JPS5838075A - Image pickup device - Google Patents

Abstract

PURPOSE:To perform the good adjustment of an aperture using simple constitution by sampling a video signal and adjusting the aperture in accordance with the ratio of counting value in which the sampled signal level becomes above or below a specified level. CONSTITUTION:A video light from an object 1 is supplied to an image pickup section 4 through an aperture 2 and a lens 3, and at the same time, a signal from a synchronizing signal generatir 4' is supplied to the image pickup section 4 and a video signal is taken out. A digital signal from an AD conversion circuit 6 is supplied to a comparator circuit 11 and compared with a digital signal from a reference signal source 12. When the signal from the circuit 11 is larger, a signal of 1 is outputted. The output is counted by a counter 14 and latched in a latch circuit 15 at timing of vertical synchronizing signal. The signal of the circuit 15 is supplied to an adjustment driving circuit 19 through a DA conversion circuit and a drive amplifier 17 to adjust the aperture 2. By this way, adjustment nearly equivalent to the mean value system can be effected, and as an accumulator is not necessary, the circuit constitution can be simplified.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to automatic aperture adjustment in an imaging device such as a television camera, and is suitable for using COD as a 49kIl function and converting video signals to ADD and processing them into digital signals. The purpose is to provide a device that provides

In the automatic adjustment of the aperture in a conventional television camera, control signals for adjustment are obtained using three methods, generally IIc and below.

1) Peak value method b) Average value method c) Combined use of a) and b) Among these, the peak value method (Otori) controls the peak value of the video signal output from the imaging unit to be constant. This method is effective when the imaging section uses an element that is prone to phosphorus, adhesion, or blue-young phenomenon. However, the content of the reflection has changed.

On the other hand, the aperture reacts sensitively, and the brightness of the green leaves changes rapidly, making it difficult to see.

On the other hand, the average value method (b) has a smooth aperture response and is good, but there is a risk of burn-in and blue-young.

Therefore, in the past, the two control signals of a) and b) were appropriately mixed and used as shown in C), and in this case, when COD was used as the sensor, k was used.

Since there is no risk of burn-in or blooming, it was considered better to use a larger proportion of b) in C). - However, using a COD as an imaging unit is problematic.

When the digital signal is processed by AD converting the Eirin signal, in the IC1b) method, the scale of the circuit for obtaining the average value, especially the klE calculator, becomes large, and the entire circuit becomes large. There were some drawbacks.

In view of these points, the present invention proposes a new control method which can obtain control signals substantially equivalent to the above-mentioned average value method and which has a simple one-circuit configuration.

By the way, as one of the iii* processing techniques, it is called the histogram equalization method (hereinafter referred to as H-B method).
There is a gradation correction method. This attempts to obtain the best gradation by forming a histogram with gradation as the horizontal axis from a 17-frame video signal, for example, and correcting it so that the histogram after one gradation correction becomes flat. .

For example, the gradation xl (for which histogram F (*) is the first
As shown in Figure AK, if there is an intermediate c/4Vc biased t video signal, this video signal is divided into eight regions along arbitrary boundaries, for example seven, along the gradation axis xltc119, and this region and A histogram is formed, the boundary positions are determined so that this second histogram becomes flat, and the gradation is corrected so that the boundary positions are equally spaced. In other words, as shown in FIG. By performing correction having the input (X) vs. output (y) % characteristic as shown in BK, the histogram '(7) for the corrected floor myK becomes flat as shown in tX1 diagram C<.

By flattening the histogram in this way, the gradation of the video signal is optimized. In this case, the gradation correction curve y = f(x) (indicated by the 19th BK broken line) is
Each boundary is defined by the gradations XI, X2, . . .

So when we think about the aperture of the imaging device.

Assuming that the gradation axis of the image light from the subject is X and the gradation axis of the image light at the photographed temple is y, the relationship between these is defined by a constant l determined by the aperture, and k such as y=ax.

Now, if you think of the aperture adjustment as Hitotsubashi's model.

The gradation correction curve f(X) at this time becomes f(z) = a x , and the degree of freedom in this gradation correction I/c is only five given to Maro.

Therefore, when performing gradation correction using the H-1 method described above,
The degree of freedom in Sakahiragi selection is 1, and the am of the gradation correction aperture diaphragm is such that the stograms of the two areas above and below this boundary are equal.
becomes possible. In other words, E with the median value on the y-axis as the boundary
, if the aperture is adjusted using the flattening of the stogram as an evaluation function, it will be possible to obtain an appropriate amount of gradation. The median value on the P7 axis needs to take into account the influence of T correction and visual distortion, and it may not necessarily be the true median value, and it may be better to choose another value. be.

The present invention has been made with attention to such points.

That is, in the present invention, the image light is passed through the aperture.
At the same time as supplying w< to obtain an Eirin signal, sample this Eirin signal, count the number of samples for which the signal level of these samples is above or below a predetermined level, and calculate the ratio of this counted value to the total number of samples. Adjust the aperture accordingly. An embodiment of the present invention will be described below with reference to the drawings.

At 1124, the subject (Eirin light from 1B is transmitted) is supplied to the COD (4) K through the lens (3), and the signal from the synchronization generator [4] is transmitted to C0D (
41, and the video signal is extracted. This C0
The video signal from D (4) passes through the preamplifier (5) to A.
It is supplied to the D conversion circuit (6). Stiff AD conversion circuit (6
) is supplied to the read-only memory ())k for r correction and -cr is corrected. This r-corrected signal is supplied to the DA conversion circuit (9)k through the synchronization signal addition circuit (8), converted into an analog signal, and output to the output terminal e.
lk @ is taken out.

At the same time, the digital signal from the conversion circuit (6) is supplied to the comparator 0υ, and the digital signal from the reference signal source α1 corresponding to the median value on the y-axis is supplied to the comparator aυ via the latch circuit a3. When the digital signal from the AD conversion circuit (6) is larger than the median value on the y-axis at this k y/(lator aυ), k
" signal is output. Note that the median value on the y-axis may not be the true median value as described above.

This:r7/<L/-The signal from 0υ is counted by the counter α fathom, and this count value is the latch a11! QSk
lI[is latched at the tie-edge of the synchronization signal. Immediately after this latch, the counter is reset to zero. This latch circuit αS signal is the D person conversion circuit (I
ek is supplied. When this DAAD conversion circuit e and the anarch signal are supplied to the drive amplifier fin,

Bias is supplied from the bias voltage GL Nishiki.

The output signal of the triple amplifier aη is then supplied to the adjustment drive circuit 69 of the aperture.

Therefore, in this circuit, the number of pixels for which the signal from the AD conversion circuit (6) is equal to or greater than the median value on the y-axis is counted, and the aperture is adjusted according to the ratio of this counted value to the total number of pixels.

In other words, in Figure @3, the person is underexposed, B is properly exposed, and C is overexposed. In these figures, the left side is a histogram of the Eirin signal, and if a second histogram is drawn for this Eirin signal with the median value O8s as the boundary, it will be as shown on the right. Here, N- is the number of pixels below the median value, and N- is the number of pixels above the median value. Therefore, N−>K+−,
Or, all pixels are out too much. CCD (
The sampled signal exceeding the Guina electric range of 41 is compressed by the white clip function and becomes marked with *. Then, by adjusting the aperture according to the value of the aperture, the gradation correction characteristics are improved as shown in Figure 4, where 1 is when the aperture (2) is slightly closed, and b is the standard , C are cases where the image is slightly opened, and by these corrections, the optimum gradation can be obtained as shown in Figures 1 and 4 BK.

In this way, automatic adjustment of the aperture is performed, and according to the present invention, it is possible to perform adjustment almost equivalent to the average value method, and since no accumulator is required, the circuit configuration is extremely simple.

Furthermore, in this circuit, the circuit can be further simplified by doing the following.

First, when the number of bits of the counter a◆ is set to n, the total number of samples N is set to 8M2n -1. Here, the number N is specified by thinning out the pixels on average from all pixels, or ignoring the top, bottom, left and right of the screen as appropriate.

In this case, the conditions for H,lii are 0 units (10...°...) for the output C of counter a◆.・0], then N, = 2f'''''t-1 to -2n-1.

From this, *crl-t can be used as an index indicating overexposure or underexposure, and Cl3-1-1... Overexposure Cn, Kaoru0 ・
... It can be considered underexposed, and if it is properly exposed.

If Cn goes back and forth between O and 1 and is averaged for one hour, it becomes 9, which is the intermediate value T.

So 11E5el! ! As shown in IK, M8B±Cn- of color α is connected to a 1-bit latch circuit rlt)'
Supply Ic.

By averaging and dispersing the ζ latch output with a low-pass filter (2), the DAAD conversion circuit can be omitted.

Also, even if N-1 per ♂-1k is not possible using this M5 diagram, if N-2''-1 is used, almost the same effect can be expected.Furthermore, K N=Go-(♂1- If x'r was fourteen.

(,,[10...-O], then N+-2n-1 N-,,21-1+norl, and if the reference signal source a is adjusted by taking l into account, the hisdigram is Although not equalized, the effect will be almost the same.

Furthermore, by using the r-corrected signal as the sample signal, the boundaries of the histogram can be selected as O,SK. In this case, it is possible to determine whether M2R of the sample signal exceeds 0.5 based on whether it is 1 or 0.

Therefore, as shown in Figure 96, read-only memory (7
) by supplying the output M2R to the counter 04, the comparator I, the reference number 51fi2, etc. can be omitted. Note that the figure shows the case where >Nzz.zu-1.

By the way, the above-mentioned circuit has almost the same effect as average value type control. Therefore, if this circuit is used for control, considerable overexposure may occur in some areas, and pluming may occur, resulting in deterioration of image quality. In order to overcome this drawback, it is desirable to perform control in consideration of the signal of the peak value method, as in the case of average value banjo.

Figure 7 shows a circuit for this purpose, in which the signal from the Ω conversion circuit (6) is supplied to the latch circuit +21), and the input signal and output signal of this latch circuit QB are supplied to the comparator (to) K. , when the input signal is large, a load signal is supplied to the latch circuit aU. Further, a signal from the latch circuit WI (to) is supplied to the latch circuit and latched by the timing of the @M synchronization signal. Note that after I of this latch, the peak value of the latch circuit is extracted from the latch circuit (to) k. Then, the signal from the latch circuit (to) and the signal from the latch circuit α9 are appropriately weighted by the circuits (to) and (to), and are added by the adder circuit (to), resulting in the DA'R conversion. It is supplied to the road slope.

In this way, the diaphragm can be adjusted in consideration of the H, E method and the peak value method. Note that in this circuit, the weighting is circuit c! 4) The adder circuit ω may be composed of a microphone computer.

Furthermore, Figure 8 is a diagram that aims to share the circuit in the method shown in @Figure 7, Figure 9 ki? The signal from the AD conversion circuit (6) and the signal from the reference signal source α2 are connected to the switch C31.
), and this switch OD signal is supplied to latch circuit 03 (also serving as latch circuit QLl) K. The signal from the comparator an (which also serves as the comparator @) and the load command deflection angle of the reference signal at appropriate timing are supplied to the switch (to) k, and the signal from this switch is supplied to the load terminal of the latch circuit a3. Ru. Others are 7th
It is done as shown in the figure.

In this circuit, by switching the switches 0υ and □□□ to the 1 field 41, the latch circuit 09,
-, control signals of the H-B method and the peak value method are respectively extracted.

Even if this circuit is not used, it is possible to perform aperture adjustment taking into account the H, B method and the peak value method.

Further, based on this figure 8k, if the signals from the AD conversion circuit (6) are sequentially red, green, and real, the peak value will be the peak value of the three signals, and the H, The R control signal is equivalent to the average value of three signals, and one circuit can control red, green,
Back signals can be processed simultaneously.

Thus, according to the present invention, the aperture can be precisely adjusted with an extremely simple configuration.

[Brief explanation of drawings]

1 [Figure 1 is a diagram for explaining the histogram identification method% Figure 2 is a block diagram of an example of the present invention, Is■,
FIG. 4 is the tth diagram of the explanation, and FIGS. 5 to 8 are configuration diagrams of other examples. 12) is the aperture, (4) is the photographic image 11. +6) is a λD conversion circuit, αυ is a comparator, (13 is a reference signal source, and I is a counter. Figure 3) Figure 9

Claims (1)

[Claims] Eirin light is supplied to the imaging unit through the aperture to obtain a video signal, and this Eirin signal is sampled, and the songs of the samples whose signal levels are above or below a predetermined level are counted, and the count value is calculated. An imaging device that adjusts the aperture according to a proportion of the total number of samples.