, and △(r, 0)=Δ(r)·exp(jθ).

Therefore, the calculation for linear interpolation is a=△x/P
x=△(r) Ncase/P xβ=△y/Py
= △(r)-sin (as J/Py, R'(m++, 1l-)1) = h, R(m, n
) + i-R(m, n+(-1)b )+ j
-R(m+(-1)a,n)+k-R(m
+(-l)a, n+(-1)b) However, h =
(1-1α1)・(1-1β1)i = (llα
1) It is given by the following formula: j = α・(1−1β1) k=lα1・1β1. In this calculation formula, coefficient a,
b is the sample point P (m, n) used for interpolation calculation, α, β
This means that the selection is made as shown in FIG. 6 according to the code.

Therefore, in this embodiment, the control unit 10 adjusts the coefficients a, b, i, j, k to the respective color registration correction data Dcr, Dcb for all sample points.
are calculated by the CPU 14 and stored in each of the correction data memories 15 and 16, respectively. Then, during the actual photographing operation, the control section 10
For each of the coefficients a, b and i, j, each color registration correction data Dcr, DCb is read out from each correction data memory 15, 16 and sent to each decoder 1.
7. The registration correction processing unit 6 via 18
The registration correction processing section 6 is operated.

In the embodiment configured as described above, at t4, each of the coefficients a, b, and i given by the control section 10.

By operating the registration J-shine correction processing section 6 according to j and k, the actual shooting During operation, high-precision registration adjustment can be performed through electrical signal processing, and high-quality color imaging can be performed.

[Effects of the Invention] In the solid-state color imaging device according to the present invention, electrical signal processing is performed based on the correction data while reading out the correction data formed in advance by interpolation processing from the storage means during actual imaging operation. Since the means performs registration correction processing on each color signal, it is possible to perform highly accurate registration adjustment by electrical signal processing of the imaging output obtained by the solid-state image sensor, and high-quality color imaging can be performed. and can fully achieve the intended purpose.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the solid-state color imaging device according to the present invention, FIG. 2 is a block diagram showing the configuration of the delay circuit of the registration correction processing section in the above embodiment, and FIG. The figure is also a block diagram showing the configuration of the correction circuit of the registration correction processing section. FIGS. 4A and 4B are schematic diagrams showing, in a one-dimensional model, the concept of overlay errors of images of each color caused by the ratio aberration of the imaging optical system and their correction processing. FIG. 5 is a schematic diagram when the concept of the registration error correction process described above is extended to a two-dimensional model. FIG. 5 is a schematic diagram showing the principle of operation of the registration error correction process in the above embodiment using the above two-dimensional model. l...Imaging lens 3...Imaging section 3R, 3G, 3B...CCD image sensor 6...
Registration correction processing section 6R, 6BΦ... Correction circuit 6G... Delay circuit 7... Fist process processing section 7R, 7G, 7B... Process processing circuit 8'' Color encoder 9... Output terminal 6011・Data selector section 61... Arithmetic processing section patent Applicant: Sony Corporation

Claims (1)

[Claims] For each color signal obtained by capturing an image using color-separated imaging light of each color with a solid-state image sensor, the difference between one color signal serving as a reference and other color signals is detected and formed by interpolation processing. storage means for storing correction data to be corrected;
A solid-state color imaging device comprising: signal processing means for performing registration correction processing on the other color signal based on the correction data read from the storage means.