JP2009038483A - Imaging apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an imaging apparatus capable of improving the image quality by suppressing the generation of false color, while performing pixel addition processing for an output signal of imaging pixels for R and B, and actualizing lower power and higher resolution by performing the pixel addition processing. <P>SOLUTION: The imaging apparatus is configured so that prior to a waveform being distorted by gamma correction (non-linear processing); a band-dividing circuit 106 divides a G signal which has not been subjected to pixel addition into a first G signal having the frequency band same as those of R and B signals subjected to pixel addition and a second G signal also, containing a signal having a frequency band higher than that of the first G signal; a color difference generating circuit 109 generates a color difference signal from the first G signal, having distortion and frequency band matching those of the R, B signals and the R, B signals; and a luminance signal generating circuit 108 generates a luminance signal from the second G signal also containing a signal, having frequency band higher than that of the first G signal and the R, B signals. Thus, a video signal in which false color will not occur at all, and the resolution is high can be generated. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an imaging apparatus that performs pixel addition and gamma correction.

  In recent years, many consumer video cameras have a function capable of recording not only moving images but also high-definition still images, as well as moving images that can shoot high-definition movies with 1125 scanning lines. Accordingly, in addition to the increase in the number of pixels of the image sensor, high-speed driving is rapidly progressing in order to increase the number of output pixels per field. However, the increase in the number of pixels and the high-speed driving of the image sensor lead to an increase in power consumption for driving.

  As one method for solving the problem, for example, as disclosed in Patent Document 1 (Japanese Patent Laid-Open No. 2006-270298), an imaging apparatus that suppresses an increase in power consumption by performing pixel addition has been proposed. The configuration is shown in FIG.

  In FIG. 8, imaging elements 801a to 801c are elements that independently photoelectrically convert imaging light that has been color-separated into R (red), G (green), and B (blue), and are configured by, for example, CCD image sensors. Has been. In this embodiment, a G image sensor 801a that photoelectrically converts G image light, an R image sensor 801b that photoelectrically converts R image light, and a B image sensor 801c that photoelectrically converts B image light. It is composed of The R image sensor 801b and the B image sensor 801c are spatially shifted by ½ pixel in the horizontal direction with respect to the G image sensor 801a. Hereinafter, such an arrangement is referred to as “horizontal pixel shift arrangement”.

  The image sensor driving circuit 802 drives the image sensors 801a to 801c.

  The drive pulse changing circuits 803a and 803b are controlled so as to thin out reset pulses output to the R image sensor 801b and the B image sensor 801c among the drive pulses output from the image sensor drive circuit 802 to the image sensors 801a to 801c. To do. Thereby, pixel addition can be performed.

  The pixel addition selection circuit 804 performs control to change the pixel addition number by performing reset pulse thinning control by the drive pulse changing circuits 803a and 803b.

  The analog signal processing circuits 805a to 805c include a CDS circuit that performs a sample operation and a hold operation for each output of the image sensors 801a to 801c, and an AGC circuit that performs gain control so as to keep the signal level constant.

  Analog / digital converters (hereinafter referred to as A / D) 806a to 806c convert analog signals output from the analog signal processing circuits 805a to 805c into digital signals.

  The digital signal processing circuit 807 performs digital signal processing on the digital signals output from the A / Ds 806a to 806c, and outputs a luminance (hereinafter referred to as Y) signal and a color difference (hereinafter referred to as C) signal.

  The operation will be described below.

  When imaging light separated into R, G, and B primary colors by a spectral means such as a prism (not shown) enters the imaging apparatus, the imaging elements 801a to 801c generate driving pulses output from the imaging element drive circuit 802. Based on each photoelectric conversion process. Analog electric signals corresponding to G, R, and B are output.

  The drive pulse output to each image sensor by the image sensor drive circuit 802 is a reset for the charge detection amplification unit that detects and amplifies the signal charge that is horizontally transferred in addition to the pulses for vertical and horizontal transfer, and outputs the signal charge as an electrical signal. There are pulses. The drive pulse changing circuits 803a and 803b perform pixel addition by thinning out the reset pulses for the image sensors 801b and 801c and changing the cycle. For example, by thinning out the pulses, if the period of the reset pulse becomes twice the period before thinning out, the signal charges for two consecutive pixels input to the charge detection amplification unit are accumulated without being reset. Pixel signals for pixels are added.

  The pixel addition selection circuit 804 is a circuit that selects the horizontal pixel addition number for setting the period of the reset pulse that is changed by the drive pulse changing circuits 803a and 803b. select. As a result, the cycle of the reset pulse input to the R and B image sensors becomes twice the cycle of the G reset pulse, and the R and B signals are output by adding pixel signals for two horizontal pixels. On the other hand, the G signal is output without being added as usual because the period of the reset pulse is not changed.

  FIG. 9 is a diagram illustrating the spatial positions of pixels output from the image sensors 801a to 801c. FIG. 9A shows the spatial position of each pixel when pixel addition processing is not performed for all of the G, R, and B signals. FIG. 9B shows the spatial position of each pixel when the R signal and the B signal are subjected to pixel addition processing as in the operation during moving image shooting. That is, at the time of moving image shooting, the R image sensor 801b performs pixel addition processing such as (R1 + R2), (R3 + R4),... And outputs them as pixel signals R12, R34,. Similarly, in the B image sensor 801c, pixel addition processing such as (B0 + B1), (B2 + B3),... Is performed and output as pixel signals B01, B23,.

  After the above control, the analog electrical signals of R, G, and B output from the image sensors 801a to 801c are processed by the analog signal processing circuits 805a to 805c and A / Ds 806a to 806c as described above. The signal is input to the signal processing circuit 807.

  For example, as in the example disclosed in Patent Document 2 (Japanese Patent Application Laid-Open No. 2006-86807), generally in the digital signal processing circuit 807, first, the brightness of the display device with respect to the R, G, and B signals. Is linearly expressed, a gamma correction having the characteristics shown in FIG. 10 is performed, and then a Y signal and a C signal are generated from the R, G, and B signals. The conversion equation for generating the Y signal and the C signal from the R, G, B signals is, for example, ITU-R BT. Defined for HDTV (High Definition TV). There is what is defined in the 709 standard, which is as follows.

Y = 0.2126R + 0.7152G + 0.0722B (Formula 1)
Pr = −0.1146R−0.3854G + 0.5000B (Equation 2)
Pb = 0.5000R-0.4542G-0.0458B (Formula 3)
In Equations 2 and 3, Pr and Pb are components of the C signal.

  In this apparatus, a portion missing from the input R, G, and B digital signals by the pixel addition processing, that is, a pixel signal between R12 and R34 in FIG. 9B. After interpolating the pixel signal between B01 and B23 by a double oversampling process, and combining the number of pixels of the G signal, the R signal, and the B signal, the Y signal and the C signal are expressed by the above Equation 1, Equation 2, Generate and output based on Equation 3. An example of the oversampling filter is a transfer function H (z) as shown in Equation 4 below.

H (z) = (1 + 2z ⁻¹ + z ⁻² ) / 4 (Formula 4)
As described above, G pixels that require a relatively high frequency characteristic due to a high ratio to the Y signal do not perform pixel addition processing, and perform pixel addition processing for other R signals and B signals, respectively, thereby degrading resolution. High-sensitivity video can be obtained while minimizing the image quality. In addition, since the R imaging element 801b, the B imaging element 801c, and the circuit that processes the signals output from these elements have the drive frequency lowered by the pixel addition process, an effect of reducing power can be obtained. Can do.
JP 2006-270298 A JP 2006-86807 A

  However, in the imaging apparatus having the above-described configuration, a false color signal (hereinafter referred to as a false color) is generated, and there is a problem that, for example, an achromatic subject is colored even if it is photographed. This will be described next.

  The R and B signals are subjected to a low-pass filter (hereinafter referred to as LPF) process of a transfer function H (z) as in Equation 5 below in the horizontal direction by pixel addition processing, and the sampling frequency is halved. It becomes equivalent.

H (z) = (1 + z ⁻² ) (Formula 5)
Further, the R and B signals are also subjected to the oversampling filter of Equation 4 in order to match the number of pixels with the G signal when generating the Y and C signals. On the other hand, no filtering process is performed on the G signal.

  FIG. 11 is a diagram illustrating frequency characteristics of the G signal and the R and B signals. Here, fs represents a sampling frequency of a signal that is not subjected to pixel addition processing. The amplitude characteristics of the G signal and the R and B signals are greatly different from each other as shown in FIG. 11 due to the difference in processing as described above with respect to the G signal and the R and B signals. Further, since the sampling frequency of the R and B signals is halved by the pixel addition process, the signal of the band of 4 / fs or higher included when the pixel is not added is turned back to the band of 4 / fs or lower. It will be different from the signal.

From the above, when an achromatic subject is photographed, the R, G, B signals are the same signal when pixels are not added.
Pr = Pb = 0
That is, it becomes an achromatic color. However, at the time of pixel addition processing, the frequency characteristics of the R, B signal and G signal are greatly different, and the amplitude of the R, B signal and G signal is also different at the same frequency.
Pr, Pb ≠ 0
And false color occurs.

  Further, since the C signal does not need a high band compared to the Y signal, it is common to apply LPF to the C signal generation signal. Therefore, it is possible to apply an LPF to the G signal so as to be in the same band as the R and B signals, but even if the bands are combined, a false color is generated in this apparatus. This is because the gamma correction processing performed on the R, G, and B signals is non-linear processing, and distortion occurs in the signal after gamma correction.

  FIG. 12 is a diagram in which a single-frequency sine wave and a waveform after performing a gamma correction process thereon are superimposed. As shown in FIG. 12, it can be seen that the waveform is distorted by the gamma correction processing, so that a new frequency component is generated in the waveform originally containing only a single frequency. As described above, a new frequency component is generated by the gamma correction processing. However, when pixels are added, the frequency characteristics of the signal are greatly different between the G signal and the R and B signals as described above. Even in the band of 4 / fs or less, which includes the aliasing component of the signal of the band of 4 / fs or more not included in the G signal, the new frequency component generated by the gamma correction processing is also the G signal. , R and B signals are different. For this reason, even when an achromatic subject is photographed, the signal after gamma correction has different waveforms for the G signal and the R and B signals. Accordingly, even when the bands of the G signal and the R and B signals are combined when the C signal is generated after the gamma correction, a false color is generated because the waveforms are different.

  An object of the present invention is to improve the image quality by suppressing the generation of false colors while performing pixel addition processing on the output signals of the R and B image sensors, and by performing pixel addition processing. An object of the present invention is to provide an imaging apparatus that can achieve low power and high resolution.

  A first configuration of an imaging apparatus according to the present invention includes a first imaging element capable of adding m pixels (m is an integer of 2 or more) in the horizontal direction, the vertical direction, or both horizontal and vertical directions, and the horizontal direction. A second imaging element capable of adding n pixels (n is an integer of 1 or more, n <m) in both the vertical direction and the horizontal and vertical directions, and an output signal of the second imaging element, Band division means for dividing and outputting a first output signal having the same signal band as the output signal of the first image sensor and a second output signal having a band including a higher frequency than the first output signal And nonlinear processing for performing nonlinear processing of input / output characteristics on the first and second output signals output from the band dividing unit and the output signal output from the first imaging device, respectively. And a luminance signal from the output signal of the nonlinear processing means A luminance signal generating means for generating; and a color difference signal generating means for generating a color difference signal from an output signal of the nonlinear processing means, wherein the color difference signal generating means is subjected to nonlinear processing by the nonlinear processing means. A color difference signal is generated from an output signal of one image pickup device and the first output signal of the band dividing unit subjected to nonlinear processing by the nonlinear processing unit.

  According to a second configuration of the imaging apparatus of the present invention, the imaging apparatus includes a plurality of imaging elements, and the horizontal direction and the vertical direction with respect to an output signal of at least one imaging element among the plurality of imaging elements. A first pixel adding unit that adds m pixels (m is an integer of 2 or more) in both the horizontal direction and the vertical direction, and the output signal of the plurality of imaging elements is the first pixel A second pixel that adds n pixels (n is an integer of 1 or more, where n <m) in the horizontal direction, the vertical direction, or both the horizontal and vertical directions, with respect to the output signal of the image sensor not subjected to pixel addition by the adding means. Output signals of the second pixel addition means, a first output signal having the same signal band as the output signal of the first pixel addition means, and a higher frequency than the first output signal. Output separately to the second output signal having a band including The input / output characteristics are nonlinear with respect to the band dividing unit, the first and second output signals output from the band dividing unit, and the output signal output from the first pixel adding unit. A non-linear processing means for performing processing, a luminance signal generating means for generating a luminance signal from the output signal of the non-linear processing means, and a color difference signal generating means for generating a color difference signal from the output signal of the non-linear processing means, The chrominance signal generation means includes an output signal of the first pixel addition means subjected to nonlinear processing by the nonlinear processing means, and the first output of the band dividing means subjected to nonlinear processing by the nonlinear processing means. A color difference signal is generated from the signal.

  According to the imaging apparatus of the present invention, before the waveform is distorted by the non-linear processing means called gamma correction, the G signal not subjected to pixel addition is converted into the first G signal in the same band as the R and B signals subjected to pixel addition, The first G signal is divided into a second G signal including a signal having a higher band than the first G signal, and the color difference signal is obtained from the first G signal and the R and B signals whose distortion and band match the R and B signals. In addition to the signal, by generating a luminance signal from the second G signal including a signal having a higher frequency than the first G signal, it is possible to obtain a video signal with no false color and high resolution. it can.

  In addition, since the sampling frequency of the R and B signals is lowered by pixel addition, an effect of reducing power can be obtained.

  The image pickup apparatus of the present invention can take the following various modes based on the above configuration.

  That is, in the first configuration of the imaging apparatus according to the present invention, the luminance signal generation unit includes the output signal of the first imaging element subjected to nonlinear processing by the nonlinear processing unit and nonlinear processing by the nonlinear processing unit. A luminance signal can be generated from the first and second output signals output from the band dividing unit that has been subjected to. According to the above configuration, since the luminance signal is generated using the output signal of the first image sensor that has been subjected to nonlinear processing by the nonlinear processing means including high frequency components, the second output signal is a pixel. Even if they are added, resolution degradation can be minimized as in the conventional imaging apparatus.

  In addition, the first and second imaging elements each independently perform photoelectric conversion on imaging light that has been color-separated into three colors of red, green, and blue, and the first imaging element includes red The imaging light that has been color-separated into blue can be photoelectrically converted, and the second imaging element can be configured to photoelectrically convert the imaging light that has been color-separated into green. According to the present invention, red. Green. Even in an imaging apparatus that independently photoelectrically converts imaging light that has been color-separated into blue three colors, generation of false colors and resolution degradation can be suppressed.

  In the second configuration of the imaging apparatus according to the present invention, the luminance signal generation unit may be configured such that the output signal of the first pixel addition unit subjected to nonlinear processing by the nonlinear processing unit and nonlinear output by the nonlinear processing unit. A luminance signal can be generated from the first and second output signals of the processed band dividing means. According to the above configuration, since the luminance signal is generated using the output signal of the first image sensor that has been subjected to nonlinear processing by the nonlinear processing means including high frequency components, the second output signal is a pixel. Even if they are added, resolution degradation can be minimized as in the conventional imaging apparatus.

  Further, the plurality of image pickup devices independently photoelectrically convert image pickup light separated into three colors of red, green, and blue, and the first pixel adding means converts the light into red and blue Pixel addition is performed for each output signal of the image sensor that photoelectrically converts the color-separated imaging light, and the second pixel adding means outputs an output signal of the image sensor that photoelectrically converts the image light that has been color-separated into green Can be configured to perform pixel addition. According to the present invention, generation of false colors and resolution degradation can be suppressed even in an imaging device that independently photoelectrically converts imaging light that has been color-separated into three colors of red, green, and blue as described above. .

(Embodiment 1)
FIG. 1 is a block diagram illustrating the basic configuration of the main part of the imaging apparatus according to the first embodiment.

  The imaging elements 101a to 101c are elements that independently photoelectrically convert imaging light that is color-separated into G (green), R (red), and B (blue) by a spectroscopic means (not shown) such as a prism. For example, a CCD image sensor or a CMOS image sensor can be cited as an example. In this embodiment, a CCD image sensor is used. The G image pickup element 101a, the R image pickup element 101b, and the B image pickup element 101c have a general arrangement in which no horizontal spatial shift is performed.

  The image sensor driving circuits 102a to 102c are circuits that drive the image sensors 101a to 101c and perform pixel addition in the horizontal direction by driving the image sensors 101a to 101c. Specifically, drive pulses for driving the image sensors 101a to 101c are generated and output to the image sensors 101a to 101c.

  The pixel addition selection circuit 103 is a circuit that controls the number of pixel additions to the image sensor driving circuits 102a to 102c.

  The analog signal processing circuits 104a to 104c are, for example, a correlated double sampling circuit (CDS circuit: CDS: Correlated Double Sampling) that performs a sampling operation and a holding operation on each output signal of the image sensors 101a to 101c, and a constant signal level. An AGC circuit (AGC: Auto Gain Control) that performs gain control so as to maintain the same is included.

  The A / Ds 105a to 105c convert analog signals output from the analog signal processing circuits 104a to 104c into digital signals.

  The band dividing circuit 106 performs a filtering process on the digitized G signal output from the A / D 105a, and outputs the divided signal into two types of signals. The signal output from the band dividing circuit 106 is input to the gamma correction circuits 107a and 107b. The band dividing circuit 106 is an example of a band dividing unit.

  The gamma correction circuits 107 a to 107 d are circuits that perform gamma correction on the output signals of the A / Ds 105 a to 105 c and the band dividing circuit 106. The gamma correction circuits 107a to 107d are examples of nonlinear processing means.

  The luminance signal generation circuit 108 generates a luminance signal (Y signal) using the output signals of the gamma correction circuits 107a to 107d and outputs it to the subsequent stage. The Y signal is generated based on Equation 1 described above. The luminance signal generation circuit 108 is an example of a luminance signal generation unit.

  The color difference signal generation circuit 109 generates a color difference signal (C signal) using the output signals of the gamma correction circuits 107b to 107d, and outputs it to the subsequent stage. Note that the C signal is generated based on Equations 2 and 3 described above. The color difference signal generation circuit 109 is an example of a color difference signal generation unit.

  Hereinafter, the shooting operation of the imaging apparatus will be described.

  When imaging light that has been color-separated into G, R, and B enters the imaging apparatus, the imaging elements 101a to 101c perform photoelectric conversion operations based on the driving pulses from the imaging element driving circuits 102a to 102c, respectively. , R, and B are output as analog electric signals. The drive pulses output to the respective image sensors by the image sensor driving circuits 102a to 102c are the charge detection amplifier for detecting and amplifying the horizontally transferred signal charges in addition to the vertical and horizontal transfer pulses, and outputting them as electric signals. There are reset pulses.

  The pixel addition selection circuit 103 sets a horizontal pixel addition number for the image sensor driving circuits 102a to 102c and performs control for changing the cycle of the reset pulse. That is, the imaging element driving circuits 102b and 102c are controlled so that the pixel addition number is “2”, that is, the cycle of the reset pulse to be output is doubled. As a result, the R and B signals output from the image sensors 101b and 101c are signals obtained by adding pixel signals for two horizontal pixels. On the other hand, the pixel addition selection circuit 103 does not change the cycle of the reset pulse for the image sensor driving circuit 102a. As a result, the G signal that is not subjected to pixel addition as usual is output from the image sensor 101a.

  FIG. 2 is a diagram illustrating the spatial positions of pixel signals output from the image sensors 101a to 101c. FIG. 2A shows the spatial position of each pixel signal when none of the G, R, and B signals has undergone pixel addition processing. FIG. 2B shows the spatial position of each pixel signal when only the R and B signals are subjected to pixel addition processing. That is, at the time of pixel addition processing, the R image sensor 101b performs pixel addition processing as (R1 + R2), (R3 + R4),..., And outputs signals R12, R34,. Further, in the B image sensor 101c, pixel addition processing is performed as (B1 + B2), (B3 + B4),..., And signals B12, B34,.

  Since the R and B signals output from the image sensors 101b and 101c are added for two horizontal pixels, the mathematical formula is similar to the signals output from the R image sensor and the B image sensor in the conventional image pickup apparatus. 5 is the same as the characteristic subjected to the LPF processing of the transfer function shown in FIG. The sampling frequency is half that of the G signal.

  After the above control, analog electrical signals of R, G, and B output from the image sensors 101a to 101c are processed in the analog signal processing circuits 104a to 104c and A / Ds 105a to 105c in the same manner as the conventional imaging device. Output to the subsequent stage. The output signal of the A / D 105a is output to the band dividing circuit 106, divided into two types of signals and output.

  FIG. 3 is a block diagram illustrating an example of the band dividing circuit 106. In FIG. 3, an LPF 301 performs LPF processing on the output signal of the A / D 105a. The pixel thinning circuit 302 is a circuit that lowers the clock frequency by thinning pixels in the horizontal direction with respect to the output signal of the LPF 301.

  When the G signal that is the output of the A / D 105a is input, the band dividing circuit 106 first divides the signal into a signal that is subjected to LPF processing by the LPF 301 and a signal that is not subjected to LPF processing. By using the same transfer function as that of Expression 5 as the transfer function of the LPF 301, the band of the G signal processed by the LPF 301 is limited to the same band as the R and B signals output from the imaging elements 101b and 101c. The output signal of the LPF 301 is input to the pixel thinning circuit 302.

  Since the pixel thinning circuit 302 thins out one pixel with respect to two horizontal pixels, the clock frequency is halved. The G signal thinned out by the pixel thinning circuit 302 is output as a GL signal to the subsequent gamma correction circuit 107b. On the other hand, the G signal input to the band dividing circuit 106 and not subjected to the LPF processing is output as it is to the subsequent gamma correction circuit 107a as a GH signal. Therefore, the GH signal output to the gamma correction circuit 107a is a signal including signals in a higher frequency range than the output R and B signals of the image sensors 101b and 101c.

  FIG. 4 shows a GH signal (FIG. 4A) and a GL signal (FIG. 4B), an R signal (FIG. 4C), and a B signal (FIG. It is a figure showing a pixel space position with d)). The spatial position of the GL signal coincides with that of the R and B signals by the processing as described above.

  Returning to FIG. 1, the gamma correction circuits 107a to 107d perform gamma correction of the characteristics shown in FIG. All the output signals of the gamma correction circuits 107a to 107d are input to the luminance signal generation circuit 108, but only the output signals of the gamma correction circuits 107b to 107d are input to the color difference signal generation circuit 109.

  In the luminance signal generation circuit 108, the GL, R, and B signals limited to the same band among all the output signals of the gamma correction circuits 107a to 107d are first oversampled twice as in the conventional imaging device. Pixel interpolation is performed by processing to match the number of pixels with the GH signal. However, since the signals output from the gamma correction circuits 107a to 107d in the present embodiment are different from the conventional imaging device in the spatial position of the GH signal, that is, the GL, R, and B signals with respect to the G signal. Pixel interpolation is performed using an oversampling filter in which the spatial position after interpolation matches the G signal. An example of such an oversampling filter is a transfer function H (z) as shown in Equation 6 below.

H (z) = (1 + 3z ⁻¹ + 3z ⁻² + z ⁻³ ) / 8 (Formula 6)
Next, a Y signal is generated based on Expressions 7 to 9 from the GL, R, and B signals and the GH signal including higher frequency components.

YL = 0.2126R + 0.7152GL + 0.0722B (Formula 7)
YH = GH-GL (Formula 8)
Y = YL + αYH (Formula 9)
In Expression 9, α is normally “α = 1”. On the other hand, when it is desired to increase the resolution of the image, the high frequency component can be increased by setting “α> 1”. Further, in the case of shooting conditions in which noise is conspicuous such as low illuminance, by setting “α <1”, high frequency components can be reduced and noise can be suppressed.

  On the other hand, the color difference signal generation circuit 109 performs the same double oversampling processing as the luminance signal generation circuit 108 on the output signals of the gamma correction circuits 107b to 107d, that is, the GL, R, and B signals, and then performs conventional imaging. Similarly to the apparatus, the C signal is generated based on the following formulas 10 and 11.

Pr = −0.1146R−0.3854GL + 0.5000B (Equation 10)
Pb = 0.5000R-0.4542GL-0.0458B (Formula 11)
By generating the Y signal and the C signal as described above, the same frequency characteristics as those of the R and B signals including not only the signal component of the band of 4 / fs or less but also the folded component of the signal of the band of 4 / fs or more are obtained. Since the GL signal has a gamma correction together with the R and B signals and generates the C signal, even if the waveform is distorted by the gamma correction, the frequency components newly generated by the GL signal and the R and B signals are It becomes the same frequency component. Therefore, in the imaging apparatus according to the present embodiment, the false color that has been a problem in the conventional imaging apparatus does not occur. In addition, since the GH signal including a high frequency component is used for generating the Y signal, the resolution deterioration can be minimized as in the conventional imaging device even if the R and B signals are pixel-added. it can.

  In addition, since the drive frequency of the R image sensor 101b, the B image sensor 101c, and the circuit that processes the output signal is lowered by the pixel addition process, the effect of reducing the power consumption is the same as that of the conventional image pickup apparatus. Obtainable.

  In the first embodiment, CCD image sensors are used as the image pickup elements 101a to 101c. However, the present invention is not limited to this, and a CMOS image sensor may be used. By using a CMOS image sensor, it is possible to achieve an effect that power can be further reduced as compared with the case of using a CCD image sensor.

  In the first embodiment, the image sensor driving circuits 102a to 102c perform pixel addition by changing the cycle of the reset pulse, but the present invention is not limited to this. For example, a circuit for performing pixel addition may be provided on a signal transfer stage in a CCD image sensor or on a pixel readout signal line in a CMOS image sensor, and drive control of the circuit may be changed.

  In the first embodiment, the pixel addition selection circuit 103 controls the image sensor drive circuits 102b and 102c that drive the R image sensor 101b and the B image sensor 101c to add two pixels. The image pickup device driving circuit 102a that drives the image pickup device 101a is controlled not to perform pixel addition, but is not limited thereto. For example, m pixels (m = 2, 3,...) May be added to the R image sensor 101b and the B image sensor 101c. Accordingly, for the G signal, the transfer function of the LPF 301 of the band dividing circuit 106 is changed to match the frequency characteristic of the signal after addition, and the thinning number of the pixel thinning circuit 302 is set to m. The oversampling processing in the luminance signal generation circuit 108 and the color difference signal generation circuit 109 is set to m times. As a result, false colors can be prevented from being generated. In addition, the circuit that processes the R and B signals has an effect that the power can be further reduced.

If the Y signal is not required to have a very high resolution, n pixels (n = 2, 3,...) May be added to the G image sensor 101a. In this case, the number of added pixels of the R image sensor 101b and the B image sensor 101c is calculated as follows:
m = k × n (k = 2, 3, 4,...)
Then, since the sampling frequency ratio between the G signal and the R and B signals becomes an integer, the transfer function of the LPF 301 of the band dividing circuit 106 can be matched with the frequency characteristics of the R and B signals. Therefore, it is possible to achieve lower power without generating a false color and hardly reducing the resolution necessary for the Y signal.

  In the first embodiment, the configuration of the band dividing circuit 106 is as shown in FIG. 3, but the configuration is not limited to this. For example, the configuration of FIG. In FIG. 5, a high-pass filter (hereinafter referred to as HPF) 501 performs HPF processing on the output signal of the A / D 105a. The transfer function H (z) of the HPF 501 is set, for example, as in the following formula 12.

H (z) = (1-z ⁻¹ ) / 2 (Expression 12)
As a result, the same processing as that of the above-described Expression 8 is performed. Therefore, in the luminance signal generation circuit 108, “YH = GH” is set instead of Equation 8, and otherwise, the same effects as in the first embodiment can be obtained by performing the same processing as in the first embodiment. it can.

  In the first embodiment, the pixel addition processing in the imaging elements 101a to 101c is performed only in the horizontal direction. However, the present invention is not limited to this, and the same applies to the pixel addition processing in the vertical direction. In that case, filtering processing by the LPF 301 in the band dividing circuit 106 and oversampling processing in the luminance signal generation circuit 108 and the color difference signal generation circuit 109 may be performed in the vertical direction. Needless to say, the same applies to the case where pixels are added simultaneously in the horizontal and vertical directions.

(Embodiment 2)
FIG. 6 is a block diagram illustrating a basic configuration of a main part of the imaging device according to the second embodiment. In FIG. 6, the same reference numerals are given to blocks that perform the same functional operations as those in FIG. 1, and thus description thereof is omitted. Pixel addition circuits 601a to 601c perform pixel addition processing on the output signals of A / D 105a to A / D 105c. The pixel addition selection circuit 602 is a circuit that controls the number of pixel additions for the pixel addition circuits 601a to 601c. The pixel addition circuits 601a to 601c are examples of pixel addition means.

  The imaging apparatus according to the second embodiment is different from the imaging apparatus according to the first embodiment mainly in the operations of the imaging elements 101a to 101c, the pixel addition circuits 601a to 601c, and the pixel addition selection circuit 602. Hereinafter, the operation of these circuits will be mainly described.

  When imaging light that has been color-separated into G, R, and B is incident on the imaging apparatus, the imaging elements 101a to 101c perform photoelectric conversion operations in response to drive pulses from the imaging element drive circuits 102a to 102c, respectively, in the first embodiment. Analog electric signals corresponding to G, R, and B are output in the same manner as described above. However, unlike the first embodiment, the imaging device according to the second embodiment does not perform pixel addition control on the imaging element driving circuits 102a to 102c. Therefore, the G image sensor 101a as well as the R image sensor 101b and the B image sensor 101c output signals that are not subjected to pixel addition, like the G image sensor 101a in the first embodiment.

  Analog signal processing circuits 104a to 104c and A / Ds 105a to 105c perform the same processing as in the first embodiment. Digital signals output from the A / Ds 105a to 105c are output to the pixel addition circuits 601a to 601c, respectively.

  The pixel addition circuits 601a to 601c perform digital pixel addition processing according to the number of pixel additions set by the pixel addition selection circuit 602.

  FIG. 7 is a block diagram illustrating an example of the pixel addition circuit 601b. The same circuit is used for the pixel addition circuits 601a and 601c. In FIG. 7, a pixel addition circuit 701 is a circuit that digitally adds horizontal pixels to the output signal of the A / D 105b. The pixel thinning circuit 702 is a circuit that thins out pixels in the horizontal direction from the signal obtained by pixel addition in the pixel addition circuit 701. The number of pixels to be added by the pixel addition circuit 701 (the number of addition pixels) and the number of pixels to be thinned by the pixel thinning circuit 702 are determined by the addition pixel number signal from the pixel addition selection circuit 602. For example, when the number of added pixels is “2”, the pixel adding circuit 701 performs a pixel addition process for two horizontal pixels, and the pixel thinning circuit 702 reduces the number of pixels by 1 to 2 pixels. After performing pixel thinning processing at the rate of times, the clock frequency is reduced to ½.

  The pixel addition selection circuit 602 outputs a signal that adds two horizontal pixels to the pixel addition circuits 601b and 601c, and does not perform horizontal one-pixel addition, that is, pixel addition, to the pixel addition circuit 601a. Output a signal.

  As a result of the operations of the pixel addition circuits 601a to 601c and the pixel addition selection circuit 602, the R and B signals input to the gamma correction circuits 107c and 107d are the signals obtained by adding two horizontal pixels as in the first embodiment. Become. Also, the G signal input to the band dividing circuit 106 is a signal that is not subjected to pixel addition as in the first embodiment. Subsequent operations are the same as those in the first embodiment, and thus detailed description thereof is omitted.

  As described above, the imaging apparatus according to the second embodiment does not generate the false color, which is a problem in the conventional imaging apparatus, as in the first embodiment, and the resolution degradation is minimal as in the conventional imaging apparatus. Can be suppressed. In addition, since the image sensor does not perform pixel addition processing, the structures of the image sensors 101a to 101c and the image sensor drive circuits 102a to 102c can be made simpler than in the first embodiment. For the R image sensor 101b and the B image sensor 101c, the drive frequency is higher than that of the first embodiment because the pixel addition is not performed, but the R and B signal processing circuits after the pixel adder circuits 601b and 601c are driven. Since the frequency can be lowered, the effect is weaker than in the first embodiment, but the effect of lowering the power can also be obtained.

  In the second embodiment, a CCD image sensor is used as an image sensor. However, the present invention is not limited to this, and a CMOS image sensor may be used as in the first embodiment. By using a CMOS image sensor, it is possible to achieve an effect that power can be further reduced as compared with the case of using a CCD image sensor.

  In the second embodiment, the pixel addition selection circuit 602 controls the R and B signal pixel addition circuits 601b and 601c to add two pixels, and the G signal pixel addition circuit 601a. Is controlled so as not to perform pixel addition, but is not limited to this. Similarly to the first embodiment, the pixel addition circuits 601b and 601c for R and B signals have m pixels (m = 2). , 3, ...) may be added. Accordingly, the G signal is oversampled in the transfer function of the LPF 301 of the band dividing circuit 106, the thinning out number of the pixel thinning circuit 302, the luminance signal generation circuit 108, and the color difference signal generation circuit 109 as in the first embodiment. The false color can be prevented from being generated by changing. In addition, the circuit that processes the R and B signals has an effect that the power can be further reduced.

  If the Y signal is not required to have a very high resolution, as in the first embodiment, n pixels (n = 2, 3,...) Are added to the G signal pixel adding circuit 601a, and the signals for the R and B signals are added. The addition pixel number m of the pixel addition circuits 601b and 601c may be m = k × n (k = 2, 3, 4,...). Thereby, the same effect as Embodiment 1 is acquired.

  In the second embodiment, the pixel addition processing in the pixel addition circuits 601a to 601c is performed only on the pixels in the horizontal direction. However, the present invention is not limited to this, and the same applies to the processing for adding the pixels in the vertical direction. Applicable to. In that case, similarly to the first embodiment, the filter processing by the LPF 301 in the band dividing circuit 106 and the oversampling processing in the luminance signal generation circuit 108 and the color difference signal generation circuit 109 may be performed in the vertical direction. Needless to say, the same can be applied to the case where pixel addition processing is performed simultaneously in the horizontal and vertical directions.

  In the second embodiment, it goes without saying that the configuration of the band dividing circuit 106 is not limited to that shown in FIG.

  The image pickup apparatus according to the present invention can be applied to a purpose of obtaining a video signal with a high resolution without generating a false color and further obtaining an effect of reducing power consumption. For example, it is useful for a photographing apparatus including at least an image sensor such as a CCD image sensor, such as a digital video camera or a digital still camera.

1 is a block diagram showing a configuration of an imaging apparatus according to Embodiment 1 of the present invention. Schematic diagram showing the spatial positional relationship of each pixel of the signal output from the R, G, B image sensor used in the first embodiment FIG. 3 is a block diagram illustrating a configuration example of a band dividing circuit used in the first embodiment. Schematic diagram showing the spatial positional relationship between each pixel of the output signal of the band dividing circuit used in Embodiment 1 and the signal output from the R and B imaging elements FIG. 3 is a block diagram illustrating a configuration example of a band dividing circuit used in the first embodiment. FIG. 6 is a block diagram illustrating a configuration of an imaging apparatus according to a second embodiment. FIG. 7 is a block diagram illustrating a configuration example of a pixel addition circuit used in Embodiment 2. Block diagram showing the configuration of a conventional imaging device Schematic diagram showing the spatial positional relationship of each pixel output from an R, G, B image sensor used in a conventional imaging device Characteristic diagram showing input / output characteristics of gamma correction circuit A characteristic diagram showing the frequency characteristics of each signal output from an R, G, B imaging element used in a conventional imaging device Waveform diagram showing input and output waveforms when a sine wave is input to the gamma correction circuit

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Image sensor 102 Image sensor drive circuit 103 Pixel addition selection circuit 104 Analog signal processing circuit 105 A / D
106 Band Division Circuit 107 Gamma Correction Circuit 108 Luminance Signal Generation Circuit 109 Color Difference Signal Generation Circuit 301 LPF
302 Pixel thinning circuit 501 HPF
601 Pixel addition circuit 602 Pixel addition selection circuit 701 Pixel addition circuit 702 Pixel thinning circuit

Claims (8)

A first imaging device capable of performing m (m is an integer of 2 or more) pixel addition in the horizontal direction, vertical direction, or both horizontal and vertical directions;
A second imaging device capable of adding n pixels (n is an integer of 1 or more, where n <m) in the horizontal direction, vertical direction, or both horizontal and vertical directions;
The output signal of the second image sensor is a first output signal having the same signal band as that of the output signal of the first image sensor, and a second having a band including a higher frequency than the first output signal. A band dividing means for outputting the output signal separately;
Nonlinear processing means for performing nonlinear processing on input / output characteristics for the first and second output signals output from the band dividing means and the output signal output from the first image sensor; ,
A luminance signal generating means for generating a luminance signal from the output signal of the nonlinear processing means;
Color difference signal generation means for generating a color difference signal from the output signal of the nonlinear processing means,
The color difference signal generating means includes
A color difference signal is generated from the output signal of the first image sensor that has been subjected to nonlinear processing by the nonlinear processing means and the first output signal of the band dividing means that has been nonlinear processed by the nonlinear processing means. An imaging device. The luminance signal generating means includes
The output signal of the first image sensor that has been subjected to nonlinear processing by the nonlinear processing means, and the first and second output signals that are output from the band dividing means that has been subjected to nonlinear processing by the nonlinear processing means. The imaging device according to claim 1, wherein a luminance signal is generated from The first and second imaging elements are
Independently photoelectrically converts imaging light that has been separated into three colors of red, green, and blue,
The first imaging element photoelectrically converts imaging light that has been color-separated into red and blue,
The imaging apparatus according to claim 1, wherein the second imaging element photoelectrically converts imaging light that has been color-separated into green. An imaging apparatus including a plurality of imaging elements,
First, m pixels (m is an integer of 2 or more) are added to the output signal of at least one of the plurality of image sensors in the horizontal direction, the vertical direction, or both the horizontal direction and the vertical direction. 1 pixel adding means;
Among the plurality of image pickup devices, n output signals are output in the horizontal direction, the vertical direction, or both the horizontal and vertical directions with respect to the output signals of the image pickup devices whose output signals are not subjected to pixel addition by the first pixel addition unit (n is 1). A second pixel addition means for performing pixel addition of the above integers, where n <m,
The output signal of the second pixel adding means has a first output signal having the same signal band as the output signal of the first pixel adding means, and a first band having a higher band than the first output signal. Band dividing means for dividing the output signal into two output signals;
Non-linear processing means for performing non-linear processing of input / output characteristics on the first and second output signals output from the band dividing means and the output signal output from the first pixel adding means. When,
A luminance signal generating means for generating a luminance signal from the output signal of the nonlinear processing means;
Color difference signal generation means for generating a color difference signal from the output signal of the nonlinear processing means,
The color difference signal generating means includes
From the output signal of the first pixel adding means subjected to nonlinear processing by the nonlinear processing means and the first output signal of the band dividing means subjected to nonlinear processing by the nonlinear processing means, a color difference signal is obtained. An imaging device that generates The luminance signal generating means includes
From the output signal of the first pixel adding means subjected to nonlinear processing by the nonlinear processing means and the first and second output signals of the band dividing means subjected to nonlinear processing by the nonlinear processing means The imaging device according to claim 4, wherein a luminance signal is generated. The plurality of image sensors are
Independently photoelectrically converts imaging light that has been separated into three colors, red, green, and blue,
The first pixel addition means performs pixel addition for each output signal of an image sensor that photoelectrically converts image light that has been color-separated into red and blue,
The imaging apparatus according to claim 4 or 5, wherein the second pixel addition unit performs pixel addition on an output signal of an imaging element that photoelectrically converts imaging light that is color-separated into green.   The imaging apparatus according to claim 1, wherein the nonlinear processing unit is configured to perform gamma correction processing.   The imaging device according to claim 1, wherein m = 2 and n = 1.

Images