CA1252561A - Registration adjusting system for color television camera - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
A registration adjusting system for color television camera detects registration error data with respect to detection marks on a test chart on the basis of image signals concerning the detection marks sent from a plural number of image pickup tubes, obtains data indicative of the registration errors in the zones other than these marks by the interpolation process based on these registration error data, and supplies both the registration error data and the data obtained by the interpolation to the deflection system of the pickup tubes as control data to compensate for the registration error, whereby the registration of the color television camera system is adjusted.

Description

~,2~i,2~G~

The present invention relates -to a systern for ad-justing the registra-tion of an image pickup equipment using a plurality of image pickup tubes, such as a color television camera.

The present invention will be illustrated by way of the accompanying drawings, in which:-Fig. 1 shows an example of a prior art test chart for automatic registration adjustmenti Fig. 2 shows an example of a test chart for auto-matic registration adjustment which is applied to the pre-sent inven-tion;

Figs.3 and 4 are diagrams for use in explaining the principle of the presen-t invention;

Fig. 5 is a block diagram of an embodiment of a registration adjusting system for a color TV camera apparatus according to the present invention; and Figs. 6 to 8 are diagrams for explaining the opera-tion of the system of Fig. 5.

There has already been known automatic registration adjusting systems whose operation is based on a computer.
In such registration adjusting systems, the registration adjustment is carried ou-t among image pickup tubes of a color TV camera under program control by a micro computer. These automatic adjustment systems have conventionally been classi-fied into several types, but these -types are the same in principle. The principle of -the prior art systems are as follows: When detec-tion marks 11 and 12 on a tes-t chart as shown in Fi~. 1 are scanned in the horizontal direction, video signals indicative of the marks 11 and 12 are generated .~25256~

at image pickup tubes. rrhe video signals are then supplied into a band-pass filter, high-pass filter, etc. to find a phase difference between the video siynals of the respective image pickup tubes, whereby the registration errors in both the horizontal and ver-tical directions are detected. To minimize these registra-tion errors, a computer controls the deflecting circuits for each image pickup tube-In the conventional automa-tic adjustment system, ~0 . ..

- la -5~

the items to be adjusted i.nclude -the centerincJ/ si.ze ].inearity skew dis-tor-tion and ro-tation o~ ~he ~,:i ek~ s image. These items have been selected, ~irstly ~)c~causc these have been used in manual adjustment, and secondly because these can be carried out by controlling -the D.C.
bias, amplitude and linearity with respect to a saw tooth current to be supplied to each of the vertical and horizontal deflec-ting coils of -the iamge pic]c-up tubes, and therefore the au-tomatic adjustmen-t can easily be realized by using the circuit technique in the manual adi ustment .
For these reasons, the same items as in the manual adjustment have long been selected up to -the present, though we are now in the computer age.
However, it has been know that a sa-tisfac-tory registration adjustment cannot be achieved over the whole screen area by adjusting the above-mentioned items only.
For example, the test chart of Fig. 1 has three de-tec-tion marks in the horizontal center of the screen, and therefore the most favourable registration can be obtained only at three points. In other words, the respective output video signals of the pickup tubes can be adjusted to be in phase with one another only at the three points.
On scanning lines other than in the center of the screen such as at the upper or lower screen areas there exists only one detection point or may not exist at all. For such screen areas, the registra-tion adjustment ~ ~ 2~ ~

can be effected through th0 skew compensation. Since the regi-stration error is usually greater at the peripheral area of the screen, the desirable registration adjustment cannot be achieved throu~h the skew compensation in such area. In addition, the above conventional ad~ustment method has anokher problem that such problems in horizontal registration adjustment similarly arise in the vertical registration ad~ustment.

At present, there has been proposed and commercialiæed a system wherein hori~ontal and vertical registration errors are detected at many points all over the screen. Fig. 2 shows a test chart used for such system. When a test chart comprising a total of 25 detection marks 21 arranged in 5 rows and 5 columns is used, the horizontal and vertical registration errors can be det-ected at 25 points where the marks 21 are located, and as thedetection marks 21 are increased in number, the registration err-ors can be detected at the increased number on the screen.

Conventionally, however, the detected registration err-ors have been used only for the evaluation of a color TV camerawith respect to the registration but have not been used for the ad~ustment of the registration error.

The present invention thus provides a registration adjusting system for a color TV camera which allows an accurate ad;ustment of horizontal 6~

and vertical registrations all over the screen.

According to -the present inven-tion there is provided a registration adjusting system for colour television camera appa-ratus in which regis-tration error is adjus-ted based on video sig-nals supplied from a plurality of image pickup tubes picking up a multiplicity of detection marks placed horizontally and verti-cally on a test chart, comprising registration error de-tecting means for detecting first data representing registra-tion errors corresponding to each of said detection marks, interpolating means for obtaining by interpolation second data representing r~gistration errors in a quadrilateral area determined by each of four mutually adjacent detection marks other than said detecting marks based on said first data, said interpolating means perform-ing interpolation with respect to one pair of the two pairs ofopposing sides of said quadrilateral area so as to obtain the second data for a plurality of points on said one pair of the opposing sides, and then performing interpolation with respect to each line connecting corresponding points of said plurality of ~0 points on said one pair of the opposing sides so as to obtain the second data for a plurality of points on said each line, and data supplying means for supplying said first and second data into deflecting systems of said image pickup tubes for adjusting said registration errors.
In accordance with the present invention, the registra-tion of the entire screen can be adjusted more accurately than the conventional registration obtained through the centering, size, linearity, skew and rotation adjustments.

~2525~

Referring once rnore -to the accompanyiny drawings, first of all, explanation will be made as to the principle of -the present invention. ~rypical colour TV camera systems using a plurality of image pickup -tubes include a three tube system wherein an image is converted to three primary colours, i.e., red (R), green (G) and blue (B), for which different three image pickup tubes are used. In this system, the red and blue registrations are adjusted according to a video signal from the green pickup tube having a spectral charac-teristic most similar -to human eyes. The green video signal (G-video signal) is compared with a reference signal such as a so-called grating or cross hatch signal to thereby pu-t the deflection distor-tion of the green video signal in its most favourable state.
In this invention, the test chart shown in Fig. 2 is used so that registration errors of Red and Blue video signals (R- and B-video signals) in the both horizontal and vertical directions with respect to the G-video signal are detected for respective 25 detection marks 21 of the ~S

,/ ---- .
/

char-t. In this connec-tion, the means for de-tectincJ the registration errors is well known ancl the explana-tion thereabout will be omitted.
Registration errors of the R- and B-video signals with respect -to the G-video siynal are the same in principle and thus the registration error of the R-video signal alone will be explained.
Fig. 3 shwos registra-tion error detec-tion points 31 corresponding to the detec-tion marks 21 in Fig. 2; and Fig. 4 shows the registration errors of the R-video signal with respect to the G-video signal detected at the respective detection points 31, with the lag being represented as positive in the upward direction of the Z
axis and lead as negative in the downward direction of the Z axis.
In Fig. ~, points, at which solid lines in the vertical direction V are intersected with solid lines in the horizontal direction H, correspond respectively to the detection points 31. Sixteen ~uadrila-terals are formed by broken lines connecting points indicative of registration error data at the detection points in the vertical position and chain-dotted lines connecting said data points in the horizontal position. A surface having the 16 quadrilaterals indicates a registration error in the horizontal direction of the rectangular screen c1efined by lines connecting points Alll ~15~ ~51an 55 g As the detection marks 21 of the test chart shown s~

in Fig. 2 are increased in number with respect to both the horizontal and vertical directions, -the number of such detection points 21 as shown in Fig. 3 is increased accordingly, whereby the registration ~rrors can be de-tected with a higher accuracy. If a registra-tion c~rr(Jr is CLIIIS~I
in the opposite direction the horizontal registr;ltion error shown in Fig. 4 will be compensated.
For the above compensation, in -the case of an image pickup ~ube oE electromagnetic deflection -type, a magnetic field to compensate for -the regis-tration error is applied to tube's deflecting magnetic field; while in the case of elec-trostatic deflection -type, an electric field to compensate for the registration error is applied to tube's deflecting electric field.
An embodiment of the invention based on the above principle will be explained with reference to Fig. 5, in which a test chart 51 has the same,arrangement as in Fig.

2. The test chart 51 is focused on respec-tive targe-ts of image pickup tubes 54, 55 and 56 through a lens system 52 and an optical system 53 for colour conversion. The tubes 54, 55 and 56 are provided with deflec-ting coils 541' 5`51 and 56l respectively. Video signals outputted from the tubes 54, 55 and 56 are supplied -to respective R-, G-and B-video circuits 57, 58 and 59 where these video signals are subjected to such predetermined operations as amplifications, etc. The video signals processed at the video circuits 57 to 59 are then supplied to a colour ~2~25~

signal mixer (not shown) and also to a recJistrat:ion-error detecting circui-t ~0 which comprises a kno~n circui-t for detecting the registration error and known ~/~ converter.
The error detector 60 de-tects the horizon-tal and vertical registration errors in the R- and B-video signals wi-th respec-t to the G-video signal for each of the detection marks 31 shown in Fig. 3 under a cornmand from an ari-thmetic circuit 60 to be explained later, and generates a digital signal. The data indicative of the detected reyistration error is applied through a write circuit 61 to a memory 62. The ari-thmetic circuit 62, when receiving via its input terminal 64 an automatic adjustment command signal, provides an activation command to the error detector 60.
For simplification of the explanation, the explanation will be directed only to the horizontal registration error of the R-video signal with respect to the G-video signal.
As has been explained earlier, the registration error data of the R-video signal detected at the error detector 60 is as depicted in Fig. 4. When such data is written in the memor~ 62, the followiny technique is employed. That is, the storage area of the memor~ 62 comprises a matrix of rows representative of respecti~e horizontal scanning lines and columns represen-tative of respective divided segments of the horizontal scanning lines divided by any integer. Codes indicative of the positions of the detection poin-ts 31 are applied to the respective data indicative of registration errors at -the ~2~;6~

respective detec-tion points 31. For example, the code concerning the horizontal reyistration error data of the R-video signal with respect ko the ~,-video siynal al- a detection point 31 of H-direction position number 1 and V-direction position number 1 is made up of (lll, Vl, R-~"
H REGI). The write circuit 61 is opera-ted in such a manner as to judge the code, selec-t a corresponding one of the storage area of the memory 62 where the horizontal registration error da-ta of -the R-video signal with respect to the G-video signal is to be stored, and store the data at the selected address (hl, vl) in the storaye area.
In this way, 2S data indicative of registration errors at the 25 detection points 31 of Fig. 3 are s-tored in the memory 62 as shown in Fiy. 6. In Fig. 6, the address (hl, vl) is expressed by All.
When all the reyis-tration errors at the 25 detection points have been detected, a completion signal is sent from the registration-error detecting circuit 60 to the arithmetic circui-t 63. The arithmetic circuit 63, for example, may comprise a microcomputer having operation, decision and storage functions. The arithmetic circuit 63, when receiviny the completion siynal, wai-ts until the final detec-tion data is written in the memory 62 by the write circuit 61. After the waitiny, the ari-thmetic circuit 63 receives all or some of the 25 s-tored registration error data from the memory 62, and executes ~.~S2S~

interpolation on regis-tration errors at the zones other than the de-tection points 31 in-the following manner. Tha-t is, assuming in Fig. ~ that the registration data on -the uppermost, second,... and lowermost rows are -termed from their left sides (all, al2---~ als), (a21~ a22' ' a25) ( 51~ a52~ ~ a55), and addresses thereof are termed from their left sides (All~ A12~ -~ Als)~ ( 21' 22' ' 25)' ' and (A51~ As2~ A55) respectively, registration error data in such a zone determined by -the 11' A12, A21 and A22 as shown in Fig. 7 can be interpolated on the basis of the four data all, al2, a21 and a22 indicative of respective registration errors at the addreSses All~ A12' A21 and 22 if the number of scanning lines between the addresses All and A21 is N, then all the registration data at the addresses between the addresses All and A21 can be interpolated by a line connecting between the data all and a21 points. Similarly, the registration error data at the addresses between the addresses A12 and A22 can be all found by the error data al2 and a22. Further, assuming that there are M addresses fromthe address All to the address A12 in the horizontal direction, all the registration e~ror data at the M addresses can be inter-polated on the basis of the error data all and al2.
Similarly, all the registration error data a-t ~1 addresses between the addresses A21 and A22 can be interpolated on the basis o~ the error data a21 and a22.

In this way, when the registra-tion error datc at respective addresses between (All and A2l), (A12 ~nd A ), (A and A12), and (A21 and A22 registration error data a-t addresses in a zone determ:ined by the addresses All, A12, A21 a 22 polated on -the basis of the found error da-ta. Since registration error data at addresses in o-ther zone determined by four other addresses can be similarly interpolated, the registration error data a-t all addresses n an area determined by the addresses A , A , A and ll 15 51 A 5 in Fig. 4 can finally be interpolated.
On the other hand, registration error data at addresses in an area other than the area determined by Eound through the interpolation operation, because the erro~ data are expressed in the form of an extension of a line connecting a point indicative of registration error data at one of the detection points 31 and a point indicative of registration error data found by the above interpolation operation.
The arithmetic circuit 63 executes the above interpolation operation to find registration error data at addresses of zones other than the zones at the de-tection points 31 shown in Fig. 3, and stores the found data at predetermined addresses in the memory 62, i.e., at addresses other than the addresses A to A , A to A ,..., and A to A shown in Fig. 6. In this way, ~5'~

stored in the memory 62 are the regis-tration error data at the addresses A1l to A15, 21 25 to A55 indicative of the detection poin-ts 31 as well as the interpolated registration error data. When all khe data have been stored at the respective addresses in the memory 62, a read command signal is supplied from -the arithmetic circuit 63 to a rea~ control circuit 65, which in turn generates a read control signal in the basis of vertical and horizontal sychroniziny signals supplied from an input terminal 66 thereof and sends it to a read cixcuit 67. The circuit 67 reads out data frorn-the memory 62 in synchronism with the electron-beam scanning operation of the pickup tubes, and applies the read daka to a digital/analog (D/A) converter 68. A signal provided from the D/A converter 68, which indicates a hor.izon-tal regis-tration error of the red video signal at an electron-beam scanning point, is supplied to a red (R) deflecting circuit 70 through an attenuator 69 which comprises, for example, a digital attenuatox capable of performing a four-quadrant multiplication. The red (R) deflecting circuit 70, green (G) deflecting circuik 71 and blue (B) deflecting circuit 72 genexake horizon-tal and vertical deflecking siynals in response to a synchroniziny signal supplied from a common input terminal 73 thereof, respectively, and send these deflecting signals to the respective deflec-ting coils 541 to 56l of the pickup tubes 54 to 56.
More specifically, the R deflecting circuit 70, ~o~

when receiving an outpuk signal from the attenuator 69, generates the deflecting signal for compensating for the horizontal red registration error. Such a deflecting signal can be generated by converting the output signal of the D/A converter 68 to a current signal by means of a voltage-to-current converting circuit (not shown) and then by superimposing the current signal on, for example, a centering current. In this connection, it is necessary ~hat the frequency characteristic of a circuitry inc-luding the horizontal deflecting coll extends sufficiently to the high frequency band. The circuitry may be as disclosed in appli-cant's Japanese Patent Application No. 56-133364 (1981) published March 2, 1983 under laid open No. 58-35 ~47 (19~3).

The above-mentioned operation is shown by steps 81 to 86 of a flow chart given in Fig. 8. when the horizontal red centering current is compensated for on the basis of the output signal of the D/A converter 68 as has been mentioned above, a magnetic field generates in the deflecting coil 541 in such a direction as to compensate for the horizontal red registration error, whereby said registration error can be compensated for.

However, when said registration error has been compen-sated for, there might sometimes be a possibility that the com-pensated amount is large or small to an excessive extent, because it is not clear whether or not the compensated amount is proper.
Accordingly, in this embodiment, the registration error data after said 25~

compensa-tion is again dekec-ted by the registra-tion-error de-tecting circuit 60 on the basis of the video s:ignals indicative of -the detec-tion marks 21 on the test chart 51 ( step 87 in Fig. 8). Now, assuming that -the de-tec-ted error data are a'll to a'l5,..., and a 51 55 values of the data are given b,y the following equations.
a'ij = aij/c ..... ~(l) or a'ij = -c-aij .... (2) where, c is a positive number, i is 1, 2, 3, 4 or 5 and j is 1, 2, 3, 4, or 5. Equation (l) explains -tha~ l~lle amount of the regis-tra-tion error after compensation is smaller than that before the compensation; while Equation (2) explains that the error direction is reversed due to over-compensation, for example, explains a state where the error in the lead direction is changed to that in the lag direction.
The arithmetic circuit 63 decides whether or not the registration error data after compensation detected by the error detecting circuit 60 is smaller than a permissible value ( step 88). When the compensaked amount is too small, the arithmetic circuit 63 genera-tes a signal and sends it to the a-ttenuator 69 to -thereby decrease the attenuating amount therein; whereas, if the compensated amount is too large~ the circuit 63 generates a signal and sends the signal to the attenuator 69 to thereby increase the attenuating amount ( s-tep ~9). ~s Z5~

a result, said signal generated by the circuit 63 is multiplied a-t the a-ttenuator 69 by the output siynal o~
the D/A converter 68, whereby a deflectiny ma~Jnetic field added to said multiplied signal is produced in the R-deflecting coil 541. The above procedure, i.e., -the steps 87 to 89 in Fig. 8 are repeated until the regiStratiOn error data falls within the permissible value. After the error data reaches -the permissible value, -the registration adjustment is completed.
Though the above explanation has been made as to the compensation of the horizontal registration error of the red video signal with respect to the green video signal, the same explanation can be applied to the compensation of the vertical red registration error, and further to the compensation of the horizontal and vertical registration errors of the blue video signal with respect to the green video signal. In addition, the deflection distortion in the green video signal may be compensated for with respect to an electric reference signal in the similar manner to the above.
According to the above embodiment, the horizontal and vertical registration errors on the whole screen are compensated for on the basis o~ the registration error d~ta at the detection points as well as the reyistration error data at positions o-ther than said detectioll poinls.
Therefore, there will not occur in the present invention a correlation among the centering, size and lineari-ty '~5~

as in the prior art. Slnce such a problem tha-t a variation in the linearity does not cause the var:iation in the centering and size, the registration error can be compensated for highly easily and accura-tely in the present invention.
In the prior art system, a total of eight adjustment items of the skew and rotation adjustments in addition to the centering, size and linearity adjustments both in the horizontal and vertical directions are se-t, and the eight adjustments are made at three poin-ts on a horizontal center line and three poin-ts on a ver-tical center line on the screen using eight attenuators or multipliers. In the present invention, on the other hand, such many control items as the centering, size, linearity, skew and rotation are unnecessary and the horizontal or vertical registration can be compensated for the whole screen by using a single attenuator. That is, the invention has an advantag~ that, the adjustment accuracy can be improved and necessary attenuators can be reduced in number, whereby the circuit configuration can be simplified and the stability can be improved.
The present invention is not limited to the above embodlment. More specifically, the test chart is not rectric-ted only to the test chart of Fig. 2, and any test chart having other arrangement can be employed as long as the chart enables the measurement a-t many points on the screen.

.

Fur-ther, in a standard -television system, interlaced scanning is carried out and thus one ~rarne consists of first and seconcl fields. For -this standard system, the reglstration shif-t is detec-ted at either one of the first and second fields or is detected at both oE the firs-t and second fields at the same -time and then the average of the de-tected results is calculatcd.
Further, the at-tenuation amount by ~lle attenuator 69 has been automatically adjusted in the above embodiment, but this adjustment may be made manually.
That is, after the steps 81 to 86 in Fiy. 8 have been carried out, the registra-tion chart is picked up to display the picked images on a monitor device and the attenuation amount by the attènua-tor 69 is manually controlled while observing the images on the monitor to compensate for horizontal regis-tration error of the red video ~ignal with respect to green video signal. In accordance with the present inven-tion, even in the case of the manual adjustment, eight adjustors for the eight adjustment i-tems which have been necessary in ~he prior art system can be reduced in number to two, and the horizontal and vertical registration errors can be compensated for independently of each other, whereby the simple adjustment can be advantageously achieved.

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A registration adjusting system for colour televi-sion camera apparatus in which registration error is adjusted based on video signals supplied from a plurality of image pickup tubes picking up a multiplicity of detection marks placed hori-zontally and vertically on a test chart, comprising registration error detecting means for detecting first data representing reg-istration errors corresponding to each of said detection marks, interpolating means for obtaining by interpolation second data representing registration errors in a quadrilateral area deter-mined by each of four mutually adjacent detection marks other than said detecting marks based on said first data, said interpo-lating means performing interpolation with respect to one pair of the two pairs of opposing sides of said quadrilateral area so as to obtain the second data for a plurality of points on said one pair of the opposing sides, and then performing interpolation with respect to each line connecting corresponding points of said plurality of points on said one pair of the opposing sides so as to obtain the second data for a plurality of points on said each line, and data supplying means for supplying said first and sec-ond data into deflecting systems of said image pickup tubes for adjusting said registration errors.