JP4725184B2 - White balance adjusting device and gain adjusting method - Google Patents

Description

  The present invention relates to a white balance adjusting device and a gain adjusting method used for a video camera or the like.

The conventional white balance adjustment device has a gain determination means for determining the gain of the color signal, determines a limited area of the gain of the color signal according to the illuminance of the subject, and determines the gain within this limited area. A balanced white balance process was performed.
Japanese Unexamined Patent Publication No. 2000-78606 (pages 2 to 6, FIG. 1)

  In the conventional white balance adjustment device, the switching of the restricted area is performed discontinuously with a single threshold value, so that the white balance processing becomes unstable before and after the switching, or the illuminance of the subject changes so as to cross the threshold value. In some cases, the color tone of the photographed image is greatly different.

  The present invention has been made to solve the above-described problem, and provides a white balance adjustment device capable of smoothly changing the color tone of a captured image before and after switching of a restricted area.

  In the white balance adjustment device according to the present invention, the color signal gain limiting region is defined as the first limiting region, the second limiting region different from the first limiting region, the illuminance of the subject, and the first limiting. The third limit area specified based on the specific value specifying the area and the second limit area is switched to one of the third limit area and the third limit area.

  The present invention is specified based on a first limit area, a second limit area that is different from the first limit area, and a specific value that specifies the illuminance of the subject, the first limit area, and the second limit area. By providing the third restricted area, the color tone of the captured image can be changed smoothly before and after the restriction area is switched.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a white balance adjusting apparatus according to Embodiment 1 for carrying out the present invention. In FIG. 1, 1 is a lens, 2 is a TG (Timing Generator) for controlling pixel accumulation time, an analog gain amplifier, and the like, and is a color signal for converting an optical image of a subject output from the lens 1 into an electrical signal. The conversion means 3 is composed of an A / D converter or the like, and converts the electrical signal converted by the color signal conversion means 2 into digital signals of B (blue), G (green), R (red), and the color of each color Color signal detection means for detecting a signal. Reference numerals 4, 5, and 6 are configured by using a multiplier or the like, and each of B color level determining means (BCL), G color level determining means (GCL), and R color for determining the color level of the B, G, and R color signals. Level determining means (RCL). Reference numeral 7 denotes an image processing circuit for performing image interpolation processing, color tone adjustment, and the like based on signals from the B color level determination means 4, the G color level determination means 5, and the R color level determination means 6. Reference numeral 8 denotes the image processing circuit 7. This is an image memory for storing the calculation results as image data. 9 is a B integrating circuit for integrating the output of the B color level determining means 4, 10 is a G integrating circuit for integrating the output of the G color level determining means, and 11 is an R integrating circuit for integrating the output of the R color level determining means. . Reference numeral 12 denotes an illuminance detection means for detecting the illuminance of the subject from the outputs from the lens 1 and the color signal conversion means 2, and reference numeral 13 denotes a restriction area determination means for determining a gain restriction area for each color signal based on the illuminance detected by the illuminance detection means. , 14 are gain determining means for determining the gain of each color signal based on the outputs of the B integrating circuit 9, the G integrating circuit 10 and the R integrating circuit 11. Reference numeral 15 denotes a microcomputer having an illuminance detection means 12, a limited area determination means 13, and a gain determination means 14.

  Next, the operation of the white balance adjustment apparatus configured as described above will be described. The lens 1 outputs an optical image of the subject to the color signal conversion unit 2 and information on the lens aperture to the illuminance determination unit 12. The color signal conversion unit 2 converts the optical image from the lens 1 into an electrical signal, and outputs the electrical signal to the color signal detection unit 3 and the pixel accumulation time and the analog gain level to the illuminance determination unit 12.

  The color signal detection means 3 converts the electrical signal from the color signal conversion means 2 into digital data of three colors B, G, and R using an internal A / D converter, and calculates a color signal for each color. Next, the B color signal is output to the B color level determination means 4, the G color signal is output to the G color level determination means 5, and the R color signal is output to the R color level determination means 5. The B color level determination means 4 uses an internal multiplier or the like, and multiplies the B color signal output from the color signal detection means 3 and the gain determined by the gain determination means 12 to the image processing circuit 7 and the B integration circuit 9. Is output. Similarly, the G color level determination unit 5 and the R color level determination unit 6 multiply the G color signal and the R color signal output from the color signal detection unit 3 by the gain determined by the gain determination unit and the image processing circuit 7 and G. The signals are output to the integrating circuit 10 and the R integrating circuit 11, respectively.

  The image processing circuit 7 performs necessary image processing such as color interpolation processing, color adjustment processing, filter processing, resolution conversion processing on the outputs of the B color level determination means 4, the G color level determination means 5 and the R color level determination means 6. Processing is performed, and the processed image data is output to the image memory 8. The image memory 8 stores this image data. The B integrating circuit 9, the G integrating circuit 10, and the R integrating circuit 11 integrate the respective inputs, and then normalize the integrated pixels to obtain the average signal levels Sb, Sg, and Sr of each color as gain determining means 14 Output to.

  The illuminance detection unit 12 calculates the illuminance of the subject based on information such as the aperture of the lens 1, the pixel accumulation time from the color signal conversion unit 2, and the analog gain level. The restricted area determination unit 13 calculates the restricted areas of the gains of the R, G, and B color signals based on the illuminance calculated by the illuminance detection unit 12. The gain determining means 14 calculates the gain of each color signal based on the average signal level of each color output from the B integration circuit 9, the G integration circuit 10, and the R integration circuit 11. When the gain calculated here is within the area calculated by the restricted area determining means 13, the calculated gain is used. When the gain is outside the area calculated by the restricted area determining means 13, the calculated gain is closest to the calculated distance. The gain in the restricted region is determined as a gain to be output to the B color level determining unit 4, the G color level determining unit 5 and the R color level determining unit 6, and the microcomputer 15 uses the determined gain to each color level determining unit. Output. Thus, the gain of each color signal is determined.

  The series of processes described above is performed for each frame, and the process of determining the gain of the color signal of the next frame using the information in the previous frame is repeatedly performed.

Next, a method for determining the limited area of the color signal in the limited area determining means 13 will be described. The restricted area determination means 13 is set with a first threshold value L1 and a second threshold value L2 less than L1. Further, a first restriction area D1 corresponding to the case where the photographing state is outdoor and a second restriction area D2 different from D1 corresponding to the case where the photographing state is indoor are set. Here, each limited region is a planar region in which the horizontal axis represents the gain Tr of the R signal with respect to the gain of the G signal, and the vertical axis represents the gain Tb of the B signal with respect to the gain of the G signal. FIG. 2 shows a diagram representing the restricted area. In FIG. 2, D1 is an area indicated by a grid, and D2 is indicated by diagonal lines so as to include an area of gain that can be taken with respect to a light source area corresponding to outdoor in the daytime, cloudy day or shade in the daytime. The area is set to include a gain area that can be taken with respect to a light source area corresponding to a fluorescent light, dusk (dusk), or outdoors during the daytime. Specifically, D1 is an area on a plane with Tr and Tb as axes expressed by (Expression 4) to (Expression 7), and D2 is expressed by (Expression 8) to (Expression 11).
ATr + BTb ≧ X1 (Formula 4)
ATr + BTb ≦ X2 (Formula 5)
ATr−BTb ≧ Y1 (Formula 6)
ATr−BTb ≦ Y2 (Expression 7)
ATr + BTb ≧ X3 (Expression 8)
ATr + BTb ≦ X4 (Expression 9)
ATr−BTb ≧ Y3 (Expression 10)
ATr−BTb ≦ Y4 (Formula 11)
In (Expression 4) to (Expression 11), X1 to X4 and Y1 to Y4 are constants, are specific values that specify D1 and D2, and A and B are coefficients of Tr and Tb, respectively. The above-mentioned area is appropriately set.

Next, the limited area determination unit 13 calculates a third limited area D3 based on the specific value and the illuminance Lx of the subject detected by the illuminance detection unit 12. Here, D3 is a region calculated by (Expression 12) to (Expression 15).
ATr + BTb ≧ X ₁ + {(X3−X1) × (Lx−L2) / (L1−L2)} (Formula 12)
ATr + BTb ≦ X ₂ + {(X4−X2) × (Lx−L2) / (L1−L2)} (Formula 13)
ATr−BTb ≧ Y ₁ + {(Y 3 −Y ₁ ) × (Lx−L 2) / (L ₁ −L 2)} (Formula 14)
ATr−BTb ≦ Y ₂ + {(Y4−Y2) × (Lx−L2) / (L1−L2)} (Formula 15)

  Next, a case where A = B = 1 will be described with reference to FIG. 2 and FIG. FIG. 3 shows the relationship between the illuminance of the subject detected by the illuminance detection means 12 and the restricted area.

  The restricted area determination unit 13 determines that the shooting state is outdoor when the illuminance Lx of the subject detected by the illuminance detection unit 12 is L1 or more, and determines that the shooting state is D1, and when Lx is L2 or less, the shooting state is indoor. And D2 is determined as a restricted area. Further, when Lx is between L1 and L2 (hereinafter referred to as an intermediate region), the third restricted region D3 calculated by (Equation 12) to (Equation 15) that changes according to Lx is restricted. Determine the area. Then, the gain determination means 14 calculates the gain of each color signal using the determined limited area.

  For example, when Lx is the intermediate value A in the intermediate region as shown in FIG. 3, the region surrounded by the alternate long and short dash line shown in FIG. 2 is shown, and when Lx is the intermediate value B shown in FIG. The areas surrounded by the broken lines shown in FIG. Accordingly, D3 changes so as to continuously interpolate D1 and D2 according to Lx when Lx is in the intermediate region.

Further, the restriction region may be determined by a straight line and a curve having a constant ratio such that ATr × BTb is constant or ATr / BTb is constant. That is, D1 may be an area expressed by (Expression 16) to (Expression 19), and D2 may be an area expressed by (Expression 21) to (Expression 24).
ATr × BTb ≧ X1 (Expression 16)
ATr × BTb ≦ X2 (Expression 17)
ATr / BTb ≧ Y1 (Expression 18)
ATr / BTb ≦ Y2 (Formula 19)
ATr × BTb ≧ X3 (Expression 20)
ATr × BTb ≦ X4 (Formula 21)
ATr / BTb ≧ Y3 (Formula 22)
ATr / BTb ≦ Y4 (Equation 23)
In this case, regarding the specific values corresponding to X1, X2,..., The same calculation as in (Expression 12) to (Expression 15) is performed, and by calculating D3, it is possible to determine the restricted area in the intermediate area.

  The trajectory of the intersection point of the boundary line of D3 is set so as to draw a straight line between the intersection point of the corresponding boundary line of D1 and the intersection point of the boundary line of D2, but the trajectory of the intersection point of the boundary line of D3 is You may set so that it may become a curve which connects these. In this case, the right side of (Expression 12) to (Expression 15) may be adjusted.

  As described above, in the present embodiment, an intermediate area is provided when switching the limited area of the gain of each color signal according to the illuminance of the subject, and the limited area is continuously changed. Even if it occurs, it is possible to realize a smooth and natural white balance process without causing the hue of the captured image to change abruptly.

  Further, since D3 is calculated from the specific values of D1 and D2 and the illuminance from the illuminance detection means 12, the stability can be improved without adding a new parameter compared to the case where the limited region is switched with a single threshold. High white balance processing can be realized.

  Next, the gain determining method in the gain determining means 14 will be described with reference to FIG. 5 when the restricted area determining means 13 determines a restricted area as shown in FIG. In FIG. 4, the horizontal axis represents the R gain Tr with respect to the G gain, the vertical axis represents the B gain Tb with respect to the G gain, and the hatched portion represents the gain limiting region. FIG. 5 is a flowchart showing a gain determination method.

First, in step S1, when the microcomputer 15 is started, the gain is initially set. In step S2, the gain determining means 14 reads the average signal levels Sb, Sg and Sr of B, G and R from the B integrating circuit 9, the G integrating circuit 10 and the R integrating circuit 11. In step S3, a gain for bringing the R average signal level Sr of the next frame closer to the G average signal level Sg is calculated from the R gain in the previous frame and the ratio of Sr and Sg. The calculation formula is shown in (Formula 1).
Kr (n + 1) = Kr (n) × Sg / Sr (Formula 1)
In (Formula 1),
Kr (n): gain of the nth frame with respect to R Kr (n + 1): gain of the (n + 1) th frame with respect to R.

In step S4, the same operation as in step S3 is performed for B. That is, the gain for bringing the average signal level Sb of B close to the average signal level Sg of G according to (Equation 2) is calculated.
Kb (n + 1) = Kb (n) × Sg / Sb (Expression 2)
In (Formula 2),
Kb (n): Gain of the nth frame with respect to B Kb (n + 1): Gain of the (n + 1) th frame with respect to B.

In step S5, the G gain of the next frame is determined. Here, the gain of G is not changed, and the set value of the previous frame is used as it is. That is, the gain of G is
Kg (n + 1) = Kg (n) (Formula 3)
It can be expressed.
In (Formula 3),
Kg (n): gain of the nth frame with respect to the G pixel Kg (n + 1): gain of the (n + 1) th frame with respect to the G pixel.

  In step S6, the gain of each color signal calculated in steps S3 to S5, Tr (n + 1) and Tb (n + 1) calculated from Kr (n + 1), Kg (n + 1), and Kr (n + 1) are the limited region determining means 13. It is determined whether or not the restriction area shown in FIG. Here, Tr (n + 1) = Kr (n + 1) / Kg (n + 1) and Tb (n + 1) = Kb (n + 1) / Kg (n + 1). As a result, if it is determined that it is not within the restricted area, the process proceeds to step S7. If it is determined that it is located within the restricted area, the process proceeds to step S8. However, the point on the boundary line is assumed to be within the restricted area.

In step S7, if it is determined in step S6 that Tr and Tb do not fall within the restricted area, within the restricted area that is closest to the coordinates of Tr and Tb indicating the gain of each color pixel calculated in steps S3 to S5. Are used as correction values. For example, in FIG. 4, when the x point is Tr or Tb indicating the gain of each color pixel calculated in steps S3 to S5, the point O within the restricted area closest to the x point is corrected. Use as Based on the determined values of Tr and Tb, the values of Kr (n + 1) and Kb (n + 1) are recalculated without changing the value of Kg (n + 1).

  In step S8, the microcomputer 15 calculates Kb (n + 1), Kg (n + 1), and Kr (n + 1), which are the gain levels of the respective colors calculated in steps S3 to S5 or recalculated in step S7, for the gain of the next frame. Is output to the corresponding color level determining means.

  By repeating the above steps S1 to S8, the gain of each color signal of the next frame can be determined based on the gain of each color signal of the previous frame. In addition, by determining the gain within the limited region of the gain, it is possible to perform white balance adjustment with high accuracy even for a subject whose specific color is biased.

Embodiment 2. FIG.
As shown in FIG. 6, the microcomputer 15 according to the first embodiment may be configured to include illuminance change suppression means 16 for suppressing temporal changes in illuminance. The configuration and operation other than the microcomputer 15 are the same as those in the first embodiment. In this case, the illuminance detected by the illuminance detection unit 12 is calculated by the illuminance change suppression unit 16 according to the illuminance of the subject in the previous frame. The restricted area determining means 13 determines the restricted area based on the illuminance calculated by the illuminance change suppressing means 16.

Next, the operation of the illuminance change suppression means 16 will be described. The illuminance change suppression unit 16 calculates the illuminance of the current frame from the illuminance of the current subject detected by the illuminance detection unit 12 and the illuminance of the subject of the previous frame. Based on the illuminance calculated by the illuminance change suppression means 16, the limited area determination means 13 determines a gain restriction area as in the first embodiment. In this case, the calculation formula for the illuminance of the subject is expressed by (Expression 24).
L (n + 1) = (1-R) × L (n) + R × Ltmp (Equation 24)
In (Equation 24),
L (n + 1): Illuminance at the (n + 1) th frame L (n): Illuminance at the nth frame Ltmp: Current illuminance R detected by the illuminance detecting means 12: Time constant (specified in a range of 0 ≦ R ≦ 1)
It is.

  FIGS. 7 and 8 show examples of changes in subject illuminance for each frame when R = 1/2. In FIGS. 7 and 8, the horizontal axis represents the frame number, and the vertical axis represents the illuminance. Further, ◯ indicates the illuminance detected by the illuminance detection means 12, and x indicates the illuminance calculated by the illuminance change suppression means 16. As is apparent from FIGS. 7 and 8, by using the illuminance change suppression means 16, it is possible to smoothly change the illuminance used by the restricted area determination means 13.

  In the white balance adjustment device configured as described above, even when there is a sudden change in the illuminance of the subject, the restricted area can be switched smoothly, and a smoother and more stable white balance process can be realized. It is possible.

It is a block diagram of the white balance adjustment apparatus which shows Embodiment 1 of this invention. It is a figure which shows switching of the restriction | limiting area in Embodiment 1 of this invention. In Embodiment 1 of this invention, it is the figure which showed the relationship between the illumination intensity of a to-be-photographed object, and a restriction | limiting area | region. It is a figure showing the restriction | limiting area | region in Embodiment 1 of this invention. It is a flowchart which shows the gain determination method in Embodiment 1 of this invention. It is a figure which shows the microcomputer in Embodiment 3 of this invention. It is a figure which shows the frame number and subject illuminance in Embodiment 3 of this invention. It is a figure which shows the frame number and subject illuminance in Embodiment 3 of this invention.

Explanation of symbols

2 color signal conversion means, 3 color signal detection means, 4 B color level determination means, 5 G color level determination means, 6 R color level determination means, 11 illuminance detection means, 12 illuminance detection means, 13 restricted area determination means, 14 Gain determining means, 16 Illuminance change suppressing means

Claims (3)

Color signal conversion means for converting an optical image of a subject into a color signal;
Illuminance detection means for detecting the illuminance of the subject;
Illuminance change suppression means for calculating the illuminance of the current frame based on the illuminance detected by the illuminance detection means and the illuminance of the previous frame;
Limiting area determining means for determining a limiting area of the gain of the color signal based on the illuminance calculated by the illuminance change suppressing means ,
Gain determining means for determining the gain of the color signal detected by the color signal converting means within the limited region;
Color level determining means for determining a color level using the color signal detected by the color signal converting means and the gain determined by the gain determining means;
A white balance adjusting device having
The restricted area determining means includes
Based on the illuminance calculated by the illuminance change suppression means for the restricted area,
A first restricted area,
A second restricted area different from the first area;
A third specified based on the illuminance calculated by the illuminance change suppression means, a first specific value for specifying the first restricted area, and a second specified value for specifying the second restricted area. Restricted area,
A white balance adjustment device characterized by switching to any of the above. The restricted area determination means determines the restricted area,
When the illuminance calculated by the illuminance change suppression means is equal to or greater than a first threshold, the first restricted area,
When the illuminance is less than the second threshold value that is less than the first threshold value, the second restricted area,
When the illuminance is between the first threshold and the second threshold, the third restricted area,
The white balance adjusting device according to claim 1, wherein An illuminance detection step for detecting the illuminance of the subject;
Illuminance change suppression step of calculating the illuminance of the current frame based on the illuminance detected in the illuminance detection step and the illuminance of the previous frame,
Based on the illuminance calculated in the illuminance change suppression step, a limited region determination step for determining a limited region of the gain of the color signal converted from the optical image of the subject;
A gain determining step of determining a gain of the color signal within the limited region;
A gain adjustment method comprising:
In the restricted area determination step,
Based on the illuminance calculated in the illuminance change suppression step the limit area,
A first restricted area,
A second restricted area different from the first area;
The illuminance calculated in the illuminance change suppression step, a first specific value that identifies the first restricted area,
A gain adjustment method characterized by switching to any one of a third limited region specified based on a second specific value that specifies the second limited region.

Images