JPH02284591A - Imaging device - Google Patents

Abstract

PURPOSE:To correct the registration deviation generated between images formed on plural image pickup elements by storing the digital image signal from a line memory signal converting means arranged for each image pickup element and performing calculation and control in accordance with correction data to read out the signal. CONSTITUTION:The light which passes a lens 6 is separated to three colors R, G, and B by a three-color separation prism 7, and respective light components are converted to electric signals by CCDs 8, 9, and 10. Electric signals read out from CCDs are converted to digital signals by A/D converting circuits 11, 12, and 13 and are temporarily written in line memories 14, 15, and 16, and they are read out while they are corrected by a registration correcting means if correction of the registration deviation is necessary. Signals which are subjected to correction of the registration deviation and are read out are converted to analog signals by D/A converting circuits 17, 18, and 19 and are handled in the same manner as signals of an ordinary color television camera thereafter.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an imaging device, and particularly to an imaging device having a registration correction means.

[Conventional technology]

First, there is a registration shift (R) that conventionally occurs in an imaging device having a plurality of imaging elements.

G, B33 original overlay deviation) will be explained.

When the lens of an imaging device has chromatic aberration of magnification, a phenomenon occurs in which the imaging magnification differs depending on the wavelength of light. Therefore,
In multi-tube and multi-plate color television cameras, etc. that split light into wavelengths and form images on two or more multiple image sensors, registration deviations occur due to differences in image size between the image sensors. arise. This phenomenon will be explained using an example where light is separated into three primary colors of R (red), G (green), and B (blue) and imaged. FIG. 5 is a side view showing the chromatic aberration of magnification in a conventional imaging device, and FIG. 6 is a plan view showing the difference in screen size for each color due to the chromatic aberration of magnification in FIG. First, in FIG. 5, 1 represents a lens and 2 represents an imaging surface. The incident light passing through the lens 1 forms an image on the imaging surface 2, but the image height differs depending on the wavelength as described above. Therefore, the effective image planes for each color of R, G, and B are the same size as the 3R, 4G, and 5B screens, as shown in Figure 6, and even if these images are superimposed again, the original image will not be accurately reproduced. I can't do anything. Also, in the case of an image pickup tube, the scanning speed of the scanning electron beam when reading out the signal charge on the imaging surface.

By adjusting the scanning interval and scanning size, it is possible to output an image as an apparently enlarged or reduced signal. Therefore, in a multi-tube color television camera, each image pickup tube enlarges or reduces the image according to the lateral chromatic aberration of the lens, and the image size in all image pickup tubes is matched to correct the registration deviation. There is.

On the other hand, solid-state image sensors form images by regularly reading out signals from pixels arranged regularly on the light-receiving surface, so 1) "It is not possible to correct the registration gap in the same building as the image pickup tube. For this reason, the current situation is that in multi-disc color television cameras, correction of misregistration is not performed.

(Problem to be Solved by the Invention) As described above, in the conventional example, when trying to correct the registration deviation, in the multi-tube color television camera, the registration deviation is corrected by the method of the conventional example. However, there is a problem in that when the amount of correction becomes large, image quality deteriorates due to occurrence of shading (signal non-uniformity).

On the other hand, as mentioned above, in the multi-plate type power radar camera,
Since there is no means to correct registration errors, there is a problem in that the image becomes unsightly with noticeable color shifts.

This invention was made in order to solve the above-mentioned problems of the conventional method. When this invention is applied to an imaging device such as a multi-tube color television, image quality deterioration such as the generation of shape ink can be suppressed. However, it is possible to correct the registration deviation caused by the lateral chromatic aberration of the lens, and when this invention is applied to an imaging device such as a multi-panel color television, it is possible to correct the registration deviation, which was previously impossible. The purpose is to make it possible.

[Means to solve the problem]

Therefore, in the present invention, a plurality of image pickup devices that convert light of multiple colors into electrical signals, a signal conversion means that converts the electric signals into digital signals, and a plurality of image pickup devices arranged for each of the image pickup devices, The image signals are formed on the plurality of image sensors by calculating and controlling and reading out the image signals stored in the line memory according to the correction data. The objective is to be achieved by an imaging apparatus configured with a registration correction means for correcting registration deviations occurring between images.

(Operation) The imaging device according to the present invention uses a plurality of imaging elements to
Converts multiple colors of light into electrical signals, and uses signal conversion means to
The electric signal is converted into a digital signal, the image signal from the signal conversion means is stored in a line memory provided for each image sensor, and the image signal stored in the line memory is converted into a digital signal by a registration correction means. The correction data is calculated and controlled and read out in accordance with the correction data to correct the registration deviation that occurs between the images formed on the plurality of image sensors.

〔Example〕

Two embodiments of this invention will be described below based on the drawings.

FIG. 1 is a block diagram showing the configuration of a first embodiment of the present invention, FIG. 2 is a plan view illustrating registration deviation in the first embodiment, and FIG. 3 is a vertical COD pixel of the first embodiment. FIG. 4, a partially enlarged schematic diagram of lines, is a configuration diagram showing the configuration of a second embodiment of the present invention.

First, a first embodiment of the imaging apparatus of the present invention will be described with reference to FIGS. 1 to 3.

This first embodiment uses three CCDs as image sensors.
2 illustrates a CCD color television camera. In FIG. 1, 6 is a lens, 7 is a three-color separation prism that separates the light entering from the lens 6 into three colors, R, G, and B;
9.10 is a CCD that converts each color of R, G, and B light into an electrical signal, and constitutes an image pickup device (a).

Reference numerals 11, 12, and 13 designate A/D conversion circuits that convert analog signals read from each of the CCDs 8, 9, and 10 into digital signals, and constitute signal conversion means (b). Reference numerals 14, 15, and 16 indicate line memories arranged for each of the plurality of image sensors, into which image signals from the signal converting means (b) are written and which store the digital signals, and 17, 18, and 19 indicate line memories, respectively. 14, 15, and 16, and 20 is a microprocessor that converts the signals read from the CCD into analog signals. A microprocessor that calculates and controls readout timing, signal writing to line memory, and readout timing, and measures and controls image signals stored in the line memory according to the correction data to read out the image signals. It constitutes a registration correction means (d) for correcting a registration shift that occurs between images formed by the image pickup device (details will be described later).

Next, the operation of this first embodiment will be explained using FIG.

In FIG. 1, light passing through a lens 6 is separated into 83 colors, R and G, by a three-color separation prism 7, and each light is converted into an electric signal by a CCD 8, 9, and 10, respectively. Electrical signals read from each CCD are converted into digital signals by A/D conversion circuits 11, 12 and 13, respectively, and then written to line memories 14, 15 and 16, respectively, to correct registration distortion. If necessary, the data is read out while being corrected by a registration correction means (d) (described later). The correction data (described later) entered into the microprocessor 2o may be input by manual adjustment, or the lateral chromatic aberration of the lens may be measured in advance and that information written to the ROM (described later). It may be read out. Based on the correction data, the microprocessor 20 controls the timing of reading signals from the CCDs 8 and 9.10 for each channel of R, G, and B, writes signals to the line memory, and controls read signals from the line memory. conduct. The signals read out after correcting the registration deviation (described later) are converted into analog signals by the D/A conversion circuits 17, 18, and 19, and then are similar to signals from a normal color television camera. be treated as such.

Although this embodiment has been described for a three-panel type CCD color television camera, the above-mentioned registration deviation correction method can be applied to a general multi-tube type or multi-plate type color television camera having two or more image pickup tubes or solid-state image sensors as image sensors. Needless to say, it is applicable to

Next, the operation of the registration correcting means (d) of this first embodiment will be explained using FIGS. 1 to 3.

In FIG. 2, reference numerals 21 and 22 indicate the image planes of the CCDs for R channel and G channel imaging, respectively. here,
In the R channel, the image reflected on the entire image plane 21 (second
(Figure (left)) is reduced by 1/α (α>1) times in the G channel and is reflected in the range of the broken line 23. Further, the number of pixels of the CCD is assumed to be xxY pixels (X pixels horizontally and Y pixels vertically). As shown in Figure 2 (
Correcting the image (right) by the registration correction means (d) will be described next.

First, (X/α)×(Y/α) on the G channel imaging plane
Consider correcting the misregistration by enlarging the image 23 in a pixel by α times both in the vertical and horizontal directions. In a CCD, signals are read out line by line in the X direction, but in order to expand the Y/α line to the Y line (1-1
/α) Y line is insufficient. By A/D converting the signal from the CCD and storing it in the line memory (Fig. 1 (c)), the signal of that line can be read out once for each line, in total (1-1/α) If the data is read out in accordance with the data of repeatedly reading out the same line signal Y times, an image multiplied by 6 in the Y direction will be obtained and formed by multiple (in this case two) image sensors. Registration deviations that occur between images are corrected.

Next, in the above correction, a specific example of extracting signals for 500 lines from an image of 450 lines will be explained with reference to FIG. In FIG. 3, the pixels of the CCD section are drawn line by line, and the signal of one line is repeatedly read out twice every 9 lines, that is, the shaded lines in the figure. Therefore, if a total of 50 lines of signals are read out twice, 500 lines of signals can be obtained in the end. At this time, if you use Raishin Sled for 2 lines, instead of reading the same line twice, you can read it twice.
It is also possible to perform vertical expansion by interpolation from line signals.

The above is an operation for enlarging in the vertical direction, but enlarging in the horizontal direction is basically performed in the same way as in the vertical direction.

That is, among the signals stored in the line memory, X/α
The signals for X pixels are routed using the pixel signals, and the total (1
-1/α) X times, the signal of the same pixel may be read out by turning the edge twice. Also, at that time, instead of reading the same pixel twice, the signal may be extracted by interpolation from the signals of two adjacent pixels.

Next, a second embodiment of the present invention will be explained using FIG. 4. In this second embodiment, the present invention is applied to a color television camera with a zoom lens. Figure 4 shows only the lens part that has the ability to output correction data.
It is assumed that the correction data output from the lens side is input to the microprocessor 20 in FIG. In addition, this color TV camera with zoom monitors the usage status of the lens (focus position, zoom ratio) depending on zooming, etc.

When the aperture value) changes, the chromatic aberration of magnification also changes accordingly. This second embodiment attempts to follow the variation of the chromatic aberration of magnification in real time and correct the registration deviation caused by the chromatic aberration of magnification.

In FIG. 4 of the drawing, 24 is a zoom lens body, 25,
Reference numerals 26 and 27 indicate potentiometers built into the zoom lens 24, respectively. Potentiometer 25 is provided to read data on the focus position, potentiometer 26 is used to read data on the zoom ratio, and potentiometer 27 is provided to read data on the aperture value. 28, 29.30 respectively.”
An A/D conversion circuit 31 converts data from the two potentiometers 25, 26, and 27 into digital signals;
ROM in which information on magnification chromatic aberration (zoom ratio, aperture value) is stored; 32 is a potentiometer 25, 26.2;
Based on the data from each of 7 and the ROM information,
A microprocessor that calculates and outputs appropriate correction data. Next, the operation of this second embodiment will be explained using FIG. 4. In Figure 4 of the drawing, potentiometer 25
, 26, 27, respectively, are input to the microprocessor 32 after passing through the A/D conversion circuits 28, 29, 30, respectively. The microprocessor 32 reads out the chromatic aberration of magnification information from the ROM 31 according to the manually entered data, performs interpolation calculations if necessary, and outputs appropriate correction data for each usage state of the lens. The correction data is input manually to the microprocessor shown in FIG. 20, and the registration correction means (d) corrects the registration deviation in the same manner as in the first embodiment. When the state of use of the lens changes, correction data appropriate for the new state of use is immediately calculated, and registration deviations are corrected based on that data, making it possible to always obtain beautiful images.

〔Effect of the invention〕

As explained above, according to the present invention, by applying the present invention to a multi-tube color television camera, etc.,
It is possible to suppress image quality deterioration due to shading, etc., and correct registration shifts caused by lateral chromatic aberration of the lens. Furthermore, if this invention is applied to multi-disc color television cameras, etc. This has the effect of making it possible to correct misregistration.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a first embodiment of the present invention, FIG. 2 is a plan view illustrating registration deviation in the first embodiment, and FIG. 3 is a vertical CCD pixel of the first embodiment. FIG. 4 is a partially enlarged schematic diagram of lines, and FIG. 4 is a configuration diagram showing the configuration of a second embodiment of the present invention. FIG. 5 is a magnification of a conventional image pickup apparatus. (a) Image sensor (b) Signal conversion means (c) Line memory (d) Registration correction means 8. 9
．． 10-...CCD 11.12.13...-A/D conversion circuit 14.
15.16-・・・Line memory 17.18.19
...-D/A conversion circuit 20.32-...Microprocessor 25.26.27...Potentiometer 31...-ROM

Claims (1)

[Scope of Claims] An imaging device characterized by comprising the following components (a) to (d). (a) Multiple image sensors that convert multiple colors of light into electrical signals. (b) Signal conversion means for converting the electrical signal into a digital signal. (c) A line memory that is arranged for each of the image pickup devices and stores image signals from the signal conversion means. (d) By calculating and controlling and reading out the image signal stored in the line memory according to the correction data,
Registration correction means for correcting registration deviations occurring between images formed on the plurality of image sensors.