JP2011023943A - Auto white balance system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an auto white balance system capable of reducing a time required for white balance adjustment. <P>SOLUTION: This auto white balance system includes: an adjustment gain calculation part 17 for calculating adjustment gain of white balance adjustment; a primary color signal average value calculation part 14 for calculating a primary color signal average value for extraction pixels extracted for calculating the adjustment gain; an adjustment gain multiplying part 15 for multiplying the primary color signal average value calculated by the primary color signal average value calculation part 14 by the adjustment gain calculated by the adjustment gain calculation part 17; and a color difference signal average value calculation part 16 for calculating a color difference signal average value by conversion from the primary color signal average value multiplied by the adjustment gain. The adjustment gain calculation part 17 calculates adjustment gain based on the color difference signal average value calculated by the color difference signal average value calculation part 16. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an auto white balance system.

  2. Description of the Related Art Conventionally, imaging apparatuses that perform white balance adjustment according to the color temperature of a light source with respect to a video signal have become widespread. An auto white balance (Automatic White Balance control, hereinafter referred to as “AWB” as appropriate) system is a system that automatically adjusts white balance (see, for example, Patent Document 1). In general, the AWB system extracts pixels suitable for white balance adjustment from one frame of data of a photographed image, and calculates an adjustment gain from the result of integrating the color difference signal values. The calculated adjustment gain is reflected in the image of the next frame. The white balance is converged by repeating the calculation of the adjustment gain and the reflection of the calculated adjustment gain until the average value of the color difference signals becomes zero. In this case, it usually takes several frames or more until the white balance converges. In particular, when an image having a hue or saturation far from the convergence point of white balance is targeted, the time required for convergence is further increased. If the adjustment gain change range is increased in order to increase the speed until the white balance is converged, a so-called hunting phenomenon may occur in which control that passes through the convergence point is repeated.

Japanese Patent Laid-Open No. 4-304093

  An object of the present invention is to provide an auto white balance system that can shorten the time required for white balance adjustment.

  According to an aspect of the present invention, an adjustment gain calculation unit that calculates an adjustment gain for white balance adjustment, and a primary color signal average that is an average value of primary color signal values for the extracted pixels extracted for calculating the adjustment gain. A primary color signal average value calculation unit that calculates a value; an adjustment gain multiplication that multiplies the primary color signal average value calculated by the primary color signal average value calculation unit and the adjustment gain calculated by the adjustment gain calculation unit; A color difference signal average value calculation unit that calculates a color difference signal average value that is an average value of color difference signal values by conversion from the primary color signal average value multiplied by the adjustment gain, and the adjustment gain An auto white balance system is provided, wherein the calculation unit calculates the adjustment gain based on the color difference signal average value calculated by the color difference signal average value calculation unit.

  According to the present invention, the time required for white balance adjustment can be shortened.

1 is a block diagram illustrating a configuration of an imaging device. 6 is a flowchart for explaining a processing procedure by the image processing apparatus. The block diagram which shows the structure of the AWB system which concerns on embodiment. The block diagram which shows the structure of the AWB system which concerns on a comparative example.

  Hereinafter, an auto white balance system according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

(Embodiment)
FIG. 1 is a block diagram illustrating a configuration of an imaging apparatus including an AWB system according to an embodiment of the present invention. The imaging device includes an imaging lens 1, an IR cut filter 2, an image sensor unit 3, an image processing device 4, and a recording unit 5.

  The imaging lens 1 constitutes an optical system for capturing light from a subject. The IR cut filter 2 removes infrared light from the light captured by the imaging lens 1. The image sensor unit 3 captures a subject image by converting light captured by the imaging lens 1 into a signal charge. The image sensor unit 3 takes in the pixel values of red (R), green (G), and blue (B) in an order corresponding to the Bayer array, generates an analog image signal, and sequentially at a gain corresponding to the imaging condition. Amplify. Furthermore, the image sensor unit 3 converts the obtained image signal from an analog system to a digital system. The image processing device 4 performs various processes on the digital image signal input from the image sensor unit 3. The image processing apparatus 4 includes an AWB system described later. The recording unit 5 records the image data input from the image processing device 4 in a memory or a recording medium.

  FIG. 2 is a flowchart for explaining a processing procedure by the image processing apparatus 4. In step S1, noise such as fixed pattern noise, dark current noise, and shot noise is removed (noise reduction). Noise reduction is useful when photographing at low illuminance or when the sensitivity of the light receiving element is low, and is preferably performed before and after resolution restoration processing. In step S2, luminance unevenness due to a light amount difference between the central portion and the peripheral portion due to the imaging lens 1 is corrected (shading correction). In step S3, resolution restoration processing is performed by the resolution restoration means.

  In step S4, contour enhancement processing is performed using a correction coefficient calculated based on the imaging conditions of the image sensor unit 3 and the position of each pixel. In step S5, pixel interpolation processing (demosaicing) is performed on the digital image signals transmitted in the Bayer array order. In demosaicing, a sensitivity level value of an insufficient color component is generated by interpolation processing of an image signal obtained by capturing a subject image. In step S6, white balance adjustment is performed by the AWB system. In step S7, linear color matrix processing for obtaining color reproducibility is performed. In step S8, the saturation and brightness of the image displayed on the display or the like are corrected by gamma correction. Note that the processing procedure described in this embodiment is merely an example, and other processing may be added, optional processing may be omitted, the order may be changed, and the like.

  FIG. 3 is a block diagram showing a configuration of the AWB system 10 according to the present embodiment. The color difference luminance signal generation unit 11 generates a color difference signal and a luminance signal based on the video signal input to the AWB system 10. The color difference luminance signal generation unit 11 generates a color difference signal and a luminance signal from the video signal before the white balance adjustment.

  The color difference luminance signal integration unit 12 calculates a color difference signal integrated value obtained by integrating the color difference signal values and a luminance signal integrated value obtained by integrating the luminance signal values for the extracted pixels extracted for calculating the adjustment gain for white balance adjustment. Calculate each. The extracted pixel is extracted by determining, for example, for each pixel whether or not it has a color difference signal value within a predetermined range and a luminance signal value within a predetermined range suitable as a reference for white balance adjustment.

  The chrominance luminance signal average value calculation unit 13 calculates, for the extracted pixels, a chrominance signal average value that is an average value of chrominance signal values and a luminance signal average value that is an average value of luminance signal values. The color difference signal average value is calculated by dividing the color difference signal integrated value by the number of extracted pixels. The luminance signal average value is calculated by dividing the luminance signal integrated value by the number of extracted pixels.

  The primary color signal average value calculation unit 14 calculates a primary color signal average value that is an average value of primary color signal values for the extracted pixels. The primary color signal average value is calculated for R, G, and B by conversion from the color difference signal average value and the luminance signal average value calculated by the color difference luminance signal average value calculation unit 13.

  The adjustment gain multiplication unit 15 multiplies the primary color signal average value calculated by the primary color signal average value calculation unit 14 and the adjustment gain calculated by the adjustment gain calculation unit 17. The adjustment gain multiplication unit 15 adjusts the white balance of the primary color signal average value by multiplying the primary color signal average value by the adjustment gain. The adjustment gain multiplication unit 15 multiplies the primary color signal average value by 1 or an arbitrary initial value as the adjustment gain in the first calculation after the video signal is input to the AWB system 10.

  The color difference signal average value calculation unit 16 calculates a color difference signal average value, which is an average value of the color difference signal values, by conversion from the primary color signal average value multiplied by the adjustment gain. The color difference signal average value calculation unit 16 generates a color difference signal average value after white balance adjustment. The adjustment gain calculation unit 17 calculates an adjustment gain based on the color difference signal average value calculated by the color difference signal average value calculation unit 16. The V latch unit 18 holds and outputs the adjustment gain calculated by the adjustment gain calculation unit 17. The AWB system 10 is realized by a software configuration, for example.

  Here, before describing the operation of the AWB system 10 according to the present embodiment, the configuration and operation of a conventional AWB system will be described. FIG. 4 is a block diagram showing a configuration of a conventional AWB system 20 according to a comparative example. The adjustment gain multiplication unit 21 adjusts the white balance of the video signal by multiplying the video signal input to the AWB system 20 by the adjustment gain calculated by the adjustment gain calculation unit 25 and V-latched by the V latch unit 26. . The color difference signal generation unit 22 generates a color difference signal by conversion from the video signal multiplied by the adjustment gain.

  The color difference signal integration unit 23 calculates a color difference signal integration value for the extracted pixels. The color difference signal average value calculation unit 24 calculates a color difference signal average value for the extracted pixels. The adjustment gain calculation unit 25 calculates the adjustment gain based on the color difference signal average value calculated by the color difference signal average value calculation unit 24. The V latch unit 26 outputs the adjustment gain calculated by the adjustment gain calculation unit 25 once per frame. In the case of the AWB system 20, since the color difference signal average value is calculated from the color difference signal multiplied by the adjustment gain, the feedback of the adjustment gain in the white balance adjustment is once per frame. For this reason, it usually takes several frames or more until the white balance converges, and it takes time to adjust the white balance.

  Returning to FIG. 3, the AWB system 10 according to the present embodiment calculates the primary color signal average value from the color difference signal and the luminance signal generated before the adjustment gain is multiplied, and the primary color signal multiplied by the adjustment gain. A color difference signal average value is calculated from the average value. For example, a video signal is sent for each pixel in one clock. A color difference luminance signal is generated by the color difference luminance signal generation unit 11 with respect to the video signal for each pixel, and integration is performed by the color difference luminance signal integration unit 12. Here, the calculation of the primary color signal average value from the calculation of the color difference luminance signal average value (operation from the color difference luminance signal generation unit 11 to the primary color signal average value calculation unit 14) is performed once per frame. Gain calculation and feedback (operation from the adjustment gain multiplication unit 15 to the adjustment gain calculation unit 17) are repeated a plurality of times, for example, every clock. The AWB system 10 repeats the calculation and feedback of the adjustment gain a plurality of times within the video blanking period.

  The adjustment gain calculation and feedback are repeated until the adjustment gain calculated by the adjustment gain calculation unit 17 converges within a predetermined range, and is stopped when the adjustment gain converges within the predetermined range within the blanking period. Is done. The AWB system 10 can sufficiently converge the adjustment gain within the blanking period by operating the adjustment gain multiplication unit 15, the color difference signal average value calculation unit 16, and the adjustment gain calculation unit 17 at high speed. . The V latch unit 18 holds the converged adjustment gain and outputs it once per frame, for example. The adjustment gain output from the AWB system 10 is reflected in the white balance adjustment of the video signal of the next frame. Thereby, the image after white balance adjustment can be obtained instantaneously.

  In this way, the AWB system 10 has an effect that the time required for white balance adjustment can be shortened by allowing the adjustment gain to converge by high-speed calculation within the blanking period. Note that the AWB system 10 is not limited to the case where the calculation of the adjustment gain is stopped when the calculated adjustment gain converges within a predetermined range. For example, the AWB system 10 may repeat the calculation of the adjustment gain from the start to the end of the blanking period regardless of whether or not the calculated adjustment gain has converged within a predetermined range. By repeating the calculation of the adjustment gain even after the calculated adjustment gain has converged within a predetermined range, highly accurate white balance adjustment can be performed using the adjustment gain that has been completely converged. When the calculation of the adjustment gain is stopped within the blanking period, the power consumption can be reduced compared to the case where the calculation of the adjustment gain is continued during the blanking period. In addition, the AWB system 10 is not limited to one that converges the adjustment gain in one frame, and may be one that converges the adjustment gain in steps in two or more frames.

  10 AWB system, 11 color difference luminance signal generation unit, 13 color difference luminance signal average value calculation unit, 14 primary color signal average value calculation unit, 15 adjustment gain multiplication unit, 16 color difference signal average value calculation unit, 17 adjustment gain calculation unit.

Claims (6)

An adjustment gain calculation unit for calculating an adjustment gain for white balance adjustment;
A primary color signal average value calculation unit that calculates a primary color signal average value that is an average value of primary color signal values for the extracted pixels extracted for calculating the adjustment gain;
An adjustment gain multiplication unit that multiplies the primary color signal average value calculated by the primary color signal average value calculation unit by the adjustment gain calculated by the adjustment gain calculation unit;
A color difference signal average value calculation unit that calculates a color difference signal average value that is an average value of color difference signal values by conversion from the primary color signal average value multiplied by the adjustment gain;
The auto white balance system, wherein the adjustment gain calculation unit calculates the adjustment gain based on the color difference signal average value calculated by the color difference signal average value calculation unit. A color difference luminance signal generation unit that generates a color difference signal and a luminance signal based on the input video signal;
A color difference signal average value that is an average value of color difference signal values and a luminance signal average value that is an average value of luminance signal values are calculated from the color difference signal and the luminance signal generated by the color difference luminance signal generation unit. A color difference luminance signal average value calculation unit,
The primary color signal average value calculation unit calculates the primary color signal average value by conversion from the color difference signal average value calculated by the color difference luminance signal average value calculation unit and the luminance signal average value. The auto white balance system according to claim 1.   3. The auto white balance system according to claim 1, wherein the adjustment gain calculation is repeated a plurality of times within a video blanking period.   4. The auto white balance system according to claim 3, wherein the calculation of the adjustment gain is stopped when the calculated adjustment gain converges within a predetermined range.   The auto white balance system according to claim 3, wherein the calculation of the adjustment gain is repeated until the blanking period ends. The extraction pixel further includes a color difference luminance signal integration unit that calculates a color difference signal integration value by integration of the color difference signal value and a luminance signal integration value by integration of the luminance signal value,
In the color difference luminance signal generation unit and the color difference luminance signal integration unit, processing is performed in units of one clock,
The auto white balance system according to claim 2, wherein the color difference luminance signal average value calculation unit and the primary color signal average value calculation unit perform processing in units of one frame.

Images