JPH0693781B2 - Automatic white balance control circuit - Google Patents

Description

Description: TECHNICAL FIELD OF THE INVENTION The present invention relates to an automatic white balance control circuit used in a color television camera or the like.

[Technical background of the invention and its problems]

Currently, most color television cameras have three primary color signals.
R, G, B information is read from the image sensor, and each of the three primary color signals R,
This is a mechanism for reproducing G and B and performing a composite operation on them to obtain a color video signal. In this case, when a white subject is imaged, it is desirable that the levels of the three primary color signals R, G, B be equal, and the gain adjustment of each signal is performed. This is
This means white balance adjustment. For example, in a studio camera or the like, a white object is photographed, and the gain control knobs are turned so that the signal levels of R, G, and B become equal to achieve white balance.

On the other hand, in portable color television cameras and home video cameras, it is necessary to easily and quickly adjust the white balance. Therefore, when the adjustment switch is turned on while a white subject is being imaged, the white balance is automatically adjusted. Things are being considered. This white balance setting circuit
It is a mechanism that has means for detecting the level of the color difference signal, and controls the gain of each primary color signal so that the level of the color difference signal becomes zero in the image pickup state of the white subject.

However, with the above method, it is necessary to perform white balance setting work after imaging a white object.
It is needed every time the lighting changes. A user who uses a home video camera often forgets such work, and sometimes the white balance setting work cannot be performed when the illumination changes during the imaging period. In order to deal with such a problem, a method of adding a means for detecting the color temperature of the illumination and taking a white balance according to the color temperature is also considered. However, with this method, it is difficult to match the characteristics of the image sensor and the characteristics of the color temperature detector, and a detection window is required. If the user's hand is located near this detection window, the white balance will be lost. To end up,
Also, there is a problem that the cost of the detector becomes high.

[Object of the Invention]

The present invention has been made in view of the above circumstances, and an object thereof is to provide an automatic white balance that can automatically obtain an appropriate white balance without requiring special work for setting the white balance. It is to provide a control circuit.

[Outline of Invention]

In the present invention, a signal corresponding to a low-saturation subject is detected, and the plurality of color difference signals generated based on the plurality of color information signals corresponding to the signal positions are integrated so that the integrated values are equal to each other. The gain of each amplifier of the color information signal is controlled by the integrated signal indicating the integrated value. In this case, in detecting the video signal position corresponding to the subject with low saturation in the color video signal, the output characteristic of the detection output with respect to the saturation range changes depending on whether the luminance signal level is high or low. When the level is high, the saturation range is widened, and when the luminance signal level is low, the saturation range is narrowed, so that the extraction threshold of the detection output is varied according to the level of the luminance signal.

Example of Invention

 Embodiments of the present invention will be described below with reference to the drawings.

According to the present invention, for example, in an apparatus for obtaining an image information signal corresponding to three kinds of color light and synthesizing the image information signals to obtain a color video signal, based on the image information signal corresponding to a subject with low saturation among the image information signals. Only detect signals. Next, when gain control is performed on the image information signal corresponding to the subject with low saturation, the white balance is achieved by making the integral values of the signals corresponding to the three types of color light equal.

FIG. 1 shows that two color difference signals (RY) and (BY) used for a television signal are divided by a luminance signal (Y) to obtain (RY) / Y, (BY) / Y. Indicates the two-dimensional coordinates indicated by the vertical axis and the horizontal axis. In this two-dimensional coordinate, the point that each point of each color of the color bar chart used for adjustment of the television camera moves according to the color temperature change for illumination is plotted. The solid line is an illumination with a color temperature of 3000 ° K, the dashed line is an illumination of 4000 ° K, and the alternate long and short dash line is an illumination of 5000 ° K. The color temperature of the illumination light is changed from 2500 ° K to 5500 ° K. Here, for a subject of the same color, if the spectrum of illumination light is the same, if the brightness changes, the color difference signal and the luminance signal level change in proportion. Therefore, a signal having a value obtained by dividing the color difference signal by the luminance signal can be considered as an amount representing color independently of brightness.

Therefore, the coordinate axes in FIG. 1 can be regarded as the base axes of the quantities representing colors. From the measurement results of FIG. 1, the range in which the point corresponding to the white subject moves on the coordinates is a range in which the absolute value is sufficiently smaller than the range in which the point corresponding to the highly saturated subject moves. Therefore, it is possible to remove the signal corresponding to the high-saturation subject and extract the signal corresponding to the low-saturation subject regardless of the illumination light.

This invention pays attention to the above-mentioned point, and among the video signals, 3
When the output ratio of the primary color signal has a large bias, it is determined as a high saturation signal. Then, when the output ratio of the three primary color signals is small, it is determined to be a low saturation signal, and the signal at this time is extracted as a signal corresponding to the low saturation subject. Then, the gains of the three primary color signals are controlled so that the integrated values of the three primary color signals at this time become equal to each other. As a result, white balance is achieved within a sufficiently small error range.

Also, for high-brightness signals, signals in a wide saturation range are extracted as signals corresponding to low-saturation subjects, and for low-brightness signals, signals in a narrow saturation range are set as low-saturation subjects. It may be extracted as a corresponding signal. This makes it possible to widen the range in which the white balanced state is maintained, without widening the error range when determining the signal corresponding to the low-saturation subject. That is, a highly saturated subject has a large bias in spectral reflectance, the reflectance as a whole does not increase, and it is unlikely that a highly bright subject is a highly saturated subject. Therefore, when extracting a signal corresponding to a low-saturation subject for a high-luminance signal,
Even if the saturation range of the extracted signal is expanded, the judgment error range is not expanded. As a result, the extraction capability of the signal corresponding to the low-saturation subject can be expanded in the high-luminance to low-luminance signal, which is effective in maintaining the white balance state.

It also compares several color-difference signals with the luminance signal and detects only when the level of the luminance signal is higher than any level of the color-difference signals to be compared, and determines the signal at this time as a signal corresponding to the low-saturation subject. You may extract. Here, the color difference signal is a signal obtained by adding and subtracting the three primary color signals, and is a signal that becomes zero when the three primary color signals are equal. Further, an arbitrary color difference signal can be created by adding / subtracting the RY signal and the BY signal at an appropriate ratio.

Next, the reason why the signal corresponding to the low-saturation subject can be extracted will be described below.

FIG. 2 is a diagram showing a part related to white in FIG. 1 in an enlarged manner. Now, consider a case where only the signals represented in the shaded area in FIG. 2 are extracted as the signals corresponding to the low-saturation subject.

The values of the points where the straight lines 2A, 2B and 2C in FIG. 2 intersect the (RY) / Y axis are a1 to a3, and the values of the points where the (BY) / Y axis intersects are the b1 to b3.
Then, the triangular area shown by the diagonal lines can be expressed as follows.

This equation can be rewritten as:

Therefore, the three color difference signals (1 / a1) (RY) + (1 / b1) (BY) (1 / a2) (RY) + (1 / b2) (BY) (1 / a3) (R−Y) + (1 / b2) (B−Y) If the level of the luminance signal Y is higher than any of the signals, the signal at this time is the signal in the shaded area, that is, the low signal. It is possible to extract only the signal corresponding to the saturation subject. Similarly,
By comparing the levels of several color difference signals and the level of the luminance signal, it becomes possible to extract signals in the low saturation region in an arbitrary form, and the slope of the straight line in FIG. 2 may be variously changed.

′ Further, in the color difference signal and the luminance signal to be compared, the black level of the luminance signal may be lower than the zero level of the color difference signal. As a result, a signal with a high luminance level can be extracted in a wide saturation range, and a signal with a low luminance level can be extracted in a narrow saturation range.

That is, now, it is considered to compare the color difference signal α (RY) + β (BY) with the luminance signal Y. If the direct current reproduction level of the luminance signal at this time is lowered by x from the zero level of the color difference signal, the portion where the luminance signal becomes larger can be expressed by the following equation.

α (R−Y) + β (B−Y) <Y−x This equation can be rewritten as follows.

α (R−Y) / Y + β (B−Y) / Y <1−x / Y Here, 1−x / Y becomes smaller as the value of Y becomes smaller. Therefore, as the brightness level is lower, only the part where the value obtained by dividing the color difference signal by the brightness signal is smaller is extracted. Therefore, with this method, the part with high brightness level is
The signal can be extracted in a wide saturation range, and the signal with a low luminance signal can be extracted in a small saturation range.

FIG. 3 is a concrete example of the configuration using the present invention.

In FIG. 3, R, G, B signals of the video camera are input to the input terminals 11a, 11b, 11c, respectively. Of the three primary color signals, the R signal and the B signal pass through the amplifiers 12a and 12c, respectively, and are input to one of the input terminals of the subtracters 15 and 16, respectively. Furthermore, each R, G, B signal passes through the amplifiers 12a, 12b, 12c and the multiplier
It is input to 13a, 13b, 13c, and is combined by the adder 14 at a ratio of 0.3, 0.54, 0.11 respectively. This produces the luminance signal Y. The luminance signal Y is also input to the subtracters 15 and 16. As a result, from the subtracters 15 and 16, the color difference signal R-
Y and BY are obtained respectively.

Here, each of the amplifiers 12a and 12c has a gain control signal G
The gain is controlled by C1 and GC2. The higher the level of the control signal, the smaller the gain. This gain control signal GC1,
GC2 is a signal that automatically adjusts the white balance.

The color difference signals R-Y, B-Y and the luminance signal Y are input to the encoder 17 and are signals for forming a video signal.

Next, using the color difference signals RY and BY, three types of difference signals S
1, S2, S3 are made.

α1 (R−Y) + β1 (B−Y) = S1 α2 (R−Y) + β2 (B−Y) = S2 α3 (R−Y) + β3 (B−Y) = S3 The difference signal S1 is the multiplier 22a. , 22b and the adder 23, the difference signal S2 is created by the multipliers 24a, 24b and the adder 25, and the difference signal S3 is created by the multipliers 26a, 26b and the adder 27.

The signals S1, S2, S3 and the luminance signal Y are clamp circuits 28, 29,
Direct current reproduction by 30,31 respectively. In this case, the luminance signal Y is DC reproduced by the clamp circuit 31, but the DC reproduction level is lower than the DC reproduction level of the difference signal by x. Further, α1 = (1-x) / a1, α2 = (1-x) / a2, α3 = (1-x) / a3, β1 = (1-x) / b1, β2 = (1-x) / The values of α1 to α3 and β1 to β3 are selected so that b2, β3 = (1−x) / b3. a1-a3 and b1-b3 are constants.

The direct current reproduction signal obtained as described above is supplied to the comparator 32 to
Compared at 34. The comparator 32 compares the outputs of the clamp circuits 28 and 31, the comparator 33 compares the outputs of the clamp circuits 29 and 31, and the comparator 34 compares the outputs of the clamp circuits 30 and 31. The outputs of the comparators 32 to 34 are input to the OR circuit 35, and the outputs of the OR circuit 35 are input to the control terminals of the blanking circuits 18 and 19 as control signals.

The blanking circuits 18 and 19 are supplied with RY and BY signals, respectively, and pass the signal only when the control signal is at the low level, and when the control signal is at the high level, the output is at zero level. And The outputs of the blanking circuits 18 and 19 are input to the integrators 20 and 21 and integrated, and the integrated outputs are used as the gain control signals GC1 and GC2 of the amplifiers 12a and 12c. The outputs of the integrators 20 and 21 continue to rise when the input is positive, continue to fall when the input is negative, and hold their values when the input is zero.

In the above system, the conditions when the RY signal and the BY signal pass through the blanking circuits 18 and 19 are as follows.

This equation can be rewritten as:

Therefore, when Y = 1, the RY signal, BY which can produce the calculation result in the shaded area shown in FIG.
The signal can pass through the blanking circuits 18,19,
When Y <1, the threshold value is varied, and the blanking circuit 18 outputs the RY signal and the BY signal which can produce the calculation result in a region narrower than the hatched region in FIG. , 1
Can pass 9 That is, the characteristics of the detection output obtained from the OR circuit 35 change according to the luminance signal level. Then, R- which has passed through the blanking circuits 18 and 19
When the Y signal and BY signal are positive, the outputs of the integrators 20 and 21 continue to rise, reducing the gain of the amplifier, and the RY signal and BY signal.
Decrease the signal level. Also, the blanking circuit 1
When the RY and BY signals passing through 8 and 19 are negative, the levels of the RY and BY signals are raised by the opposite effect. In this way, the amplifier 12 is controlled so that the RY and BY signals passing through the blanking circuits 18 and 19 become zero.
The gains of a and 12c are adjusted and converge. RY signal, BY
The zero signal means that the gain-controlled R, G, B signals are equal. Therefore, in the above circuit, when the convergence state is reached, the integrated values of the R, G, and B signals of the part of the video signal where only the low-saturation part is extracted are equal, and this condition is the white balance. Can be said to have been obtained within a sufficient error range.

As described above, according to this embodiment, it is possible to achieve white balance without requiring any artificial work for achieving white balance.

The values of α1 to α3 and β1 to β3, x in the above embodiment are not limited to the above examples, and it is possible to select appropriate values according to the characteristics of the image pickup device. Further, the number of difference signals used for detecting the low saturation signal is not limited to three, and it is possible to use more signals.

Further, there is an image pickup apparatus which obtains a Y signal and R, B signals from an image pickup element instead of the three primary color signals. For such an apparatus, the R, G, B signals are obtained from the above embodiment. The Y signal obtained from the image sensor may be directly used by omitting the portion for generating the Y signal by using.

The present invention uses (RY) / Y, (BY) / Y as a method for detecting a signal corresponding to a low-saturation subject in the above embodiment.
The axis coordinate is used as a concept, and a signal (difference signal) represented on this coordinate is created and detected.

However, the method of detecting the signal corresponding to the low-saturation subject is not limited to this method, and the xy chromaticity diagram, u
A signal corresponding to a low-saturation subject may be detected by creating a signal representing coordinates on a chromaticity diagram such as a v chromaticity diagram and detecting the level thereof.

Further, the present invention can be applied not only to a color image pickup device but also to the white balance by utilizing the signal itself used for a video signal. Therefore, the present invention can be applied to a color television receiver and a video tape. It can also be incorporated into a recorder. Then, it is also possible to correct the broadcast signal or the recorded signal in which the white balance is deviated to a good white balance state. Furthermore, the present invention can be applied to a color photographic printer.

When the present invention is used in a color photographic printer, in order to control the gain of the three primary color signals, the incident light on the negative is passed through a filter for correction of the three primary colors, and the density of this filter is changed.

FIG. 4 shows an embodiment applied to a color photographic printer. The light from the light source 41 is collimated by the condenser lens 42 and passes through the cyan filter 43, the magenta filter 44, and the yellow filter 45. Filter 42,43,44
, The concentration is changed according to the control signal. The light transmitted through the filter portion passes through the negative film 46, then passes through the lens 47, and is focused on the color image sensor 49 through the mirror 48. The signal processing circuit 50 detects only the low-saturation portion of the signal obtained from the image pickup device 49, and integrates the R-Y, G-Y, and BY signals for that portion.
The density of each of the cyan, magenta, and yellow filters 43, 44, 45 is controlled by ▼ and ▲ ▼. Then, when the average value of each of the RY, GY, and BY signals for the signal extracted as the low saturation signal becomes zero, ▲ ▼, ▲ ▼, ▲ at this time
Fix each control signal of ▼. Next, the signal processing circuit 50 controls the mirror 48 to form an image of the incident light on the photographic printing paper 51 for printing.
By matching the characteristics of the image sensor 49 and the signal processing circuit 50 with the characteristics of the printing paper, proper color correction can be performed.

〔The invention's effect〕

As described above, according to the present invention, it is possible to automatically obtain an appropriate white balance without requiring artificial work. Further, since a signal obtained from the image pickup device itself is used, a detector for illumination light is not required, which is inexpensive and does not cause a problem due to a mismatch between the characteristics of the detector and the image pickup device. In addition, it is possible to automatically rebalance the white balance even for a signal that has already been recorded, reproduced, or broadcast.

[Brief description of drawings]

FIGS. 1 and 2 are explanatory views showing how the (RY) / Y signal and (BY) / Y signal for each subject are changed by the change of illumination light, and FIG. FIG. 4 is a structural explanatory view showing an embodiment of the invention, and FIG. 4 is a structural explanatory view showing a usage example of the present invention. 12a to 12c ... amplifier, 13a to 13c, 22a to 26b ... multiplier, 14,2
3,25,27 ... Adder, 18,19 ... Blanking circuit, 20, 21 ...
Integrator, 28-31 ... Clamp circuit, 32-34 ... Comparator, 35 ...
OR circuit.

Claims (1)

[Claims] 1. An apparatus for obtaining a color video signal by using each image information signal corresponding to a plurality of types of color light, for detecting a video signal position corresponding to a subject with low saturation in the color video signal, When the luminance signal level is high and when the luminance signal level is low, the output characteristic of the detection output with respect to the saturation range changes. When the luminance signal level is high, the saturation range is widened, and when the luminance signal level is low, the saturation range is narrow. As described above, the first means having a varying means for varying the extraction threshold of the detection output according to the level of the luminance signal, and the respective signals based on the respective image information signals corresponding to the detection position are By the second means for extracting by the detection output of the first means and integrating each extracted signal, and by each integrated signal obtained by the second means,
An automatic white balance control circuit comprising: third means for controlling the gain of the amplifier of each image information signal so that the gain of each image information signal corresponding to the detection position becomes equal.