JP3348320B2 - Video camera - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video camera having non-linear processing means for performing non-linear processing such as gamma correction and knee correction on three primary color signals obtained by an image sensor, and more particularly to a video camera having an automatic white adjustment function. Things.

[0002]

2. Description of the Related Art Generally, a video camera system has a non-linear processing function of performing non-linear processing such as gamma correction and knee correction on a three primary color signal obtained by an image sensor, and a signal of the three primary color signals for an achromatic portion of a subject image. It has a white adjustment function for matching levels.

Conventionally, white adjustment (white balance) of a video camera system is performed by detecting the signal level of each of the three primary color signals on one of the former stage and the latter stage of the non-linear processing means for the three primary color signals obtained by the image sensor. And the gain of the variable gain amplifying means is controlled based on the detected output.

[0004]

In a conventional video camera system in which the signal levels of the three primary color signals are detected at the stage prior to the nonlinear processing means to perform white adjustment, the detection result by the level detection means and the variable gain amplification are used. Since the relationship with the gain of the means is simple and does not depend on the level, the amount of gain correction can be easily calculated from the above detection result. If there is an element that changes the signal level of each signal (for example, variation in circuit, temperature drift, intentional setting of level offset), the white balance is shifted. Although it is possible to perform white adjustment in consideration of the level change amount at the subsequent stage of the level detecting means, it is necessary to calculate the level change amount accurately because the contents to be considered are various in practice. Is extremely difficult.

On the other hand, if the white level is adjusted by detecting the signal levels of the three primary color signals at the stage prior to the output stage, high accuracy can be guaranteed. Is complicated, and it is difficult to calculate the amount of gain correction from the detection result. For example, in the γ correction, the input is raised to the power of 0.45, so that the correction amount cannot be accurately obtained only by the four arithmetic operations.

Accordingly, in the conventional video camera system,
When performing white adjustment, it is extremely difficult to simultaneously improve the convergence time and the convergence accuracy.

Here, it is known that, in the control means for controlling the gain of the variable gain amplifying means, controlling the gain in logarithmic proportion (decibel) is more advantageous than controlling the gain in proportion to the magnification. For example, the gain change amount (painting amount) intentionally for a special effect does not change even if the gain changes for white adjustment. The addition of the gain can be accurately performed, and the relative value can be easily calculated. If the output level of the variable gain amplifying means is constant, the level change amount with respect to the unit gain change becomes constant regardless of the input level, which is suitable for performing feedback control.

However, conventionally, since the calculation of the logarithm is complicated, an accurate correction amount is not calculated, the difference between each level of the red signal and the blue signal with respect to the green signal is obtained, and the difference is multiplied by a coefficient to prevent a divergence. The correction has been performed so that the gain is 1 or less. Therefore, since the correction amount is not appropriate, the number of corrections increases, and a long time is required until convergence.

Accordingly, an object of the present invention is to provide a video camera capable of performing white adjustment in a short time and with high accuracy in view of the above-mentioned conventional problems.

[0010]

According to a first aspect of the present invention, there is provided a video camera system including a non-linear processing unit for performing non-linear processing on three primary color signals output from an image sensor. Variable gain amplifying means capable of variably controlling the gain to be amplified, first level detecting means for detecting the signal level of the three primary color signals before the nonlinear processing means, and signal level of the three primary color signals after the nonlinear processing means And a control for controlling the gain of the variable gain amplifying means so as to match the signal levels of the three primary color signals of the white portion of the subject image based on the respective detection outputs of the respective level detecting means. Means for controlling the gain of the variable gain amplifying means based on the detection output of the first level detecting means. After performing the coarse adjustment for matching the signal level of the three primary color signals of the white portion of the subject image by,
On the basis of the detection output of the second level detecting means, a fine adjustment for matching the signal levels of the three primary color signals of the white portion of the subject image is performed by controlling the gain of the variable gain amplifying means.

A video camera according to a second aspect of the present invention is a video camera system including a non-linear processing means for performing non-linear processing on the three primary color signals output from the image sensor, wherein the gain for amplifying each of the three primary color signals output from the image sensor is variable. Controllable variable gain amplifying means, first level detecting means for detecting the signal levels of the three primary color signals before the nonlinear processing means, and second level detecting means for detecting the signal levels of the three primary color signals after the nonlinear processing means. Level detecting means, and control means for controlling the gain of the variable gain amplifying means in logarithmic proportion so as to match the signal levels of the three primary color signals of the white portion of the subject image based on each detection output by each of the level detecting means. Wherein the control means controls a gain of the variable gain amplifying means based on a detection output of the first level detecting means. After performing a coarse adjustment to match the signal levels of the three primary color signals of the white portion of the subject image, the gain of the variable gain amplifying unit is controlled based on the detection output of the second level detecting unit. Fine adjustment is performed to match the signal levels of the three primary color signals of the white portion.

[0012]

The video camera according to the first aspect of the present invention detects the signal levels of the three primary color signals by first level detecting means provided before the nonlinear processing means for performing nonlinear processing on the three primary color signals output from the image sensor. A signal level of the three primary color signals is detected by a second level detecting means provided at a stage subsequent to the non-linear processing means, and a gain of the variable gain amplifying means is controlled by a control means based on a detection output of the first level detecting means. After performing the coarse adjustment to make the signal levels of the three primary color signals of the white portion of the subject image coincide with each other,
The gain of the variable gain amplifying means is controlled based on the detection output of the level detecting means for fine adjustment for matching the signal levels of the three primary color signals of the white portion of the subject image.

In the video camera system according to the second aspect of the present invention, the signal levels of the three primary color signals are detected by first level detecting means provided before the nonlinear processing means for performing nonlinear processing on the three primary color signals output from the image sensor. A second level detecting means provided at a subsequent stage of the non-linear processing means for detecting a signal level of the three primary color signals, and a control means for controlling a gain of the variable gain amplifying means based on a detection output of the first level detecting means; Is controlled in a logarithmic proportion to make a coarse adjustment to match the signal levels of the three primary color signals of the white portion of the subject image, and then the gain of the variable gain amplifying means is adjusted based on the detection output of the second level detecting means. By performing control in a logarithmic proportion, fine adjustment is performed to match the signal levels of the three primary color signals of the white portion of the subject image.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the video camera according to the present invention will be described in detail with reference to the drawings.

A video camera according to the present invention is, for example, shown in FIG.
It is configured as shown in FIG. This video camera is one in which the present invention is applied to a three-panel color video camera that performs color imaging using three CCD image sensors 101R, 101G, and 101B.

The CCD image sensors 101R, 10R
In 1G and 101B, imaging light from a subject is color-separated into three primary colors of R, G and B by a prism or the like in an imaging optical system 100 and is incident on each imaging surface. The CCD image sensors 101R, 101G, 101B
Captures a subject image formed by the three primary color lights on each imaging surface, and outputs an electrical three primary color signal.

The CCD image sensors 101R, 101R
G and 101B are respectively used for correlated double sampling (CDS) circuits 102R and 102R.
After the clock component is removed by G and 102B, the sensitivity is adjusted by amplifiers 103R, 103G and 103B.

The amplifiers 103R, 103G,
The three primary color signals, the sensitivity of which has been adjusted by 103B, are subjected to processing such as shading correction, gain up, and clamping in analog signal processing circuits 104R, 104G, and 104B, and then to variable gain amplifiers 105R, 105G, and 105B.
105B. The variable gain amplifier 105R,
Each gain of 105G and 105B is controlled by a system controller 114 using a microcomputer.

The variable gain amplifier 105R, 1
The three primary color signals output from the 05G and 105B are low-pass filters 106R and 106R for removing high frequency components.
A / D converter 107R, 1 via 106G, 106B
07G and 107B and are digitized.

The A / D converters 107R, 107G, 1
The digital three primary color signals digitized by 07B are
Digital signal processing circuits 109R, 109G, 10 via first level detection circuits 108R, 108G, 108B.
9B.

The digital signal processing circuits 109R and 109R
In 9G and 109B, γ is used for the digital three primary color signals.
Performs process processing including non-linear processing such as correction, detail, aperture addition, linear matrix, and flare correction.

The digital signal processing circuits 109R and 109R
The three primary color signals output from 9G and 109B are supplied to the encoder 111 via the second level detection circuits 110R, 110G and 110B.

The encoder 111 is, for example, an NTSC
And converts the three primary color signals from the digital signal processing circuits 110R, 110G and 110B into an NTSC composite video signal. The digital composite video signal obtained by the encoder 111 is converted into an analog signal by a D / A converter 112 and supplied to a video tape recorder, a monitor device, and the like (not shown) via a low-pass filter 113.

The first level detection circuit 108
R, 108G and 108B correspond to the A / D converter 107
Each signal level of the digital three primary color signals digitized by R, 107G, and 107B is detected, and the detected output is supplied to the output 114. Similarly, the second level detection circuits 110R, 110G, and 110B detect the signal levels of the digital three primary color signals that have been subjected to process processing including non-linear processing by the digital signal processing circuits 109R, 109G, and 109B. , And supplies the detection output to the output 114.

The system controller 114 detects the positions of a gain setting switch 122 and an auto white balance activation switch 114 provided on the camera body, and receives a command from a remote controller 116 as an external device, and Has a function of performing an auto white balance control operation in a control procedure as shown in the flowchart of FIG.

That is, in the control operation of the automatic white balance, first, in step 1, a green signal is output for the white portion of the subject image based on the detection output by the second level detection circuits 110R, 110G and 110B. Each level difference R of the red signal R and the blue signal B with respect to G
-G and BG are obtained.

In the next step 2, each level difference R-
It is determined whether coarse adjustment is necessary based on G, BG.

If the determination result in step 2 is YES
That is, when each of the above-mentioned level differences RG and BG is equal to or more than a predetermined value and coarse adjustment is necessary, the process shifts to the coarse adjustment mode and the processing of steps 3 to 6 is repeated.

In step 3, based on the detection outputs of the first level detection circuits 108R, 108G, and 108B, each level of the red signal R and the blue signal B with respect to the green signal G for the white portion of the subject image. Difference RG, B
-Find G. In the next step 4, each level difference R
It is determined whether or not -G and BG satisfy the coarse adjustment standard. Further, in step 5, when the determination result in step 4 is NO, that is, when the coarse adjustment standard is not satisfied, the gains required for the red signal R and the blue signal B are calculated. In step 6, the gains of the variable gain amplifiers 105R and 105G for the red signal R and the blue signal B are set, and the process returns to step 3.

The calculation of the gain required for the red signal R and the blue signal B in step 5 is as follows: new R gain [dB] = current R gain + 20 × log (G / R)
= Current R gain + 20 × (logG−logR) 1st
Formula New B gain [dB] = Current B gain + 20 × log (G / B)
= Current R gain + 20 × (logG−logB) 2nd
The value of the expression is obtained by a table reference method (lookup table).

Here, in the table reference system, for example, if each signal level of the green signal G and the red signal R is represented by 12 bits, the size of the table is 2 <12> 4096.
It requires calculation results for different G levels and R levels. In order to reduce this, the lower 4 bits of 12 bits are truncated and the gain calculation is performed with the upper 8 bits, and the size of the table becomes only the calculation result for 256 kinds of G levels and R levels of 2 to the eighth power. However, the accuracy is deteriorated. In particular, when the G level and the R level are small and the values are close to each other, the difference in the upper 8 bits is small, and the accuracy is deteriorated.

Therefore, the above first equation is calculated as follows: new R gain [dB] = current R gain + 20 × log (G / R)
= Current R gain + 20 × (logG−logR) = Current R gain + 20 × (log (G × M) −log (R × M) (where M is a coefficient) Equation 3 Modified into Equation 3 Then, the gain is obtained by the table reference method.

That is, after the signal levels of the green signal G and the red signal R are multiplied by the same coefficient M and logarithmically converted, the result is the same even if the difference is obtained. After converting to a value as large as possible, lower bits are rounded and the value is obtained by a table reference method according to the number of bits in the table, so that the accuracy of gain calculation can be improved.

When the G level and the R level are 12 bits, the processing is performed according to, for example, the procedure shown in the flowchart of FIG.

First, it is checked whether the G level or the R level is higher (step a), and the higher level is set as a reference X (steps b and c).

Next, it is checked whether or not the reference X is 800H or more (a state where 1 is set in the most significant bit) (step d). Shift the level one bit to the left (×
2) Do (step e).

Then, when it becomes 800H or more, 008H is added to the G level and the R level to round off (step f), and then the G level and the R level are shifted right by 4 bits (× （). To 8 bits (step g).

As a result, the lower 4 bits can be rounded after bit shifting to the left (× 2) within a range where the G level and the R level do not overflow.

In the coarse adjustment mode, the result of determination in step 4 is YES, that is, the level differences RG, BG based on the detection outputs of the first level detection circuits 108R, 108G, 108B. Step 3 to Step 6 until
Is repeated.

If the decision result in the above step 2 is NO, that is, if the second level detection circuits 110R, 11R
0G, each level difference R based on detection output by 110B
-G and BG are not equal to or greater than predetermined values, and the coarse adjustment is not required, and if the determination result in step 4 is YES, that is, the first level detection circuits 108R, 108G,
108B, each level difference RG based on the detection output,
If BG satisfies the coarse adjustment standard, the mode shifts to the fine adjustment mode, and the processing of steps 7 to 10 is repeated.

In step 7, based on the detection outputs of the second level detection circuits 110R, 110G, and 110B, the respective levels of the red signal R and the blue signal B with respect to the green signal G for the white portion of the subject image Difference RG, B
-Find G. In the next step 8, each level difference R
It is determined whether or not -G and BG satisfy the white adjustment standard. Further, in step 9, when the determination result in step 8 is NO, that is, when the white adjustment standard is not satisfied, the gains required for the red signal R and the blue signal B are calculated. In step 10, the gains of the variable gain amplifiers 105R and 105G for the red signal R and the blue signal B are set, and the process returns to step 7.

The calculation of the gain required for the red signal R and the blue signal B in step 9 is as follows: new R gain [dB] = current R gain + 20 × log (GR)
/ N ··· Equation 4 New B gain [dB] = Current B gain + 20 × log (GB)
/ N: Performed according to the fifth equation. Here, N is a coefficient for determining the feedback amount.

In the fine adjustment mode, the result of determination in step 8 is YES, that is, the level differences RG, BG based on the detection outputs of the second level detection circuits 110R, 110G, 110B. Steps 7 to 8 are repeatedly performed until the state satisfies the white adjustment standard.

As described above, the digital signal processing circuits 108R, 108G, and 10 include nonlinear processing such as gamma correction.
8B, a first level detection circuit 108R, 1
After performing coarse adjustment based on the detection outputs of the digital signal processing circuits 108R and 108B,
The second level detection circuit 110 provided at the subsequent stage of 108B
By performing fine adjustment based on the detection outputs of R, 110G, and 110B, white adjustment can be performed in a short time and with high accuracy.

[0045]

As described above, in the video camera according to the first aspect of the present invention, the three primary color signals are detected by the first level detecting means provided before the nonlinear processing means for performing the nonlinear processing on the three primary color signals output from the image sensor. In addition to detecting the signal level, the signal level of the three primary color signals is detected by a second level detecting means provided at a stage subsequent to the non-linear processing means, and the control means controls the signal level based on the detection output of the first level detecting means. The gain of the variable gain amplifying means is controlled to make a coarse adjustment to match the signal levels of the three primary color signals of the white portion of the subject image, and then the variable gain amplifying means is adjusted based on the detection output of the second level detecting means. By performing the fine adjustment to control the gain of the three primary color signals of the white portion of the subject image so as to match the signal levels, the white adjustment can be performed in a short time and with high accuracy. That.

In the video camera system according to the second aspect of the present invention, the signal levels of the three primary color signals are detected by first level detecting means provided before the nonlinear processing means for performing nonlinear processing on the three primary color signals output from the image sensor. In addition, the signal level of the three primary color signals is detected by second level detecting means provided at the subsequent stage of the nonlinear processing means, and the variable gain amplifying means is controlled by the control means based on the detection output of the first level detecting means. The gain of the variable gain amplifying means is controlled based on the logarithmic proportionality to make the signal levels of the three primary color signals of the white portion of the subject image coincide with each other, and based on the detection output of the second level detecting means. By controlling the gain in logarithmic proportion and making fine adjustments to match the signal levels of the three primary color signals of the white part of the subject image, the white level can be reduced in a short time and with high accuracy. It is possible to perform an integer.

[Brief description of the drawings]

FIG. 1 is a block diagram showing one embodiment of a video camera according to the present invention.

FIG. 2 is a flowchart showing a processing procedure of an automatic white balance control operation of the video camera.

FIG. 3 is a flowchart showing a procedure of a rounding process required for a gain calculation in the control of the auto white balance of the video camera.

[Explanation of symbols]

100 imaging optics 101R, 101G, 101B CCD image sensor 102R, 102G, 102B correlated double sampling circuit 103R, 103G, 103B amplifier 104R, 104G, 104B ... analog signal processing circuit 105R, 105G, 105B ... variable gain amplifier 107R, 107G, 107B ... A / D converter 108R, 108G, 108B ... first level detection circuit 109R, 109G, 109B ... Digital signal processing circuit 110R, 110G, 110B ... Second level detection circuit 111 ... Encoder 112 ... ······ D / A converter 114 ······ System controller 115 ···・ ・ ・ ・ ・ Auto white balance start switch 116 ・ ・ ・ ・ ・ ・ Remote controller

Claims (2)

(57) [Claims] 1. A video camera having a non-linear processing means for performing non-linear processing on three primary color signals output from an image sensor, wherein the gain for amplifying each of the three primary color signals output from the image sensor is variably controllable. Gain amplifying means, first level detecting means for detecting the signal level of the three primary color signals before the nonlinear processing means, and second level detecting means for detecting the signal level of the three primary color signals after the nonlinear processing means. , and control means for controlling the gain of said variable gain amplifying means to match the signal level of the three primary color signals of the white portion of the subject image based on each detection output of the respective level detection means, said control means, Detection output of the first level detection means
Controlling the gain of the variable gain amplifying means based on
The signal level of the three primary color signals of the white part of the subject image is
After performing the coarse adjustment to match the above, the second level detection is performed.
Based on the detection output of the variable gain amplifying means.
By controlling the gain, the three primary color signals of the white part of the subject image
A video camera for making fine adjustments to match the signal level of the video camera . 2. A video camera having non-linear processing means for performing non-linear processing on three primary color signals output from an image sensor, wherein the gain for amplifying each of the three primary color signals output from the image sensor is variably controllable. Gain amplifying means, first level detecting means for detecting the signal level of the three primary color signals before the nonlinear processing means, and second level detecting means for detecting the signal level of the three primary color signals after the nonlinear processing means. , and control means for controlling the gain of said variable gain amplifying means to match the signal level of the three primary color signals of the white portion of the subject image based on each detection output of each of the level detection means logarithmically, the control The means is a detection output of the first level detection means.
Controlling the gain of the variable gain amplifying means based on
The signal level of the three primary color signals of the white part of the subject image is
After performing the coarse adjustment to match the above, the second level detection is performed.
Based on the detection output of the variable gain amplifying means.
By controlling the gain, the three primary color signals of the white part of the subject image
A video camera for making fine adjustments to match the signal level of the video camera .